Evolutionary psychology: Conceptual foundations

Full text

3GC01 06/09/2015 12:40:42 Page 3
CHAPTER 1
The Theoretical Foundations
of Evolutionary Psychology
JOHN TOOBY and LEDA COSMIDES
THE EMERGENCE OF EVOLUTIONARY
PSYCHOLOGY: WHAT IS AT STAKE?
THE THEORY OF evolution by natural selection has revolutionary implications for
understanding the design of the human mind and brain, as Darwin himself was
the first to recognize (Darwin, 1859). Indeed, a principled understanding of the
network of causation that built the functional architecture of the human species offers
the possibility of transforming the study of humanity into a natural science capable of
precision and rapid progress. Yet, more than a century and a half after The Origin of
Species was published, many of the psychological, social, and behavioral sciences
continue to be grounded on assumptions that evolutionarily informed researchers
know to be false; the rest have only in the past few decades set to work on the radical
reformulations of their disciplines necessary to make them consistent with findings in
the evolutionary sciences, information theory, computer science, physics, the neuro-
sciences, molecular and cellular biology, genetics, behavioral ecology, hunter-gatherer
studies, biological anthropology, primatology, and so on (Pinker, 1997, 2002; Tooby &
Cosmides, 1992). Evolutionary psychology is the long-forestalled scientific attempt to
assemble out of the disjointed, fragmentary, and mutually contradictory human
disciplines a single, logically integrated research framework for the psychological,
social, and behavioral sciences—a framework that not only incorporates the evolu-
tionary sciences and information theory on a full and equal basis, but that systemati-
cally works out all the revisions in existing belief and research practice that such a
synthesis requires (Tooby & Cosmides, 1992).
The first long-term scientific goal toward which evolutionary psychologists are
working is the mapping of our universal human nature. By mapping human nature,
we mean the progressive construction and refinement of a set of empirically validated,
high-resolution models of the evolved adaptations (genetic, developmental, anatomi-
cal, neural, information processing, etc.) that collectively constitute universal human
We dedicate this chapter to the late Irven DeVore, professor emeritus, Department of Anthropology,
Harvard University, dear mentor and friend.
3
Tooby, J. & Cosmides, L. (2015). The theoretical foundations
of evolutionary psychology. In Buss, D. M. (Ed.), The
Handbook of Evolutionary Psychology, Second edition.
Volume 1: Foundations. (pp. 3-87). Hoboken, NJ: John Wiley
& Sons.

3GC01 06/09/2015 12:40:42 Page 4
nature. Because the focus in the behavioral and social sciences is on explaining mind,
behavior, and social interactions, initially the emphasis has been placed on adapta-
tions that are behavior-regulating, and which researchers may call a variety of names,
such as evolved psychological (mental, cognitive) programs, neurocomputational
programs, behavior-regulatory programs, adaptive specializations, “modules,”infor-
mation-processing mechanisms, and so on. However, because the architecture of the
human species evolved as a set of functional interactions at all physical and temporal
scales, it follows that genetic, cellular, developmental, anatomical, physiological,
endocrinological, and life-historical processes are also considered as fully part of
human nature, and, therefore, part of the systems of evolved interrelationships that
evolutionary psychology needs to deal with. Because the evolved function of a
regulatory mechanism is computational—to regulate behavior, development, and
the body adaptively (over the short term and the long term) in response to informa-
tional inputs—such a model consists of a description of the functional circuit logic or
information-processing architecture of the mechanism, in a way that eventually
should incorporate its physical implementation (Cosmides & Tooby, 1987; Tooby &
Cosmides, 1992). More completely, these models must sooner or later include
descriptions of the regulatory logic of the developmental programs that, in interaction
with environments, lead to the unfolding succession of designs that constitute the
organism’s changing phenotype across its life history (Tooby & Cosmides, 1992;
Tooby, Cosmides, & Barrett, 2003—see review in Del Giudice, Gangestad, & Kaplan,
Chapter 2, this volume). As scientific knowledge grows in the longer term, these
models will eventually come to incorporate descriptions of the neural and genetic
implementations of these mechanisms.
The second long-term scientific goal toward which evolutionary psychologists and
their allies are working is a comprehensive reconstruction of the social sciences (and
many of the humanities) that an accurate, natural science-based model of human
nature will both make possible and require. At present, the social sciences are a stew of
mutually contradictory claims, with no theoretical unity or clear progressive direction.
Major components of the social sciences are sufficiently incoherent to qualify—in Paul
Dirac’s phrase—as not even wrong. Genuine, detailed specifications of the circuit
logics of the neuroregulatory programs that compose human nature are expected to
become the theoretical centerpieces of a newly reconstituted set of social sciences. This
is because each model of an evolved component of human nature (e.g., the human-
language competence) makes predictions about (and explains) those sets of develop-
mental, psychological, behavioral, and social phenomena that its circuits generate or
regulate (e.g., the existence of and the patterns found in human language; Pinker, 1994;
the existence of and patterns found in incest aversion and kin-directed altruism;
Lieberman, Tooby, & Cosmides, 2007). The resulting changes to the social sciences are
expected to be dramatic and far-reaching because the traditional conceptual frame-
work for the social and behavioral sciences—what we have called the Standard Social
Science Model (SSSM)—was built from defective assumptions about the nature of the
human psychological and developmental architecture (for an analysis of the SSSM, see
Pinker, 2002; Tooby & Cosmides, 1992). The most consequential assumption is that the
human psychological architecture consists predominantly of learning and reasoning
mechanisms that are general purpose, content independent, and equipotential (Pinker,
2002; Tooby & Cosmides, 1992). That is, the mind is blank-slate-like, and lacks
specialized circuits that were designed by natural selection to respond differentially
to inputs by virtue of their evolved significance. This presumed psychology justifies a
4F
OUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 5
crucial foundational claim: Just as a blank piece of paper plays no causal role in
determining the content that is inscribed on it, the blank-slate view of the mind
rationalizes the belief that the evolved organization of the mind plays little causal role
in generating the content of human social and mental life. The mind with its learning
capacity absorbs its content and organization almost entirely from external sources.
These processes are thought to be analogous to the operation of a video camera—the
content of the recording originates in the world, whereas the mechanism of recording
adds no content of its own to the mix. As Thomas Aquinas put this seemingly self-
evident view, “There is nothing in the mind that was not first in the senses.”Hence,
according to the standard model, the social and cultural phenomena studied by the
social sciences are autonomous and disconnected from any nontrivial causal pattern-
ing originating in our evolved psychological mechanisms. Organization flows inward
to the mind from the processes in the social world (what we call the Durkheimian
causal arrow). More importantly, social scientists have considered it to be unshakably
well-established that content does not flow outward from evolved organization in
individual minds to organize culture or the social world (Geertz, 1973; Sahlins, 1976).
Now that this hypothesis is being empirically tested, however, it is regularly falsified
(e.g., Buss, 1989; Lieberman, Tooby, & Cosmides, 2003, 2007; Peterson, Sznycer, Sell,
Cosmides, & Tooby, 2013; Sell, Tooby, & Cosmides, 2009).
Yet if—as evolutionary psychologists have been demonstrating—the blank slate
view of the mind is wrong, then the social-science project of the past century is not only
wrong but radically misconceived. The blank slate assumption removes the central
causal organizers of social phenomena—evolved psychological programs—from the
analysis of social events, rendering the social sciences powerless to understand the
animating logic of the social world. Evolutionary psychology provokes so much
reflexive opposition because the stakes for many social scientists, behavioral scientists,
and humanists are so high: If evolutionary psychology turns out to be well founded,
then the existing superstructure of the social and behavioral sciences—the Standard
Social Science Model—will have to be dismantled. Instead, a new social science
framework will need to be assembled in its place that recognizes that models of
psychological mechanisms are essential constituents of social theories (Boyer, 2001;
Sperber, 1994, 1996; Tooby & Cosmides, 1992).
Within such a framework, the circuit logic of each evolved mechanism contributes
to the explanation of every social or cultural phenomenon it influences or helps to
generate. For example, the nature of the social interactions between the sexes are partly
rooted in the design features of evolved programs that underlie sexual behavior, mate
choice, attractiveness, intrasexual competition, intersexual conflict, and mateship
maintenance, reviewed in many chapters in this volume (for notable earlier work,
see Buss, 1994, 2000; Daly & Wilson, 1988; Symons, 1979). The patterned incidence of
violence is partly explained by the evolved programs governing our species’psychol-
ogy of aggression, parenting, and sexuality (Campbell, Daly & Wilson, 1988; Simp-
son & Campbell, Chapter 3, this volume); the foundations of trade can be located in
evolved cognitive specializations for social exchange (Cosmides & Tooby, 1992;
Tooby & Cosmides, Chapter 25, this Handbook, Volume 2); both incest avoidance
and love for family members are rooted in evolved mechanisms for kin recognition
(Lieberman, Tooby, & Cosmides, 2003, 2007). Similarly, the evolutionarily specialized
mechanisms underlying human alliance psychology help to explain phenomena such
as racism, coalitions, morality, social sanctions, and group dynamics (e.g., Delton,
Cosmides, Guemo, Robertson, & Tooby, 2012; Kurzban, Tooby, & Cosmides, 2001;
The Theoretical Foundations of Evolutionary Psychology 5

3GC01 06/09/2015 12:40:42 Page 6
Pietraszewski, Cosmides, & Tooby, 2014; Tooby & Cosmides, 2010; Tooby, Cos-
mides & Price, 2006).
A growing inventory of such models will catalyze the transformation of the social
sciences from fields that are predominantly descriptive, soft, and particularistic into
theoretically principled scientific disciplines with genuine predictive and explanatory
power. Evolutionary psychology in the narrow sense is the scientific project of
mapping our evolved psychological and developmental mechanisms; in the broad
sense, it includes the project of reformulating and expanding the social sciences (and
medical sciences, as somatic adaptations become incorporated into the synthesis)
in the light of the progressive mapping of our species’evolved architecture. This
Handbook contains reviews of the rich harvest of projects and discoveries that have
already emerged out of this young paradigm. Even though the field is in its infancy,
evolutionary psychologists have already identified a very large set of examples that
touch almost every aspect of human life. In the light of such rapidly accumulating
findings, many hallowed beliefs in anthropology, sociology, political science, social
psychology, cognitive psychology, and (to a lesser extent) economics will have to be
completely revised. However, we are only in the earliest phases of what is expected to
be an ever-widening transformation of the human and nonhuman behavioral sciences,
an enterprise so large that may take remainder of this century, and which is sure to
include surprises as more and more strands of conceptual unification proceed.
It is important to emphasize that evolutionary psychology in the broad sense is not
just about the design of the individual, nor is it just a revision of the present academic
field of psychology. Instead, this reformulation encompasses and integrates the entire
sweep of the human sciences. This is because our minds’programs evolved in ancestral
social, demographic, and informational environments that gradually produced and
refashioned various epidemeological and population- and group-level phenomena such
as cultural traditions, languages, social groups, and demographic structures. These,
in turn, acted as selection pressures that collectively engineered our constellation of
evolved programs to operate functionally with respect to these supra-individual
phenomena. That is, these programs evolved to functionally produce some of these
phenomena (e.g., alliances; language); they also evolved to act functionally within
environments that included these phenomena (e.g., fitness-promoting behavior guided
by an alliance detector;communication made possible by language competence). Hence,
the extended phenotypes (in Dawkins’1982 sense) that these programs produce are not
only individual traits (in the folk sense), but are designed to interact with each other to
produce or exploit complex collective phenotypes (e.g., languages, cultural elements,
traditions, exchange networks, social groups, agent-like coalitions, mobs, wars, small-
scale hierarchies, some small-scale institutions). Moreover, on their way to producing
the functional socially extended phenotypes they were designed ancestrally to produce,
our evolved programs andtheir outputs also produce many modern and complex group
and population-level phenomena as byproducts (e.g., global networks of exchange,
fashions, supply and demand curves, aristocracies, social classes, complex hierarchies,
complex institutions, religions, different languages, etc.).
All these are objects of study for social scientists, and because these are patterned by
our evolved programs, evolutionary psychology provides the integrating framework
for the social sciences. It is the specifics of our adaptations’decision-making architec-
tures that strongly shape how individuals assemble themselves into larger social
structures in the modern world, and that generate the cultural outputs that our minds
dynamically build and reshape over time.
AU: Please confirm
Campbell chapter
reference (3 okay?
Or 18?)
6FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 7
Because the Standard Social Science Model’s claim for the source of essentially all
human mental content is free-form culture downloaded into individual minds, it is
vital to realize how different the evolutionary psychological explanation is of the
origin of mental content, and the nature of culture. The evolutionary psychological
claim is that—for our evolved computational problem-solvers to actually solve the
adaptive problems faced by our ancestors (food acquisition, parenting, mate acquisi-
tion)—they had to be richly structured by selection in a content-specific way. That is,
they are endowed by what philosophers would once have called innate ideas or a priori
concepts (e.g., food, child, my child, male-female, ingroup-outgroup, mother, kin,
cheater, free-rider, snake, spider, animacy, number, noun, object, aggressive formi-
dability, friend, enemy, predator, leader, and perhaps thousands of others). These may
be built in to evolved modes of interpretation, conceptual-motivational systems, or
evolved intuitive ontologies, in what might be thought of as a Darwinian-Kantian-
computational synthesis of how our evolved programs organize experience (Boyer
& Barrett, Chapter 5, this volume; Cosmides & Tooby, 1994b; Tooby, Cosmides,
& Barrett, 2003). This different approach explains and often predicts the (previously
unappreciated) set of human universals (see, e.g., Brown, 1991) as reliably developing
adaptations, their byproducts, and their interactive products. It predicts and explains
principled cross-cultural variation; for example, adaptations have been designed by
selection to take relevant local conditions as input to produce output that is calibrated
to local circumstances (e.g., Gaulin & Schlegel, 1980; Schmitt, 2005; Sznycer et al.,
2012). This approach can even explain highly particularistic cultural phenomena as
unique patterns of activation of species-typical evolved mechanisms (e.g., Boyer, 2001).
Hence, our content-inflected mental adaptations reliably develop, as well as
generate, some of the particular content of human culture, and form the raw materials
out of which the rest is developmentally and socially elaborated in an immense and
endlessly shifting play of combinatorics. This content is then present to be adopted
or modified by evolved programs situated in other members of the population, or
shaped by social interactions. This gives rise to epidemiological and historical
population-level processes, located in particular ecological, economic, demographic,
and intergroup social contexts or environments, which themselves impact their cross-
individual and cross-generational dynamics. From this perspective, culture is the
manufactured product of our evolved neurocomputational programs situated in
individuals living in groups. To flag how different this theory of culture is from
classical general learning transmission approaches, we gave our book The Adapted
Mind (Barkow, Cosmides, & Tooby, 1992) the subtitle Evolutionary Psychology and the
Generation of Culture. The recognition that the mind contains a large array of evolved
programs leads to another departure from standard thought: Culture is not a unitary
stuff, nor is culture in any way independent of evolved psychological processes.
Instead, “culture”is located inside our evolved programs, and different kinds of
culture are located inside different programs (and their combinations). Different types
of information live inside distinct computational habitats as their native settings—that
is, habitats built out of different evolved mental programs. The computational
specifics of these different habitats give meaning to these data structures; they impose
meaningful structure on content; they determine the rules by which potential changes
to content can happen; they determine what inputs from which other programs
provide the raw material that a given program operates on; they determine which
contents in which internal habitats can become outputs. “Culture”and “learning”are
not theoretical rivals to evolutionary psychology; they are instead phenomena to be
The Theoretical Foundations of Evolutionary Psychology 7

3GC01 06/09/2015 12:40:42 Page 8
explained by reference to and in terms of the design of the evolved neural programs
that produce them (plus a description of the local inputs provided to these
mechanisms).
So there is fear-of-snakes culture (living “inside”the snake-phobia system, that can
be transmitted as intensities of fear-response passed to others), grammar culture
(living “inside”the language competence), food-preference culture, group- identity
culture, disgust culture, contempt culture, sharing culture, aggression culture, and so
on. The set of cultural competences arose as a response to the opportunity afforded by
the fact that other humans with their own calibrated programs are rich potential
sources of information. Any time a program can cost effectively improve its perform-
ance by censusing programs situated in other brains, then selection will favor the
evolution of inference systems to do so. All these distinct effects have been confusingly
aggregated under the single name “culture,”misleading people into thinking “cul-
ture”is a homogeneous stuff moving according to unitary principles free of the
influence of our evolved psychology. Instead, brains are linked by many causally
distinct pathways, built to perform distinct functions. Each brain is bristling with
many independent “tubes”that propagate many distinct kinds of stuff to and from a
diversity of brain mechanisms in others. This is why evolutionary psychology is not
restricted to studying the static determinants of individual behavior taken in isolation
from culture or social and historical setting. Instead, this is why evolutionary
psychology in the broad sense integrates with and provides a nonoptional founda-
tional framework for the social sciences (e.g., Boyer, 2001; Pinker, 2002; Sperber, 1996;
Tooby, 2014; Tooby & Cosmides, 1992).
For almost a century, adherence to the Standard Social Science Model has been
strongly moralized within the scholarly world, immunizing key aspects from criticism
and reform (Pinker, 2002; Tooby & Cosmides, 1992). As a result, in the international
scholarly community, criteria for belief fixation have often strayed disturbingly far
from the scientific merits of the issues involved, whenever research trajectories
produce results that threaten to undermine the credibility of the SSSM. Nevertheless,
in recent decades, the strain of ignoring, exceptionalizing, or explaining away the
growing weight of evidence contradicting traditional theories has become severe.
Equally, reexaminations of the arguments advanced in favor of the moral necessity of
the SSSM suggest that they—at best—result from misplaced fears (Pinker, 2002;
Tooby & Cosmides, 1992). Indeed, we may all have been complicit in the perpetuation
of vast tides of human suffering—suffering that might have been prevented or
alleviated if the scientific community had not chosen to postpone or forgo a more
accurate social and behavioral science.
THE INTELLECTUAL ORIGINS OF EVOLUTIONARY PSYCHOLOGY
Despite the marginalization of Darwinism within the behavioral and social sciences
during the 20th century, a diverse minority of thinkers tried to think through how
Darwinian insights could be applied to behavior. These efforts led to many valuable
approaches, including: the instinct psychology of William James and William
McDougall; the ethological approach of Tinbergen, Lorenz, and von Frisch, which
integrated the careful observation of animal behavior in natural contexts with
investigations of its adaptive significance and physiological basis; the sociobiological
approach of Richard Alexander, William Hamilton, Robert Trivers, Edward O. Wilson
8F
OUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 9
and many others, which often tried to explain patterns of social behavior—differences
as well as universals—in humans and other species in terms of their fitness conse-
quences; nativist approaches to language pioneered by Chomsky (1959, 1966), Lenne-
berg (1967) and others, which brought to wider attention the question of whether one
general-purpose learning system could account for all learning; and even behaviorist
psychology—quite orthodox with respect to the Standard Social Science Model—
looked for phylogenetic continuities in the laws of learning that would apply across
species. As valuable as each of these approaches turned out to be, conceptual
handicaps internal to each program limited their scope of application and their
capacity to usefully reorganize the human psychological, behavioral, and social
sciences.
The way past these limitations involved isolating or deriving a core set of
foundational concepts from the intersection of physics, biology, and information
theory, elucidating their logical and causal interrelationships, and then building back
upward from this groundwork. (A few representative concepts are function, regulation,
information, computational architecture, adaptation, organization, design, entropy, selection,
replication, selection pressure, byproduct, environment of evolutionary adaptedness, and
task environment.) These concepts could then be used to trace out the necessary
interconnections among several previously distinct scientific programs, so that the
previously independent (and often inconsistent) disciplinary building blocks could be
integrated into a single unified framework (for discussion, see Tooby & Cosmides,
1992). The building blocks from which evolutionary psychology was assembled
include (a) the modern adaptationist revolution in theoretical evolutionary biology
(Williams, 1966); (b) the rise of information theory and the computational sciences
(Shannon, 1948; Weiner, 1948); (c) the emergence of serious attempts to reconstruct the
ancestral conditions and ways of life of humans and prehumans and the selection
pressures they imposed on our lineage (e.g., Cheney, Seyfarth, Smuts, & Wrangham,
1987; Isaac, 1989; Kaplan & Hill, 1985; Lee & DeVore, 1968, 1976); and (d) an
adaptationist/computationalist resolution of the debate between environmentalists
and nativists (e.g., Cosmides & Tooby, 1987; Pinker, 1997; Tooby & Cosmides, 1990a,
1990b, 1992; Tooby, Cosmides, & Barrett, 2003).
The first building block of evolutionary psychology was the strain of theoretical
evolutionary biology that started in the late 1950s and early 1960s, especially with the
work of George Williams (Williams & Williams, 1957; Williams, 1966); William D.
Hamilton (1964); and John Maynard Smith (1982). By being placed on a more rigorous,
formal foundation of replicator dynamics, evolutionary biology was transformed over
the ensuing decades from a vaguely conceptualized and sometimes implicitly teleo-
logical field into a principled discipline that rivals physics in its theoretical beauty and
explanatory power. Oneface of this transformation has been the derivation of a series
of elegant selectionist theories—theories of how natural selection acts on altruism,
kinship, cooperation, mating, foraging, reproduction, parenting, risk-taking, aggres-
sion, senescence, host-parasite interactions, intragenomic conflict, life-history, com-
munication, and many other dimensions of life. Research in biology (and the human
sciences informed by these theories) has been called sociobiology, behavioral ecology,
evolutionary ecology, or simply evolutionary biology. In addition to evolutionary
genetics, a key foundation of the improvements in our understanding of replicator
dynamics was the application of game theory (von Neumann & Morgenstern, 1944) to
genetic and organismal interactions—a program that rapidly developed into evolu-
tionary game theory (Maynard Smith, 1982). (We think this process will continue as
The Theoretical Foundations of Evolutionary Psychology 9

3GC01 06/09/2015 12:40:42 Page 10
evolutionary game theory morphs into what might be called adaptationist game theory
(e.g., Delton, Krasnow, Cosmides, & Tooby, 2011; Krasnow, Delton, Cosmides &
Tooby, 2015).
1
The other face of this revolution in biology is modern adaptationism
(Williams, 1966)—a set of deductions that are still often misunderstood, even in
biology (Alcock, 2001; Thornhill, 1997; Tooby & Cosmides, 1992; Tooby, Cosmides, &
Barrett, 2003). Adaptationism is based on the recognition that selection is the only
known natural physical process that builds highly ordered functional organization
(adaptations) into the designs of species, in a world otherwise continuously assaulted
by the ubiquitous entropic tendency of physical systems to become increasingly
disordered with time. Thus, although not everything is functional, whenever complex
functional organization is found in the architectures of species, its existence and form
can be traced back to a previous history of selection. Moreover, for a given selection
pressure to drive an allele systematically upward until it is incorporated into the
species-typical design, the same selective cause-and-effect relationship must recur
across large areas and for many generations. Complex adaptations necessarily reflect
the functional demands of the cross-generationally long-enduring structure of the
organism’s ancestral world, rather than modern, local, transient, or individual condi-
tions. This is why evolutionary psychology as an adaptationist field concerns the
functional design of mechanisms given a recurrently structured ancestral world,
rather than the idea that behavior is the fitness striving of individuals tailored to
unique circumstances (Symons, 1992; Tooby & Cosmides, 1990a).
Consequently, systems of complex, antientropic functional organization (adapta-
tions) in organisms require explanation wherever they are found; their correct
explanation (barring supernatural events or artificial intervention) always involves
a specific history of selection in ancestral environments; and so the prediction,
discovery, mapping, and understanding of the functional architecture of organisms
can be greatly facilitated by analyzing the recurrent structure of a species’ancestral
world, in conjunction with the selection pressures that operated ancestrally. The
foundational recognition that psychological (neurocomputational) mechanisms are
evolved adaptations connects evolutionary biology to psychology in the strongest
possible fashion, allowing everything we know about the study of adaptations to be
applied to the study of psychological mechanisms. Whatever the sociology of
1
We think that, although there are many valuable results that have emerged from evolutionary game theory,
other widely cited and influential results are not truly applicable to real species such as humans. The goal of
adaptationist game theory is to replace a series of limitations in standard evolutionary game theory (such as
highly biologically implausible conditions, radically impoverished strategy-types, etc., that were adopted to
make the mathematics tractable or other reasons of preference or convenience) with modeling decisions
chosen to make the results more biologically realistic. This is made possible by moving from primarily
analytic approaches to agent-based population simulations; by endowing the simulated world with, for
example, more plausible information ecologies; giving agents locations; by endowing agents with richer and
more realistic strategies—specified psychologies—that include background capacities humans actually
have, such as individual recognition; by allowing relevant decision-making variables to evolve—that is, not
restricting strategy sets to a small number of discontinuous types such as defector and cooperator but,
instead, for example, allowing the probability of cooperation allowed to evolve from 0 to 1. For example, it
was widely thought that humans were irrationally generous, by cooperating in one-shot games, purportedly
showing individual selection could not explain human game performance. By simply recognizing that
interactions don’t come pre-typologized for the agent as either one-shot or repeated, and that the organism
must make this discrimination under uncertainty, simulations demonstrate that reciprocity under biologi-
cally plausible conditions spontaneously evolves to manifest the observed (and no-longer mysterious)
generosity-bias (Delton, Krasnow, Cosmides, & Tooby, 2011; see also Krasnow, Delton, Cosmides, & Tooby,
2013).
10 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 11
academic tribes, scientifically psychology (along with the other social and behavioral
sciences) is a subbranch of evolutionary biology and can no longer be defensibly
divorced from it.
George Williams’s 1966 volume, Adaptation and Natural Selection: A critique of Some
Current Evolutionary Thought, was central to both the adaptationist and selectionist
revolutions. In it, Williams provided the first fully modern statement of the relation-
ship between selection and adaptive design; clarified that selection operates at the
genic level; developed strict evidentiary standards for deciding what aspects of a
species’phenotype were adaptations, byproducts of adaptations, or noise; and
usefully distinguished the present usefulness of traits (if any) from their evolved
functions (if any).
2
The second building block of evolutionary psychology was the rise
of the computational sciences and the recognition of the true character of mental
phenomena. Boole (1848) and Frege (1879) formalized logic in such a way that it
became possible to see how logical operations could be carried out mechanically,
automatically, and hence through purely physical causation, without the need for an
animate interpretive intelligence to carry out the steps. This raised the irresistible
theoretical possibility that not only logic but other mental phenomena such as goals
and learning also consisted of formal relationships embodied nonvitalistically in
physical processes (Weiner, 1948). With the rise of information theory (Shannon,
1948), the development of the first computers (von Neumann, 1945), and advances in
cybernetics and neuroscience (Weiner, 1948), it became widely understood that mental
events consisted of transformations of structured informational relationships embod-
ied as aspects of organized physical systems in the brain. This spreading appreciation
constituted the cognitive revolution. The world of the mental was no longer a
mysterious, indefinable realm, but locatable in the physical world in terms of precisely
describable, highly organized causal relations. Why do these informational relation-
ships emerge in physical systems in organisms? The adaptive problem of regulating
behavior in a fitness-promoting way could be seen as the selection pressure that led to
the emergence of systems for natural computation—that is, as naturally engineered
behavior control systems for organisms—adaptationist cybernetics.
Evolutionary psychology can, therefore, be seen as the inevitable intersection of the
computationalism of the cognitive revolution with the adaptationism of Williams’
evolutionary biology: Because mental phenomena are the expression of complex
functional organization in biological systems, and complex organic functionality is
the downstream consequence of natural selection, then it must be the case that the
sciences of the mind and brain are adaptationist sciences, and psychological mecha-
nisms are computational adaptations. In this way, the marriage of computationalism
with adaptationism marks a major turning point in the history of ideas, dissolving the
intellectual tethers that had limited fundamental progress, and opening the way
forward. Like Dalton’s wedding of atomic theory to chemistry, computationalism and
adaptationism solve each other’s deepest problems, and open up new continents of
2
The arguments that not every trait is an adaptation, not all beneficial effects of a trait are its functions, that
phenotypes are full of byproducts, and that there are constraints on developing systems were all central to
Williams’s 1966 critique of evolutionary biology. Thus, many of us were surprised when, 13 years later,
Stephen Jay Gould and Richard Lewontin (1979) began to repeat the same critique without attribution,
writing as if it were unknown to the evolutionary community they were criticizing. One striking difference
between the two critiques was Williams’development of strict standards of evidence can be used to
distinguish adaptations from nonadaptations, rendering the issue a matter of empirical research rather than
post hoc rhetoric.
The Theoretical Foundations of Evolutionary Psychology 11

3GC01 06/09/2015 12:40:42 Page 12
scientific possibility (Cosmides & Tooby, 1987; Tooby & Cosmides, 1992; Tooby,
Cosmides, & Barrett, 2003, 2005).
Sociologically speaking, the single most significant factor in triggering the renewed
efforts to apply evolution to behavior was the selectionist revolution in evolutionary
biology, which subsequently (if temporarily) became known as sociobiology (Wilson,
1975). Across the world, biologists and allied researchers were electrified by the
potential predictive and explanatory power of the new selectionist theories that were
emerging, together with how elegantly and systematically they could be derived.
Dynamic research communities formed at Oxford, Cambridge, Sussex, Michigan,
Harvard, the University of California, and elsewhere. As a result of the flood of
empirical and theoretical work coming out of these communities, the adaptationists/
selectionist revolution rapidly established itself in the biological journals as the
dominant theoretical approach biologists apply to understanding the behavior of
nonhumans—a position behavioral and social scientists are surprised to find that it
occupies today (often under other names such as behavioral or evolutionary ecology.
3
At Harvard, for example, under the sponsorship of Irven DeVore and E.O. Wilson,
one of the most influential and dynamic of these communities gathered and matured.
This research community fluoresced in Irven DeVore’s living room, where Harvard’s
Simian Seminar was held from 1971 through the mid-1980s. In this atmosphere of
ongoing discovery, ideas and findings sparked each other in an endless chain reaction.
A remarkable procession of figures in evolutionary biology, behavioral ecology,
primatology, and ethology spoke at DeVore’s Simian Seminar, participating in this
chain reaction, and sometimes staying for extended periods. Among many others,
these included George Williams, Bill Hamilton, John Maynard Smith, Ernst Mayr,
Edward O. Wilson, Richard Alexander, Richard Dawkins, Tim Clutton-Brock, Paul
Harvey, Lionel Tiger, Robin Fox, Diane Fosse, Jane Goodall, Robert Hinde, Richard
Leakey, Joseph Shepher, Richard Lee, Stephen Jay Gould, Martin Daly, and Margo
Wilson, and the editor of this Handbook, David Buss. Among the students or proteges
DeVore mentored in this environment were Bob Bailey, Peter Ellison, John Fleagle,
Steve Gaulin, Henry Harpending, Paul Harvey, Sarah Blaffer Hrdy, Melvin Konner,
Jeff Kurland, Jim Moore, Nadine Peacock, Peter Rodman, Robert Sapolsky, John Seger,
Marjorie Shostak, Barbara Smuts, Karen Strier, Bob Trivers, Carol Worthman, Richard
Wrangham, John Yellen, and ourselves (John Tooby and Leda Cosmides). Although
Wilson’s contributions are deservedly famous through his books and publications,
DeVore’s intellectual impact is less well known because his ideas were realized
through his students, proteges, and colleagues. Deeply interested in human origins,
DeVore pioneered three major research movements. He initiated and then champ-
ioned the systematic study of primate social behavior under natural conditions
(DeVore, 1962, 1965). Thisled him to want to incorporate human hunter-gatherers
into the same careful scientific framework. With Lee and many other colleagues, in
1963 he inaugurated the systematic, empirical, quantitative investigation of living
3
Intellectuals wedded to the blank slate generated an unslakable demand for seemingly authoritative
dismissals of the new biology. As a result, the handful of biologists who were willing to ignore the data and
supply these dismissals came to be seen as the authentic voices of scientific biology to the intellectual world
at large (e.g., Gould & Lewontin, 1979). The decisive empirical success of the paradigm within biology
itself—what Alcock (2001) calls “the triumph of sociobiology”—is largely unknown outside of the field, and
the majority of nonbiologists labor under the misimpression that sociobiology was substantively discredited
by “real”biologists.
12 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY
Thisled

3GC01 06/09/2015 12:40:42 Page 13
hunter-gatherers with the Kalahari Research Project and the famous Man the Hunter
meetings (Lee & DeVore, 1968, 1976). Then, together with Chagnon, Irons, and many
anthropologists, he worked on applying the new selectionist biology to anthropologi-
cal questions.
DeVore and his colleague, Richard Lee, eschewed the “lone anthropologist”model
(with its typological baggage), in which a single individual spends time documenting
“the”culture of a people. In its place, they innovated a team-based approach like that
found in other sciences. (Imagine the state of physics if one physicist studied the
electron, another the mu meson, etc.) Their Kalahari San project brought scientists and
scholars from a broad array of disciplines—anthropologists, demographers, physi-
cians, linguists, folklorists, psychologists, ethologists, archeologists—in an attempt to
document as completely as possible the behavior, health, and lives of the !Kung San
people in Botswana’s Kalahari desert, before hunting and gathering as a way of life
disappeared from the planet. His goal in studying the San was to provide a detailed
database that, when triangulated with other similarly detailed databases drawn from
other hunter-gatherer groups, would allow new and powerful inferences to be made
about the selection pressures that operated on hunter-gatherers to shape human
design. Behavioral ecologists would be able to test optimal foraging models by
matching foraging patterns to ecological conditions. Archaeologists could better
interpret patterns found at ancestral sites by seeing patterns of campfires, animal
remains, tool-making debris, and midden heaps produced by the social life of living
hunter-gatherers. Medical researchers could gain insight into diseases of civilization
by comparing diets and conditions in industrialized countries to the diets and stressors
produced by a way of life that more closely resembles the conditions in which our
species evolved. Developmental psychologists could gain insights into the mother-
infant bond and human attachment by seeing the demands placed on infants and
mothers in foraging contexts. Anthropologists could learn what social conditions
foster risk pooling and food sharing; what kinds of knowledge hunter-gatherers have
about animal behavior and plant life; how they use this knowledge in foraging; and
how people negotiate the problems and opportunities of social life in a tiny commu-
nity of interdependent, extended families (see, e.g., Lee & DeVore, 1976; Shostak,
1981). Although commonplace now, these ideas were pathbreaking at the time. After
all, if the human mind consists primarily of a general capacity to learn, then the
particulars of the ancestral hunter-gatherer world and our prehuman history as
Miocene apes left no interesting imprint on our design. In contrast, if our minds—
as evolutionary psychologists argue—are collections of mechanisms designed to solve
the adaptive problems posed by the ancestral world, then hunter-gatherer studies and
primatology become indispensable sources of knowledge about the adaptations that
constitute modern human nature, and how our evolved psychology and soma
organizes modern social, cultural, and economic processes. DeVore’s insistence on
situating the operation of natural selection within the detailed contexts of hunter-
gatherer and nonhuman primate life was a signal contribution to the application of the
evolutionary sciences to humans.
Many members of the evolutionary research communities believed that the new
selectionist theories straightforwardly applied to humans, although others continued
to welcome the SSSM arguments that learning had insulated human life from evolu-
tionary patterning. On the one hand, human behavior exhibited many patterns that
offered ready selectionist interpretations (e.g., sex differences in the psychology of
mating), but many other phenomena resisted easy interpretation and seemed to lack
The Theoretical Foundations of Evolutionary Psychology 13

3GC01 06/09/2015 12:40:42 Page 14
clear nonhuman analogues (e.g., morality, the arts, language, culture, etc.). The result
was a rich and contradictory pluralism of ideas about how evolution relates to human
affairs—a pluralism that is still with us.
One of the most widespread approaches to emerge is what might be called fitness
teleology. Teleological explanations are found in Aristotle (invited by his observa-
tions, because he was in fact largely a biologist), and arguably constitute an evolved
mode of interpretation built into the human mind. Humans find explaining things in
terms of the ends they lead to intuitive and often sufficient (Baron-Cohen, 1995;
Dennett, 1987; Leslie, 1987, 1994). Social science theories have regularly depended
on explicitly or implicitly teleological thinking. Economics, for example, explains
choice behavior not in terms of its antecedent physical or computational causes
but in terms of how the behavior serves utility maximization (involving the
future pursuit and realization of valued goals). Of course, the scientificrevolution
originated in Renaissance mechanics, and seeks ultimately to explain everything
(non-quantum mechanical) using forward physical causality—averydifferent
explanatory system in which teleology is not admissible. Darwin outlined a forward
causal physical process—natural selection—that produces biological outcomes that
had once been attributed to natural teleological processes (Darwin, 1859). The
theoryofnaturalselectionexplainshowbiologicalsystemscouldhavesetsof
properties (adaptations) that naturally emerged because of the functions they
served. Williams (1966) mounted a systematic critique of the myriad ways teleology
had nonetheless implicitly infected evolutionary biology (where it persists in
Darwinian disguises). Computationalism assimilated the other notable class of
apparently teleological behavior in the universe—the seeming goal directedness
of living systems—to physical causation by showing how informational structures
in a regulatory system can operate in a forward causal way and yet be directed
toward goals (either apparently or actually) (Weiner, 1948). The teleological end that
seems to exist in the future as the point toward which things tend is in reality a
feedback-driven regulatory process—a regulatory process that need not but some-
times does include a representation of a goal state in the organism in the present. The
modern scientific claim would be that adaptationism and computationalism in
combination can explain by forward physical causation all events that once would
have been explained teleologically.
Yet, because the human mind evolved in the midst of biological and mental
phenomena that can be compactly and efficiently represented and predicted using
intuitive teleology, our brains evolved teleological representations as one natural
causal format: we are all implicitly drawn to explain things in teleological terms.
Hence, the implicit or explicit substrate underlying many attempts to apply Darwin-
ism to human behavior was a return to the intuition that human behavior was
explained by the ends it serves. For a Darwinian, it was argued, choices, practices,
culture, and institutions were explained to the extent that they could be interpreted as
contributing to individual (or sometimes group) reproduction: That is, the explanation
for individual human behavior is that it naturally tends toward the end of maximizing
reproduction in the present and future. This theory—Darwinism transmuted into
fitness teleology—parallels the economic view of individuals as selfish utility maxi-
mizers, except that Hamilton’s (1964) concept of inclusive fitness is substituted for the
economists’concept of utility. Both approaches implicitly assume that unbounded
rationality is possible and that the mind is a general-purpose computer that can figure
out, in any situation, what will maximize a given quantity over the long term (whether
14 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 15
utility, children, or inclusive fitness). Indeed, the concept of “learning”within the
SSSM itself tacitly invokes unbounded rationality, in that “learning”is often implicitly
treated as the tendency of the general-purpose, equipotential mind to grow—by an
unspecified and as yet undiscovered computational means—whatever functional
information-processing abilities it needs to serve its purposes, given time and expe-
rience in the task environment.
Evolutionary psychologists depart from fitness teleologists, nonmodularist cogni-
tive scientists, blank-slate learning theorists, and traditional economists (but not
neuroeconomists or behavioral economists) by arguing that neither human engineers
nor evolution can build a computational device that exhibits these forms of
unbounded rationality, because such architectures are impossible, even in principle
(for arguments, see Cosmides & Tooby, 1987; Symons, 1989, 1992; Tooby & Cosmides,
1990a, 1992). In any case, observed human behavior dramatically and systematically
departs from the sociobiological predictions of generalized fitness striving (as well as
the predictions of economic rationality and blank-slate learning abilities). To take one
simple contrast, large numbers of men will pay to have nonreproductive sex with
prostitutes they believe and hope are contracepting, but have to be paid to contribute
to sperm banks that, with high probability, may lead to offspring. More generally,
across a range of wealthy nations, those able to afford more children choose to have
fewer children—a striking disconfirmation of the prediction that humans teleologi-
cally seek to maximize reproduction or fitness (Vining, 1986). Human life is permeated
with systematic deviations away from rationally maximized child-production and kin
assistance. Humans are not mesmerized by accounts of Hutterites or Tsimane—people
who average roughly 10 children per family.
For those eager to leap directly from theories of selection pressures to predictions of
fitness maximization, there remains a missing level of causation and explanation: the
level of informational or computational adaptations. This level cannot be avoided if
the application of Darwin’s theory to humans is ever to achieve the necessary level of
scientific precision. Natural selection does not operate on behavior per se; it operates
on a systematically caused relationship between information and behavior. Running—
a behavior—is neither good nor bad. Running away from a lion can promote survival
and reproduction; running toward a lion will curtail both. To be adaptive, behavioral
regulation needs to be functionally contingent on information; for example, flee when
you see a stalking lion. But a systematic relationship between information and a
behavioral response cannot occur unless some reliably developing piece of organic
machinery causes it. These causal relations between information and behavior are
created by reliably developing neural circuits in the brain, which function as programs
that process information. By altering the neural circuitry that develops, mutations can
alter the information processing properties of these programs, creating alternative
information-behavior relationships. Selection should retain or discard alternative
circuit designs from a species’neural architecture on the basis of how well the
information-behavior relationships they produce promote the propagation of the
genetic bases of their designs. Those circuit designs that promote their own prolifera-
tion will be retained and spread, eventually becoming species-typical (or stably
frequency-dependent); those that do not will eventually disappear from the popula-
tion. The idea that the evolutionary causation of behavior would lead to rigid,
inflexible behavior is the opposite of the truth: Evolved neural architectures are
specifications of richly contingent systems for generating responses to informational
inputs.
The Theoretical Foundations of Evolutionary Psychology 15

3GC01 06/09/2015 12:40:42 Page 16
As a result of selection acting on information-behavior relationships, the human
brain is predicted to be densely packed with programs that cause intricate relation-
ships between information and behavior, including functionally specialized learning
systems, domain-specialized rules of inference, default preferences that are adjusted
by experience, complex decision rules, concepts that organize our experiences and
databases of knowledge, and vast databases of acquired information stored in
specialized memory systems—remembered episodes from our lives, encyclopedias
of plant life and animal behavior, banks of information about other people’s proclivi-
ties and preferences, and so on. All these programs and the databases they create can
be called on in different combinations to elicit a dazzling variety of behavioral
responses. These responses are themselves information, subsequently ingested by
the same evolved programs, in endless cycles that produce complex eddies, currents,
and even singularities in individual, social, and cultural life. To get a genuine purchase
on human behavior and society, researchers need to know the architecture of each of
these evolved programs. Knowing the selection pressures will not be enough. Our
behavior is not a direct response to selection pressures or to a “striving”to increase our
reproduction.
Hence, one of several reasons that evolutionary psychology is distinct from the
fitness-teleological branch of human sociobiology and other similar approaches lies in
its rejection of fitness maximization as an explanation for behavior (Cosmides &
Tooby, 1987; Daly & Wilson, 1988; Symons, 1987, 1989, 1992; Tooby & Cosmides,
1990a, 1992). The relative degree of fitness promotion under ancestral conditions is
simply the design criterion by which alternative mutant designs were sorted in the
evolutionary past. (The causal role that fitness plays in the present is in changing the
relative frequencies of alternative designs with respect to future generations.)
Although organisms sometimes appear to be pursuing fitness on behalf of their
genes, in reality they are executing the evolved circuit logic built into their neural
programs, regardless of whether this corresponds to current fitness maximization.
Organisms are adaptation executers, not fitness pursuers. Mapping the computational
architecture of the mechanisms will give a precise theory of behavior, whereas relying
on predictions derived from fitness maximization will give a very impoverished and
unreliable set of predictions about behavioral dynamics.
To summarize, evolutionary psychology’s focus on psychological mechanisms as
evolved programs was motivated by new developments from a series of different fields:
Advance 1: The cognitive revolution was providing, for the first time in human
history, a precise language for describing mental mechanisms as programs that
process information. Galileo’s discovery that mathematics provided a precise
language for expressing the mechanical and physical relationships enabled the
birth of modern physics. Analogously, cognitive scientists’discovery that compu-
tational-informational formalisms provide a precise language for describing the
design, properties, regulatory architecture, and operation of psychological mecha-
nisms (and developmental regulation) enables a modern science of mind (and its
physical basis). Computational language is not just a convenience for modeling
anything with complex dynamics. The brain’s evolved function is inherently and
fundamentally computational—to use information to adaptively regulate the body
and behavior—so computational and informational formalisms are by their nature
the most appropriate to capture the functional design of behavior regulation.
16 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 17
Advance 2: Advances in paleoanthropology, hunter-gatherer studies, primatology,
and behavioral ecology were providing data about the adaptive problems our
ancestors had to solve to survive and reproduce and the environments in which
they did so.
Advance 3: Research in animal behavior, linguistics, and neuropsychology was
showing that the mind is not a blank slate, passively recording the world. Orga-
nisms come “factory-equipped”with knowledge about the world, which allows
them to learn some relationships easily and others only with great effort, if at all.
Skinner’s hypothesis—that there is one keylearning process governed by reward
and punishment—was wrong.
Advance 4: Evolutionary biology was revolutionized by (a) being placed on a more
rigorous, formal foundation of replicator dynamics (e.g., Hamilton, 1964; Maynard
Smith, 1982; Williams, 1966), leading to the derivation of a diversity of powerful
selectionist theories, and by the development of adaptationism, which includes the
analytic tools to recognize and differentiate adaptations, from byproducts, and
stochastically generated evolutionary noise (Williams, 1966). Enduring selection
pressures (recurrent adaptive problems), operating over evolutionary time within
sets of enduring environmental regularities, act to construct in species reliably
developing solutions (adaptations) to their enduring adaptive problems. Evolu-
tionary change involves the change in a population’s gene frequencies, and those
environmental characteristics that are transient and variable cannot, by their very
nature, systematically push gene frequencies directionally upward for long enough
to cumulatively produce complex functional species-typical design. Hence adap-
tationists necessarily emphasize the role that a species-particular history of endur-
ing selection pressures and environmental regularities plays in explaining complex
functional design (see the discussion of the environment of evolutionary adaptedness or
EEA following). The composite of enduring selection pressures (the EEA) that
pushed the alleles underlying adaptation upward to stably high frequencies are
that specific part of the past that caused the adaptation and hence explains its
existence and design.
Ethology had brought together Advances 2 and 3, sociobiology had connected
Advances 2 and 4, sometimes with 3; nativist cognitive science connected Advances
1 and 3, but neglected and still shrinks from Advances 2 and 4. Standard cognitive
neuroscience partially and erratically accepts 1 and 3, but omits 2 and 4. Aside from
cognitive approaches, the rest of psychology lacks much of Advance 1, most of
Advance 3, and all of Advances 2 and 4. Evolutionary anthropology appreciates
Advances 2 and 4, but neglects 1 and 3. Social anthropology and sociology lack all
four. So it goes. If one counts the adaptationist/computationalist resolution of the
nature-nurture issue as a critical advance, the situation is even bleaker.
We thought these new developments could be painstakingly pieced together into
an integrated framework that successfully addressed the difficulties that had plagued
evolutionary and nonevolutionary approaches alike. The reason that the synthesis had
not emerged earlier in the century was because the key concepts and theories (e.g.,
adaptationism, computationalism, etc.) were scattered across fields that were institu-
tionally and intellectually distant from each other. Consequently, relatively few were
in the lucky position of being professionally equipped to see all the necessary
The Theoretical Foundations of Evolutionary Psychology 17
keylearning

3GC01 06/09/2015 12:40:42 Page 18
connections at once. This limited the field’s initial appeal, because what seems self-
evident from the synoptic vantage point seems esoteric, pedantic, or cultish (and
immoral) from other vantage points. Nevertheless, those researchers working along
these and similar lines were confident that by bringing all four advances together, the
evolutionary sciences could be united with the computationalist revolution in a way
that provided a framework not only for psychology but for all of the social and
behavioral sciences. To signal its distinctiveness from other approaches, the field was
named evolutionary psychology.
4
Its long-term goal is to eradicate disciplinary bounda-
ries, and unify the evolutionary, genetic, neural, cognitive, psychological, behavioral,
and social sciences, because the idea that these are different fields is a sociological
vestige rooted in the isolated perspectives native to the independent disciplines when
they were founded. Reality has no such boundaries, and the eventual theoretical
unification of these fields should reflect the undivided nature of the reality we are
studying.
EVOLUTIONARY PSYCHOLOGY
Like other cognitive scientists, when evolutionary psychologists refer to the mind, they
mean the set of information processing devices, embodied in neural tissue, that is
responsible for all conscious and nonconscious mental activity, that generates all
behavior, and that regulates the body. Like other psychologists, evolutionary psy-
chologists test hypotheses about the design of these computational devices using
laboratory methods from experimental cognitive and social psychology, developmen-
tal psychology, experimental economics, cognitive neuroscience, and cross-cultural
fieldwork.
The primary tool that allows evolutionary psychologists to go beyond traditional
psychologists in studying the mind is that they take full advantage in their research of
4
We sometimes read that the term evolutionary psychology is simply sociobiology, with the name changed to
avoid the bad political press that sociobiology had received. Although it is amusing (given the record) to be
accused of ducking controversy, these claims are historically and substantively wrong. In the first place,
evolutionary psychologists are generally admirers and defenders of sociobiology (or behavioral ecology, or
evolutionary ecology). It has been the most useful and most sophisticated branch of modern evolutionary
biology, and various evolutionary psychologists have themselves made contributions to this literature.
Nonetheless, the lengthy and intense debates about how to apply evolution to behavior made it increasingly
clear that markedly opposed views needed different labels if any theoretical and empirical project was to be
clearly understood. In the 1980s, Martin Daly, Margo Wilson, Don Symons, John Tooby, Leda Cosmides, and
David Buss had many discussions about what to call this new field, some at Daly and Wilson’s kangaroo rat
field site in Palm Desert, some in Santa Barbara, and some at the Center for Advanced Study in the
Behavioral Sciences. Politics and the press did not enter these discussions, and of course we anticipated
(correctly) that the same content-free ad hominem attacks would pursue us throughout our careers. What we
did discuss was that this new field focused on characterizing the adaptations comprising the psychological/
developmental architecture—whereas sociobiology had not. Sociobiology had focused mostly on selection-
ist theories, with no consideration of the computational level, and little interest in mapping psychological
mechanisms. Both the subject matter of evolutionary psychology and the theoretical commitments were
simply different from that of sociobiology, in the same way that sociobiology was quite different from the
ethology that preceded it and in the same way that cognitive psychology was different from behaviorist
psychology—necessitating a new name in each case.
18 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 19
an overlooked reality: The programs comprising the human mind were designed by
natural selection to solve the adaptive problems regularly faced by our hunter-
gatherer ancestors—problems such as finding a mate, cooperating with others,
hunting, gathering, protecting children, navigating, avoiding predators, avoiding
exploitation, and so on. Knowing this allows evolutionary psychologists to approach
the study of the mind like an engineer. You start by carefully specifying an adaptive
information processing problem; then you do a task analysis of that problem. A task
analysis consists of identifying what properties a program would have to have to solve
that problem well. This approach allows you to generate hypotheses about the
structure of the programs that comprise the mind, which can then be tested. Indeed,
evolutionary psychology is unique among theoretical orientations in psychology in
the degree to which it derives from independently established theories principled
predictions about previously unknown aspects of the species-typical psychological
architectures of humans and other species (see, e.g., Buss, 1999; Daly & Wilson, 1988;
Gaulin, 1995; Symons, 1979).
From this point of view, there are precise causal connections that link the four
developments discussed earlier into a coherent framework for thinking about human
nature and society (Tooby & Cosmides, 1992):
•Each organ in the body evolved to serve a function: The intestines digest, the
heart pumps blood, and the liver detoxifies poisons. The brain’s evolved function
is to extract information from the environment and use that information to
generate behavior and regulate physiology. Hence, the brain is not just like a
computer. It is a computer—that is, a physical system that was designed to
process information (Advance 1). Its programs were designed not by an engi-
neer, but by natural selection, a causal process that retains and discards design
features based on how well they solved adaptive problems in past environments
(Advance 4).
The fact that the brain processes information is not an accidental side effect of
some metabolic process. The brain was designed by natural selection to be a
computer. Therefore, if you want to describe its operation in a way that captures
its evolved function, you need to think of it as composed of programs that
process information. The question then becomes: What programs are to be found
in the human brain? What are the reliably developing, species-typical programs
that, taken together, comprise the human mind?
•Individual behavior is generated by this evolved computer, in response to
information that it extracts from the internal and external environment
(including the social environment, Advance 1). To understand an individual’s
behavior, therefore, you need to know both the information that the person
registered and the structure of the programs that generated his or her
behavior.
•The programs that comprise the human brain were sculpted over evolutionary
time by the ancestral environments and selection pressures experienced by the
hunter-gatherers from whom we are descended (Advances 2 and 4). Each evolved
program exists because it produced behavior that promoted the survival and
reproduction of our ancestors better than alternative programs that arose during
human evolutionary history. Evolutionary psychologists emphasize hunter-gath-
erer life because the evolutionary process is slow—it takes hundreds of genera-
tions to build a program of any complexity. The industrial revolution—even the
The Theoretical Foundations of Evolutionary Psychology 19

3GC01 06/09/2015 12:40:42 Page 20
agricultural revolution—is too brief a period to have selected for new neuro-
computational programs of any complexity.
5
•Although the behavior our evolved programs generate would, on average, have
been adaptive (reproduction promoting) in ancestral environments, there is no
guarantee that it will be so now. Modern environments differ importantly from
ancestral ones, particularly when it comes to social behavior. We no longer live in
small, face-to-face societies, in seminomadic bands typically of 50 to 150 people,
many of whom were close relatives. Yet, our cognitive programs were designed
for that social world.
•Perhaps most importantly, natural selection will ensure that the brain is com-
posed of many different programs, many (or all) of which will be specialized for
solving their own corresponding adaptive problems. That is, the evolutionary
process will not produce a predominantly general-purpose, equipotential,
domain-general architecture (Advance 3).
In fact, this is a ubiquitous engineering outcome. The existence of recurrent
computational problems leads to functionally specialized application software.
For example, the demand for effective word processing and good digital music
playback led to different application programs because many of the causal
design features that make a program an effective word processing program are
different from those that make a program a good digital music player. Indeed,
the greater the number of functionally specialized programs (or subroutines)
your computer has installed, the more intelligent your computer is, and the more
things it can accomplish. The same is true for organisms. Armed with this
insight, we can lay to rest the myth that the more evolved organization the
human mind has, the more inflexible its response. Interpreting the emotional
expressions of others, seeing beauty, learning language, loving your child—all
these enhancements to human mental life are made possible by specialized
neural programs built by natural selection.
To survive and reproduce reliably as hunter-gatherers required the solution of
large and diverse arrays of adaptive information-processing problems. These
ranged from predator vigilance and prey stalking to plant gathering, mate
selection, childbirth, parental care, coalition formation, and disease avoidance.
Design features that make a program good at choosing nutritious foods, for
example, are ill suited for finding a fertile mate or recognizing free riders. Some
sets of problems would have required differentiated computational solutions.
This difference can be most clearly seen by using results from evolutionary
game theory (Advance 4) and data about ancestral environments (Advance 2) to
define adaptive problems and then carefully dissecting the computational
requirements of any program capable of solving those problems. For example,
5
Simple, unidimensional traits, caused by quantitative genetic variation (e.g., taller, shorter), can be adjusted
in less time; see Tooby and Cosmides, 1990b. Moreover, intense selection pressures, such as those caused by
diseases (e.g. malaria) or new food sources (milk from domesticated animals) can propel some alleles rapidly
upward in frequency on a timescale of centuries. For example, all mammals have the adaptations to digest
milk in infancy and then lose it after weaning, but some human populations who get milk from livestock
benefited from the tweaking of the lactose-digesting enzyme production system so that the preexisting
ability to digest milk is maintained into adulthood. In contrast, despite being surrounded, for millions of
years, by forests of sugar (cellulose) whose digestion would have prevented all starvation, no humans have
evolved the appropriate enzymes to break down the beta acetal linkages that prevent the digestion of
cellulose. Complex adaptations are difficult to evolve rapidly.
20 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 21
game theoretic analyses of conditional helping show that programs designed for
logical reasoning would be poorly designed for detecting cheaters in social
exchange and vice versa; this incommensurability selected for programs that are
functionally specialized for reasoning about reciprocity or exchange (Cosmides &
Tooby, Chapter 20, this volume).
•Finally, descriptions of the computational architecture of our evolved mecha-
nisms allows a systematic understanding of cultural and social phenomena. The
mind is not like a tape recorder, passively recording the world but imparting no
content of its own. Domain-specific programs organize our experiences, create
our inferences, inject certain recurrent concepts and motivations into our mental
life, give us our passions, and provide cross-culturally universal frames of
meaning that allow us to understand the actions and intentions of others.
They invite us to think certain kinds of thoughts; they make certain ideas,
feelings, and reactions seem reasonable, interesting, and memorable. Conse-
quently, they play a key role in determining which ideas and customs will easily
spread from mind to mind and which will not (Boyer, 2001; Sperber, 1994, 1996;
Tooby & Cosmides, 1992). That is, they play a crucial role in shaping human
culture.
Instincts are often thought of as the opposite of reasoning, decision- making, and
learning. But the reasoning, decision-making, and learning programs that evolu-
tionary psychologists have been discovering (a) are complexly specialized for solving
an adaptive problem, (b) reliably develop in all normal human beings, (c) develop
without any conscious effort and in the absence of formal instruction, (d) are applied
without any awareness of their underlying logic, and (e) are distinct from more
general abilities to process information or behave intelligently. In other words, they
have all the hallmarks of what we usually think of as an instinct (Pinker, 1994). In fact,
we can think of these specialized circuits as reasoning instincts, decision instincts, and
learning instincts. They make certain kinds of inferences and decisions just as easy,
effortless, and natural to us as humans as catching flies is to a frog or burrowing is to
a mole.
Consider this example from the work of Simon Baron-Cohen (1995). Like adults,
normal 4-year-olds easily and automatically note eye direction in others, and use it to
make inferences about the mental states of the gazer. For example, 4-year-olds, like
adults, infer that, when presented with an array of candy, the gazer wants the
particular candy he or she is looking at. Children with autism do not make this
inference. Although children with this developmental disorder can compute eye
direction correctly, they cannot use that information to infer what someone wants.
Normal individuals know, spontaneously and with no mental effort, that the person
wants the candy he or she is looking at. This is so obvious to us that it hardly seems to
require an inference at all. It is just common sense. But “common sense”is caused: It is
produced by neurocomputational mechanisms. To infer a mental state (wanting) from
information about eye direction requires a computation. There is an inference circuit—
a reasoning instinct—that produces this inference. When the circuit that does this
computation is broken or fails to develop, the inference cannot be made. Those with
autism fail this task because they lack this reasoning instinct, even though they often
acquire very sophisticated competences of other sorts. If the mind consisted of a
domain-general knowledge-acquisition system, narrow impairments of this kind
would not be possible.
The Theoretical Foundations of Evolutionary Psychology 21
20,
this
volume).
s allows a

3GC01 06/09/2015 12:40:42 Page 22
Reasoning instincts are invisible to our intuitions, even as they generate them. They
are no more accessible to consciousness than our retinal cells and line detectors but are
just as important in manufacturing our perceptions of the world. As a species, we have
been blind to the existence of these instincts, not because we lack them but precisely
because they work so well. Because they process information so effortlessly and
automatically, their operation disappears unnoticed into the background. Moreover,
these instincts structure our thought and experience so powerfully we mistake their
products for features of the external world: Color, beauty, status, friendship, charm—
all are computed by the mind and then experienced as if they were objective properties
of the objects we attribute them to. These mechanisms limit our sense of behavioral
possibility to choices people commonly make, shielding us from seeing how complex
and regulated the mechanics of choice is. Indeed, these mechanisms make it difficult to
imagine how things could be otherwise. As a result, we take normal behavior for
granted: We do not realize that normal behavior needs to be explained at all.
As behavioral scientists, we need corrective lenses to overcome our instinct
blindness. The brain is fantastically complex, packed with programs, most of which
are currently unknown to science. Theories of adaptive function can serve as corrective
lenses for psychologists, allowing us to see computational problems that are invisible
to human intuition. When carefully thought out, these functional theories can lead us
to look for programs in the brain that no one had previously suspected.
PRINCIPLES OF ORGANIC DESIGN
Biology is the study of organisms, and psychology is—in a fundamental sense—a
branch of biology. It is the study of the evolved designs of the behavior-regulating
tissues of organisms. To be effective researchers, psychologists will need to become at
least minimally acquainted with the principles of organic design.
NATURAL SELECTION ISANENGINEER THAT DESIGNS ORGANIC MACHINES
The phenomenon that Darwin was trying to explain is the presence of functional
organization in living systems—the kind of organization found in artifacts, such as
clocks, spectacles, or carriages; indeed, the kind of organization that appeared to be
designed by an intelligent engineer to solve a problem. Darwin realized that orga-
nisms can be thought of as self-reproducing machines. What distinguishes living from
nonliving machines is reproduction: the presence in a machine of devices (organized
components) that cause it to produce new and similarly reproducing machines. Given
a population of living machines, this property—self-reproduction—drives a system of
positive and negative feedback—natural selection—that can explain the remarkable fit
between the design of organisms and the problems they must solve to survive and
reproduce.
In contrast to human-made machines, which are designed by inventors, living
machines acquire their intricate functional design over immense lengths of time, as a
consequence of the fact that they reproduce themselves. Indeed, modern Darwinism
has an elegant deductive structure that logically follows from Darwin’s initial insight
that reproduction is the defining property of life:
When an organism reproduces, genes that cause the development of its design
features are introduced into its offspring. But the replication of the design of the
22 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 23
parental machine is not always error free. As a result, randomly modified designs (i.e.,
mutants) are introduced into populations of reproducers. Because living machines are
already exactingly organized so that they cause the otherwise improbable outcome of
constructing offspring machines, random modifications will usually introduce dis-
ruptions into the complex sequence of actions necessary for self-reproduction. Con-
sequently, most newly modified but now defective designs will remove themselves
from the population: a case of negative feedback.
However, a small number of these random design modifications will, by chance,
improve the system’s machinery for causing its own reproduction. Such improved
designs (by definition) cause their own increasing frequency in the population: a case
of positive feedback.
This increase continues until (usually) such modified designs outreproduce and
thereby replace the alternative designs in the population, leading to a new species-
standard (or population-standard) design: a new retinal design, or blood cell, or
reasoning circuit, or food preference ordering. After such an event, the population
of reproducing machines is different from the ancestral population. The population
has taken a step “uphill”toward a greater degree of functional organization for
reproduction than it had previously. Over the long run, down chains of descent,
this feedback cycle pushes designs through state-space toward increasingly well-
engineered—and increasingly improbable—functional arrangements. These arrange-
ments are functional in a specific sense: The elements are well organized to cause their
own reproduction in the environment in which the species evolved.
For example, if a mutation appeared that caused individuals to find family
members sexually repugnant, they would be less likely to conceive children incestu-
ously. They would produce children with fewer genetic diseases, and more of these
children would mature and reproduce than would the children of those who were not
averse to incest. Such an incest-avoiding design would produce a larger set of healthy
children every generation, down the generations. By promoting the reproduction of its
bearers, the incest-avoiding circuit thereby promotes its own spread over the genera-
tions, until it eventually replaces the earlier-model sexual circuitry and becomes a
universal feature of that species’design (for a map of the design of this system, see
Lieberman, Tooby, & Cosmides, 2007). This spontaneous feedback process—natural
selection—causes functional organization to emerge naturally, without the interven-
tion of an intelligent designer or supernatural forces.
Genes and Design Self-reproducing systems could not exist unless there were
adaptations that conserved the functional design against entropy from one generation
to the next. Genes are the means by which functional design features replicate
themselves from parent to offspring. They can be thought of as particles of design.
These elements are transmitted from parent to offspring and together with stable
features of an environment, cause the organism to develop some design features and
not others. Genes have two primary ways they can propagate themselves: by
increasing the probability that offspring will be produced by the organism in which
they are situated or by increasing reproduction in others who are more likely than
random members of the population to carry the same gene.
An individual’s genetic relatives carry some of the same genes, by virtue of having
received some of the same genes from a recent common ancestor. Thus, a gene in an
individual that causes an increase in the reproductive rate of that individual’s kin will,
by so doing, tend to increase its own frequency in the population. A circuit that
The Theoretical Foundations of Evolutionary Psychology 23

3GC01 06/09/2015 12:40:42 Page 24
motivates individuals to help feed their sisters and brothers, if they are in sufficiently
greater need, is an example of a program that increases kin reproduction (for evidence
about the design of such a system, see Lieberman et al., 2007). As Hamilton (1964)
pointed out, design features that promote both direct reproduction and kin reproduc-
tion and that make efficient trade-offs between the two will replace those that do not
(a process called kin selection).
Reproduction and Function How well a design feature systematically promotes direct
and kin reproduction is the bizarre but real engineering criterion determining whether
a specific design feature will be added to or discarded from a species’design.
The concept of adaptive behavior can now be defined with precision. Adaptive
behavior, in the evolutionary sense, is (in the general case) behavior that increased the
frequency of the alleles underlying the behavior; typically, this means behavior that
systematically promoted the net lifetime reproduction of the individual and/or (with
appropriate trade-offs) that individual’s genetic relatives. By promoting the replica-
tion of the genes that built them, circuits that—systematically and over many
generations—cause adaptive behavior become incorporated into a species’neural
design. In contrast, behavior that undermines the reproduction of the individual or his
or her genetic relatives removes the circuits causing those behaviors from the species.
Such behavior is maladaptive.
Evolutionists analyze how design features are organized (in ancestral environ-
ments) to contribute to lifetime kin-weighted reproduction because reproduction was
the final causal pathway through which a functionally improved design feature
caused itself to increase in frequency until it became standard equipment in all (or
in an enduring subset of) ordinary members of the species.
Adaptive Problems Select for Adaptations Darwin’s detailed studies of plants and
animals revealed complex structures composed of parts that appeared to be organized
to overcome reproductive obstacles (e.g., the presence of predators) or to take
advantage of reproductive opportunities (e.g., the presence of fertile mates). Enduring
conditions in the world that create reproductive opportunities or obstacles constitute
adaptive problems, such as the presence of pathogens, variance in the food supply, the
vulnerability of infants, or the presence of family in an individual’s social group.
Adaptive problems have two defining characteristics. First, they are conditions or
cause-and-effect relationships that were regularly encountered by members of a
population or species, and that recurred across sufficiently many generations such
that natural selection has enough time to design adaptations in response. Second, they
are that subset of enduring relationships that could, in principle, be exploited by some
property of an organism to increase its reproduction or the reproduction of its
relatives. Alternative designs are retained or discarded by natural selection on the
basis of how well they function as solutions to adaptive problems.
Over evolutionary time, more and more design features accumulate that fit together
to form an integrated structure or device (e.g., a retina, a claw, an incest avoidance
program) that is well engineered to solve its particular adaptive problem. Such a
structure or device is called an adaptation. Indeed, an organism can be thought of as a
collection of adaptations, together with the engineering byproducts of adaptations,
and evolutionary noise. The functional subcomponents of the ear, hand, intestines,
uterus, or circulatory system are examples. Each of these adaptations exists in the
human design now because it contributed to the process of self- and kin reproduction
24 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:42 Page 25
in the ancestral past. Adaptive problems are the only kind of problem that natural
selection can design machinery for solving. They are the source of and explanation of
our evolved functional design.
The Environment of Evolutionary Adaptedness One key to understanding the func-
tional architecture of the mind is to remember that its programs were not selected for
because they solved the problems faced by modern humans. Instead, they were
shaped by how well they solved adaptive problems among our hunter-gatherer
ancestors. The second key is to understand that the developmental processes that
build each program, as well as each program in its mature state, evolved to use
information and conditions that were reliably present in ancestral environments. The
design of each adaptation assumes the presence of certain background conditions
and operates as a successful problem solver only when those conditions are met. The
environment of evolutionary adaptedness (EEA) refers jointly to the problems hunter-
gatherers had to solve and the conditions under which they solved them (including
their developmental environment).
Although the hominin line is thought to have originated in African open wood-
lands, the EEA is not a particular place or time. The EEA for a given adaptation is the
statistical composite of the enduring selection pressures or cause-and-effect relation-
ships that pushed the alleles underlying an adaptation systematically upward in
frequency until they became species-typical or reached a frequency-dependent equi-
librium (most adaptations are species-typical; see Hagen, Chapter 4, this volume).
Because the coordinated fixation of alleles at different loci takes time, complex
adaptations reflect enduring features of the ancestral world. The adaptation is the
consequence of the EEA, and so the structure of the adaptation reflects the structure of
the EEA. The adaptation evolved so that when it interacted with the stable features of
the ancestral task environment, their interaction systematically promoted fitness (i.e.,
solves an adaptive problem). The concept of the EEA is essential to Darwinism, but its
formalization was prompted by the evolutionary analysis of humans because human
environments have changed more dramatically than the environments most other
species occupy. The research problems faced by most biologists do not require them to
distinguish the modern environment from a species’ancestral environment. Because
adaptations evolved and assumed their modern form at different times and because
different aspects of the environment were relevant to the design of each, the EEA for
one adaptation may be somewhat different from the EEA for another. Conditions of
terrestrial illumination, which form (part of) the EEA for the vertebrate eye, remained
relatively constant for hundreds of millions of years—and can still be observed by
turning off all artificial lights. In contrast, the social and foraging conditions that
formed (part of) the EEA that selected for neural programs that cause human males to
provision and care for their offspring (under certain conditions) is almost certainly less
than 2 million years old.
When a program is operating outside the envelope of ancestral conditions that
selected for its design, it may look like a poorly engineered problem solver. Efficient
foraging, for example, requires good probability judgments, yet laboratory data
suggested that people are poor intuitive statisticians, incapable of making simple
inferences about conditional probabilities (Kahneman, Slovic, & Tversky, 1982).
Evolutionary psychologists recognized that these findings were problematic, given
that birds and bees solve similar problems with ease. The paradox evaporates when
you consider the EEA for probability judgment. Behavioral ecologists presented birds
The Theoretical Foundations of Evolutionary Psychology 25

3GC01 06/09/2015 12:40:43 Page 26
and bees with information in ecologically valid formats; psychologists studying
humans did not.
Being mindful of the EEA concept changes how research is designed and what is
discovered. Giving people probability information in the form of absolute frequen-
cies—an ecologically valid format for hunter-gatherers—reveals the presence of
mechanisms that generate sound Bayesian inferences (Brase, Cosmides, & Tooby,
1998; Cosmides & Tooby, 1996a; Gigerenzer, 1991; Gigerenzer, Todd, & the ABC
Group, 1999). Indeed, EEA-minded research on judgment under uncertainty is now
showing that the human mind is equipped with a toolbox of “fast-and-frugal
heuristics,”each designed to make well-calibrated judgments quickly on the basis
of limited information (Gigerenzer & Selten, 2002; Gigerenzer et al., 1999; Todd,
Hertwig, & Hoffrage, Chapter 37, this Handbook, Volume 2). These procedures are
ecologically rational, providing good solutions when operating in the task environments
for which they evolved (Cosmides & Tooby, 1996b; Delton, Krasnow, Cosmides, &
Tooby, 2011; Tooby, Cosmides, & Barrett, 2005, Tooby & Cosmides, in press).
Knowing the Past It is often argued that we can know nothing about the past that is
relevant to psychology because behavior doesn’t fossilize. Thus, the whole field of
evolutionary psychology is claimed to rest on uncertain speculation or conjecture.
In reality, we know with certainty thousands of important things about our ancestors
and the world they inhabited, many of which can be useful in guiding psychological
research. Some of these should be obvious, although their implications may not be. For
example, it is a certainty that our ancestors lived in a world in which certain principles
of physics governed the motions of objects: facts that allowed Shepard (1984, 1987) to
discover how the mind represents the motion of objects, both in perception and
imagination. It is equally certain that hominins had eyes, looked at what interested
them, and absorbed information about what they were looking at, making eye-gaze
direction informative to onlookers: facts that helped Baron-Cohen (1995) and others to
create a far-reaching research program on the cognitive basis of mind reading, the
ability to infer the mental states of others. It is certain that our ancestors, like other
Old World primates, nursed; had two sexes; chose mates; had color vision calibrated
to the spectral properties of sunlight; lived in a biotic environment with predatory
cats, venomous snakes, and spiders; were predated on; bled when wounded; were
incapacitated from injuries; were vulnerable to a large variety of parasites and
pathogens; and had deleterious recessives rendering them subject to inbreeding
depression if they mated with siblings. All these conditions (along with tens of
thousands of others) are known, and all pose adaptive problems. By considering
these selection pressures, a careful, well-informed, intelligent researcher can develop
plausible, testable theories of the adaptations that arose in response to them. Selection
would not plausibly have built an equipotential cognitive architecture that had to
encounter the world as if it were unprepared for functionally significant sets of
evolutionarily recurrent relationships. It is remarkable that such a model is so
vigorously defended.
By triangulating the work of researchers in many disciplines, many other sound
inferences can be made. Evolutionary psychologists, behavioral ecologists and evolu-
tionary biologists have already created a library of sophisticated models of the
selection pressures, strategies, and trade-offs that characterize fundamental adaptive
problems (Advance 4), which they use in studying processes of attention, memory,
decision-making, and learning in nonhuman animals. Which model is applicable for a
26 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 27
given species depends on certain key life-history parameters. Findings from paleo-
anthropology, hunter-gatherer archaeology, and studies of living hunter-gatherer
populations locate humans in this theoretical landscape by filling in the critical
parameter values (Advance 2). Ancestral hominins were ground-living primates;
omnivores,
6
exposed to a wide variety of plant toxins and meat-borne bacteria and
fungi; they had a sexual division of labor involving differential rates of hunting and
gathering. They were mammals with altricial young, long periods of biparental
investment in offspring, enduring male-female mateships, and an extended period
of physiologically obligatory female investment in pregnancy and lactation. They
were a long-lived, low-fecundity species in which variance in male reproductive
success was higher than variance in female reproductive success. They lived in small,
typically nomadic, kin-based bands often of 20 to 150; they would rarely have seen
more than 1,000 people at one time; they had only modest opportunities to store
provisions for the future; they engaged in cooperative hunting, raiding, defense, and
aggressive coalitions; and they made tools and engaged in extensive amounts of
implicit and explicit exchange, food-sharing, cooperation, and deferred reciprocation.
When these parameters are combined with formal models from evolutionary biology
and behavioral ecology, a reasonably consistent picture of ancestral life begins to
appear (e.g., Tooby & DeVore, 1987). From this, researchers can refine theories of
adaptive problems, develop models of their computational requirements, and test for
the presence of mechanisms equipped with design features that satisfy these require-
ments. Most chapters in this volume provide examples of this process.
Many adaptive problems can be further illuminated by the use of evolutionary
game theory (see Cosmides & Tooby, Chapter 25, this Handbook, Volume 2) and/or
optimal foraging models. For example, variance in the food supply can be buffered
through food sharing, a method of pooling risk, which is stable only when the variance
is primarily due to luck rather than effort. Studies of modern hunter-gatherers have
allowed quantitative estimates of how much variance there is in successfully finding
different kinds of foods; for example, among the Ache of Paraguay, meat and honey
are high-variance foods even for skilled foragers, whereas the variance in gathering
vegetable foods is low and comes from effort rather than luck. As might be predicted
from an analysis of the adaptive problems posed by variance in the food supply, Ache
hunter-gatherers risk-pool with meat and honey by sharing widely at the band level,
but they share gathered vegetable foods only within nuclear families (Kaplan & Hill,
1985). This analysis suggests that our minds house at least two different decision rules
for sharing, each creating a different sense of what is appropriate or fair, and each
triggered by a different experience of variance. This, in turn, led to the successful
prediction that we have mechanisms designed to be effectively calibrated to variance
and its causes (e.g., Rode, Cosmides, Hell, & Tooby, 1999; Wang, 2002). Indeed, the
“irrational”risk aversion posited in Kahneman and Tversky’s (1979) prospect theory
can be replaced by an evolutionarily revised prospect theory (Rode et al., 1999), in
which individuals can be shown to be adaptively risk-seeking or adaptively risk
averse depending on their need level and the probability distribution they faced.
Knowledge of ancestral life, ancestral conditions, and ancestral adaptive problems
are like treasure maps that can supercharge the discovery of previously unknown
6
Fossil sites show extensive processing sites for animal products. Large East African woodland primates
hunt and eat meat. Hunter-gatherers are observed to get a major fraction of their diet from hunting, and for
hunting to be a dispropoportionately male activity not only in humans but in chimpanzees and baboons.
The Theoretical Foundations of Evolutionary Psychology 27

3GC01 06/09/2015 12:40:43 Page 28
psychological and developmental mechanisms. Although behavioral scientists can be
certain about a huge inventory of facts about the ancestral world that has not yet been
harnessed to guide psychological research, certainty about the past is not necessary for
building better hypotheses. We can derive valuable experimental hypotheses from
likely rather than certain features of the ancestral world. At worst, such a hypothesis is
no more likely to be falsified than the hypotheses advanced by nonevolutionary
researchers, who have no principled source from which to derive their hypotheses and
must rely on the random walk of blind empiricism. There are certainly many features
of the ancestral world about which we are completely ignorant: These features simply
do not form the basis for experiments. Traditional research programs involve pro-
ceeding either with blind empiricism, on the basis of no theory of function, or to
proceed guided by necessarily false theories of function. It is difficult to see any valid
argument for doing either, because (a) random empirical tests are unlikely to
efficiently guide researchers to the correct experimental procedures that are capable
of detecting and mapping of complex neural or developmental programs, and (b)
invalid nonevolutionary theories are even less likely to be productive. (The physics of
entropy together with replicator dynamics tell us that the only origin of complex
functional design in undomesticated species is natural selection; hence, all correct
theories of function will be and must be evolutionary.)
PSYCHOLOGY ISREVERSE ENGINEERING
As engineers go, natural selection is superlative. It has produced exquisitely engi-
neered biological machines—the vertebrate eye, the four-chambered heart, the liver,
and the immune system—whose performance at solving problems is unrivaled by any
machine yet designed by humans. (Consider the poor quality of machine vision
compared to evolved vision, artificial pacemakers compared to the evolved system
regulating the heart, pharmaceuticals with their negative side effects compared to the
body’s immune and detoxification systems.)
Psychologists—evolutionary or otherwise—are engineers working in reverse. The
brain/mind is a complex functional system, composed of programs whose design was
engineered by natural selection to solve specific adaptive problems. Our job is to
reverse-engineer the human brain/mind: to dissect its computational architecture into
functionally isolable information processing units—programs—and to determine how
these units operate, both computationally and physically. To arrive at the appropriate
construal, the neurocomputational and developmental architecture must be concep-
tualized as a set of parts designed to interact in such a way that they solve adaptive
problems. This conceptualization requires theories of adaptive function—engineering
specifications, which provide analyses of what would count as good design for a
particular problem. In so doing, they also provide the criteria necessary to decide
whether a property of an organism is a design feature, a functionless byproduct, or
noise.
Many Properties of Organisms Are Not Adaptations The cross-generationally recurrent
design of an organism can be partitioned into (a) adaptations, which are present
because they were selected for, (b) byproducts of adaptations, which were not
themselves targets of selection but are present because they are causally coupled
to or produced by traits that were, and (c) noise, which was injected by the stochastic
28 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 29
components of evolution. Consider, for example, that all brain-intact persons learn to
speak (or sign) the language of their surrounding community without explicit
instruction, whereas reading and writing require explicit schooling, are not mastered
by every individual, and are entirely absent from some cultures. The neural programs
that allow humans to acquire and use spoken language are adaptations, specialized by
selection for that task (Pinker, 1994; Pinker & Bloom, 1990). But once an information-
processing mechanism exists, it can be deployed in activities that are unrelated to its
original function. Because we have evolved learning mechanisms that cause language
acquisition, we can, through laborious study and schooling, learn to write and read.
But the learning mechanisms that enable these activities were not selected for because
they caused reading and writing. The ability to read and write are byproducts of
adaptations for spoken language, enabled by their causal structure. Random evolu-
tionary noise exists as well—for example, the gene variants that cause dyslexia
(difficulties with learning to read). Indeed, entropy is pervasive, and so the designs
of organisms are the product of mutation-selection balance. All organisms contain
many negative genetic mutations, on the way to being selected out, and the environ-
ments of development change, generating environmental “mutations”—changes—
that induced developmental perturbations. Moreover, organisms may only have been
exposed to an adaptive problem recently. So evolutionarily informed researchers do
not expect optimality, and are not confounded when nonoptimality is found. They
only expect that designs are to be found in regions of design-space that are vastly
better than random from a functional perspective, and that by modeling or consider-
ing “optimality”or good design, these rare regions can be identified.
Adaptations are present because of a prior history of selection. They are not defined
as any ability or trait, however rare or modern, that is beneficial by virtue of enabling a
particular individual to have more children. Suppose, for example, that a computer
programmer were to become wealthy through writing code and used that wealth to
conceive many children. This would not make computer programming, which is a
very recent cultural invention, an adaptation, nor would it mean that the cognitive
mechanisms that enable computer programming are adaptations designed for pro-
ducing computer programs. The ability to write code is a beneficial side effect of
cognitive adaptations that arose to solve entirely different problems, ones that
promoted reproduction in an ancestral past.
7
Thus, although selection creates functional organization, not all traits of organisms
are functional. In fact, most “parts”of an organism are not functional for a simple
reason: Most ways of conceptually dissecting a species’phenotype into parts will fail
to capture functional components.
8
To see the organization that exists in a complex
7
In the case of computer programming, these adaptations might include the numerical abilities that
underwrite foraging (Wynn, 1998), recursion for producing metarepresentations (Leslie, 1987), grammatical
mechanisms (Pinker, 1994), certain deductive capacities (Rips, 1994), and so on. To determine which
adaptations underwrite the ability to program computers would require cognitive experimentation aimed at
discovering which information processing mechanisms are activated when someone is engaged in this
evolutionarily novel activity. Moreover, different constellations of mechanisms might be activated when
different individuals program, precisely because there has not been enough time for natural selection to
produce an integrated design specifically for this purpose.
8
Imagine you are looking inside a television and considering ways to conceptually divide its innards into
parts. A random parsing is unlikely to isolate the functional units that allow a TV to transduce electro-
magnetic radiation into a color bitmap (its function). Indeed, most ways of dividing its insides will fail to
capture any functional components, and any such nonfunctional “parts”will be byproducts of the functional
ones (Hagen, Chapter 4, this volume).
The Theoretical Foundations of Evolutionary Psychology 29

3GC01 06/09/2015 12:40:43 Page 30
system, researchers need to be able to distinguish its functional components from the
byproducts and noise.
With a well-specified theory of an adaptive problem, researchers can identify
functional and nonfunctional parts of an organism. Of the three kinds of properties,
adaptations are the most important and illuminating because they explain why a
system has certain parts, why these participate in certain cause-and-effect relation-
ships with one another, and why they interact with the world in the way that they do.
Adaptations are problem-solving machines and can be identified using design
evidence. This entails probability judgments about the degree to which a set of design
features nonrandomly solve an independently defined ancestral adaptive problem.
DESIGN EVIDENCE
To determine a system’s adaptive function, researchers need to produce evidence of a
fit between its design and the proposed function, This requires the application of
engineering standards. As an analogy, consider the relation between design and
function in human-made artifacts. A ceramic mug is made of an insulating material
that does not dissolve or melt when it contacts hot drinks; its shape stably contains
about 8 ounces of liquid and allows a mouth access to it; and it has a heat-dissipating
handle that allows it to be lifted without burning the lifter. These properties of a mug
are design features: properties that exist because they are good solutions to the problem
of drinking hot beverages without burning your hands.
These properties are unlikely to occur together by chance. Moreover, other uses to
which mugs are put (e.g., paperweights, pencil holders) neither predict nor explain
these features (paperweights need only be heavy; pencil holders must have a
containing shape, but many materials will do—the container need not be watertight,
and no handle is needed). A mug can produce many beneficial effects, but only one of
these is its function, that is, the explanation for why it was constructed in the way that
it was. We can tell which design explanation is correct by analyzing the fit between the
mug’s design and a proposed function. Mugs have many interlocking properties that
are good solutions to the problem of drinking hot drinks, and their properties are
poorly explained by alternative theories of their function; that is how we know that
they were designed for that function. The more complex the architecture, the more
powerful design evidence can be. For example, there are many design features that can
decide whether a toaster was intended to be a vehicle, a nutrient, a cleaner, a geological
accident, a device for executing bathers, or a means for toasting slices of bread (for
discussion, see Dawkins, 1996).
In the same way, design evidence is the criterion for claiming that a property of an
organism is an adaptation, whether that property is a knee, a heart, or a neural circuit
that processes information. Does the organic machinery in question have properties
that cause it to solve an adaptive problem precisely, reliably, and economically? If
not, then its ability to solve the problem at issue may be incidental, a side effect of a
system that is well designed to perform some alternative adaptive function (Williams,
1966). For example, zoologists found that nocturnal bats have a sonar system with
many of the same intricate and interlocking features of human-engineered sonar and
radar systems, including features that make bat sonar a good design for finding insects
and avoiding obstacles at night (e.g., higher pulse rates when hunting small moving
targets than when cruising; for discussion, see Dawkins, 1986). At the same time, bat
30 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 31
sonar is poorly suited for solving most other problems (e.g., judging the relative
ripeness of fruit during the day). And there is no physical law or general metabolic
process that produces bat sonar as a side effect.
Finding and pursuing small flying food items in the dark without crashing into
things pose intricate computational problems, which very few arrangements of matter
can solve. The bat’s sonar solves these problems well. There is a tight fit between the
problems’requirements and the evolved solution. It is by virtue of this excellence in
design that we recognize finding insects and avoiding obstacles at night as the
adaptive function of bat sonar.
Researchers can identify an aspect of an organism’s physical, developmental, or
psychological structure—its phenotype—as an adaptation by showing that (a) it has
many design features that are improbably well suited to solving an ancestral adaptive
problem, (b) these phenotypic properties are unlikely to have arisen by chance alone,
and (c) they are not better explained as the byproduct of mechanisms designed to solve
some alternative adaptive problem or some more inclusive class of adaptive problem.
Finding that a reliably developing feature of the species’architecture solves an
adaptive problem with reliability, precision, efficiency, and economy is prima facie
evidence that an adaptation has been located. This is like showing that an oddly
shaped piece of metal easily opens the lock on your front door. It is almost certainly a
key designed for your door because door locks are not easily opened by random bits of
metal, by can openers or candlesticks, or even by keys designed for other doors.
To show that something is a byproduct, researchers must first establish that
something else is an adaptation (e.g., blood as an oxygen transport system) and
then show how the feature is a side effect of the adaptation (e.g., the redness of blood is
a side effect of the oxygen-carrying iron in hemoglobin). Features that are
uncoordinated with functional demands are evolutionary noise (e.g., the locations
of flecks of color in the eye).
THEORIES OF GOOD DESIGN ARE A HEURISTIC FOR DISCOVERY
If design evidence were important only for explaining why known properties of
organisms have the form that they do (i.e., why the lens of the eye is transparent rather
than opaque), its use in psychology would be limited. After all, most properties of the
human mind are currently unknown. The concept of good design for solving an
adaptive problem is important because it allows researchers to discover new mecha-
nisms within the human mind. There is a systematic method for using theories of
adaptive function and principles of good design for discovering new programs.
One starts with an adaptive problem encountered by human ancestors, including
what information would potentially have been present in past environments for
solving that problem (i.e., its information ecology). From the model of an adaptive
problem, the researcher develops a task analysis of the kinds of computations
necessary for solving that problem, concentrating on what would count as a well-
designed program given the adaptive function under consideration. Based on this task
analysis, hypotheses can be formulated about what kinds of programs might actually
have evolved. Next, their presence can be tested experimentally, using methods from
cognitive, social, and developmental psychology, cognitive neuroscience/neuro-
psychology, experimental economics, cross-cultural studies—whichever methods
are most appropriate for illuminating programs with the hypothesized properties.
The Theoretical Foundations of Evolutionary Psychology 31

3GC01 06/09/2015 12:40:43 Page 32
If the predicted properties are found, tests can be conducted to make sure they are not
better explained by alternative hypotheses about the programs responsible. Testing
includes making sure the program in question is distributed cross-culturally in the
way predicted by the theory (usually adaptations are predicted to be species-typical).
However, a universal program may often produce different expressions triggered by
different environmental or social conditions, or show local calibration by specific
circumstances.
Research on the architecture of kin detection in humans provides an example of
how this process of discovery can work (Lieberman, Tooby, & Cosmides, 2003, 2007).
Avoiding the deleterious effects of inbreeding was an important adaptive problem
faced by our hominin ancestors. The best way to avoid the costs of inbreeding is to
avoid having sex with close genetic relatives. This, in turn, requires a system for
distinguishing close genetic relatives from other individuals: a kin detection system,
which computes a kinship estimate for each individual with whom one lives in close
association. Because genetic relatedness cannot be directly observed, it is important to
consider what information relevant to estimating degrees of kinship would have been
available to a ancestral hunter-gatherers. To be useful, kinship estimates would have
to be based on cues that reliably predicted genetic relatedness in the social conditions
under which our ancestors lived. We are looking for cues that would have been stably
present across a broad variety of ancestral social conditions and habitats. For example,
hunter-gatherers often live and forage in groups that fuse and fission along nuclear
family lines, such that parents more frequently stay together with children, adult
siblings and their families maintain association, but to a lesser degree, and so on. This
would allow the cumulative duration of childhood co-residence to function as a cue to
genetic relatedness. An individual who observed his or her mother caring for another
infant (what we call maternal perinatal association) would be a more direct cue that
the infant was a sibling. A third cue might be an olfactory signature indicating
similarity of the major histocompatibility complex. Based on the stable information
structure of the ancestral world, the kin detection system is expected to evolve to
monitor ancestrally valid cues, and use them to compute a relatedness estimate (that
we call a kinship index) for each individual in the person’s social world. The kinship
index serves as an input to systems that compute the sexual value of another
individual to himself or herself: All else equal, close genetic relatives should be
assigned a lower sexual value than unrelated people. This sexual-value estimate—
another internal regulatory variable—should regulate the motivational system that
generates sexual attraction. A low kinship estimate should upregulate sexual attrac-
tion whereas a high kinship estimate should downregulate sexual attraction, perhaps
by activating disgust in response to the prospect of sex with that person. Indepen-
dently, the kinship index in one individual’s mind about a particular other individual
should regulate altruism: The higher the kinship index, the more an individual should
be motivated to sacrifice for the relative. These and other theoretically derived
predictions about the existence and architecture of the human kin detection system
were empirically confirmed, along with a parallel set of predictions about kin-directed
altruism. The two predicted cues—maternal perinatal association and duration of
childhood co-residence—regulate sexual disgust toward genetic relatives and kin-
directed altruism as well (as predicted by Hamilton, 1964). The cues used by older
siblings in detecting younger ones differ from those used by younger siblings
detecting older ones. The results are incompatible with a variety of alternative theories
that could be put forth to explain the results (e.g., Leiberman, Tooby, & Cosmides,
32 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 33
2003, 2007). So far, the pattern found holds in a variety of different cultural settings,
consistent with the hypothesis that the kin detection system develops cross-culturally
as a universal mechanism of the human mind.
Note that by starting with an adaptive problem—inbreeding avoidance—and
analyzing the computational requirements of a system that solves this problem, a
significant neurocomputational system was predicted, tested for, and discovered—a
system that was previously unknown and uninvestigated by traditional psychologists
and cognitive scientists.
It may not seem so at first glance, but notice that the kin detection system is a
learning mechanism. Its function is to learn which individuals in a person’s environment
are kin and which are not, and it is designed to make this categorization on the basis of
certain cues present during development, while ignoring others. For example, an
individual’s consciously held beliefs about who is a sibling do not predict degree of
sexual aversion, once duration of childhood coresidence is controlled for (but cor-
esidence does predict sexual aversion, controlling for beliefs about who is a sibling;
Lieberman, Tooby, & Cosmides, 2003, 2007). The kin detection system is not, however,
ageneral-purpose learning mechanism. It is highly specialized for a narrow task and has
nothing in common with mechanisms of classical and operant conditioning, the way
facts are learned in school, or any other more general-purpose method of learning.
9
NATURE AND NURTURE:ANADAPTATIONIST PERSPECTIVE
To fully understand the concept of design evidence, we need to consider how
evolutionary psychologists think about nature and nurture. Debates about the relative
contribution (as it is misleadingly put) of genes and environment during development
have been among the most contentious in psychology. The premises that underlie
these debates are flawed, yet they are so deeply entrenched that many people,
scientists and nonscientists alike, have difficulty seeing that there are better ways
to think about these issues (For an excellent, early treatment of these issues, see
Tinbergen, 1963).
Rather than there being one nature-nurture issue, there are many independent
issues. Unfortunately, they have become so tangled that most discussions in psychol-
ogy and the social sciences are hopelessly confused. We pull the major questions apart
and look at them one by one. Some of them are conceptual confusions, whereas others
are genuine scientific questions whose resolution will depend on research, rather than
on clear thinking alone.
Despite widespread belief to the contrary, evolutionary psychology is not another
swing of the nature-nurture pendulum (Tooby & Cosmides, 1992). It shatters the
traditional framework and the old categories entirely, rather than siding with any
position within the old debate. Indeed, a defining characteristic of the field is the
explicit rejection of the usual nature-nurture dichotomies—instinct versus reasoning,
innate versus learned, biological versus cultural, nativist versus environmentalist,
socially determined versus genetically determined, and so on—because they do not
correspond to the actual distinctions that need to be made in the real world.
9
It is not known how children learn facts in school—the notion that it is via some form of general-purpose
learning is an assumption, not a finding for which there is evidence. Indeed, there is starting to be evidence
that school learning piggybacks off domain-specific inference mechanisms such as being fed linguistic
representations (e.g., Hirschfeld & Gelman, 1994).
The Theoretical Foundations of Evolutionary Psychology 33

3GC01 06/09/2015 12:40:43 Page 34
Evolutionary psychologists do not see nature and nurture as existing in an explanatory
zero-sum relationship. Nonevolutionary researchers have typically assumed that
there is a spectrum in the nature-nurture debate, that it runs from the nativist extreme
(most things are “genetically determined”) to the environmentalist extreme (most
things are “environmentally determined”), and that the true position (the subject of
the debate) lies somewhere along this spectrum. But all properties of the organism
equally develop through 100% gene-environment interaction. The key point at which
the adaptationist approach pivots to a new framework for understanding develop-
ment lies in the following recognition: As will be explained in greater depth, a species’
history of selection acted over evolutionary time to organize and tune the interaction
between genes and environment to produce the reliable development of that
species’adaptations—adaptations whose program logic, in turn, specifies how envi-
ronmental inputs are operated on to become behavioral outputs. (Whereas selection
acts antientropically to functionally tune gene-environment interactions, random
genetic and environmental changes—mutations—act to entropically disrupt reliable
development.)
Innate Is Not the Opposite of Learned Everyone is a nativist, regardless of whether she
knows it. Even the most extreme advocates of the role of the environment in shaping
human behavior, from Skinner to the postmodernists, make nativist claims about the
“innate”structure of the evolved neural machinery that learns or responds to the
environment. The only difference is whether they make the nature of their claims
about this machinery explicit or allow them to remain implicit, forcing the reader to
deduce them from their arguments about why people act as they do.
Imagine that you are an engineer and your project is to create a brain that can learn.
To be able to learn, this brain would have to have a certain kind of structure—after all,
3-pound cauliflowers do not learn, but 3-pound brains do. To get your brain to learn,
you would have to arrange the neurons in particular ways. You would have to create
circuits that cause learning to occur. In short, you would have to equip your brain with
programs that cause it to learn. The same is true when natural selection is the engineer.
Even if a program that causes a particular kind of learning was itself learned, there
had to be a prior program that caused that learning to occur, and so on. Logic forces us
to conclude that there had to be, at some point in the developmental causal chain, a
program that caused learning but that was itself unlearned. These unlearned pro-
grams are a part of the brain by virtue of being part of its evolved architecture. They
are programs that reliably develop across the ancestrally normal range of human
environments.
Both environmentalists and nativists—Pavlov, Skinner, and Chomsky alike—must
agree on this point. They may disagree strongly about the computational structure of
the evolved programs that cause learning but not about whether evolved learning
programs exist. For example, classical and operant conditioning are widely viewed as
the simplest and most general forms of learning in humans and other animals. Yet,
even operant conditioning presumes the existence of evolved mechanisms that change
the probability of a behavior by a certain amount, as a function of its consequences
(and according to particular equations). It also presumes that a handful of conse-
quences—food, water, pain—are “intrinsically”reinforcing (i.e., the fact that these
consequences are capable of changing the probability of a subsequent behavior is a
design feature of the brain). Classical conditioning presumes the existence of a great
deal of innate equipment. In addition to the programs that compute contingencies, the
34 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 35
animal is filled with unconditioned—that is, unlearned—responses, such as salivating
in response to meat. Salivating in response to meat is considered to be part of the dog’s
evolved architecture, and what the evolved learning program does is calculate when
an arbitrary stimulus, such as a bell, predicts the appearance of the meat (Gallistel &
Gibbon, 2000). Thus, even in classical conditioning, the learned link between infor-
mation and behavior—salivating to the sound of the bell—is caused by an evolved
learning program, which takes as input both innate stimulus-response pairs (meat and
salivation) and information from the external environment (the contingency between
the sound of the bell and the appearance of meat). The only substantive disagreement
between a Skinner and a Chomsky is about the structures of the evolved programs that
cause learning.
Consequently, any learned behavior is the joint product of “innate”equipment
interacting with environmental inputs and, therefore, cannot be solely attributed to the
action of the environment on the organism. Thus, innate cannot be the opposite of
learned. It is just as mistaken to think of evolved as the opposite of learned because our
evolved learning programs were organized by evolution to learn some things and not
others.
To say a behavior is learned in no way undermines the claim that the behavior was
organized by evolution. Behavior—if it was learned at all—was learned through the
agency of evolved mechanisms. If natural selection had built a different set of learning
mechanisms into an organism, that organism would learn a different set of behaviors
in response to the same environment. It is these evolved mechanisms that organize the
relationship between the environmental input and behavioral output and thereby
pattern the behavior. For this reason, learning is not an alternative explanation to the claim
that natural selection shaped the behavior, although many researchers assume that it is.
The same is true for culture. Given that cultural ideas are absorbed via learning,
inference, and interaction payoffs—which themselves are caused by evolved pro-
grams of some kind in interaction with the environment—a behavior can be, at one
and the same time, cultural, learned, and evolved. (For excellent discussions of how
evolved inference mechanisms produce and structure cultural transmission, see
Boyer, 2001; Sperber, 1996.)
Moreover, there does not appear to be a single program that causes learning in all
domains (consider kin detection, food aversions, snake phobias, alliance detection,
and grammar acquisition). Evidence strongly supports the view that learning is
caused by a multiplicity of programs (Gallistel, 2000; Tooby & Cosmides, 1992).
Without specifying which program is the cause (and knowing its functional architec-
ture), nothing whatsoever is explained by invoking learning as an explanation for a
behavior. Labeling something as learning does not remove the requirement to spell
out the evolved machinery involved; it only makes the weak claim that interaction
with the environment participated in the process (which is always the case, anyway, in
all anatomical and behavioral phenotypes). In short, learning (like culture) is a
phenomenon that itself requires explanation, rather than being any kind of explan-
ation itself. A coherent explanation for how humans and nonhumans learn about a
given domain must include (a) a description of what the evolved learning program
looks like (that is, its circuit logic, code, or program architecture); (b) what selection
pressures and other evolutionary effects led it to acquire its present structure over
evolutionary time; (c) what set of gene-environment interactions lead it to develop its
structure at any given point in the organism’s life history, (d) what information was
and is available to the organism that is executing that evolved program.
The Theoretical Foundations of Evolutionary Psychology 35

3GC01 06/09/2015 12:40:43 Page 36
Everyone is also an environmentalist, regardless of whether she knows it. Even the
most die-hard nativist understands that organisms learn—or, even more broadly, that
an organism’s evolved mechanisms extract information from the environment and
process it to regulate behavior. Hence the environment regulates behavior, and it is the
presence of evolved mechanisms that makes this possible. Indeed, the entire function
of a brain is to allow the organism’s responses to be sensitively contingent to the
information provided by the environment.
Thus, evolved programs—instincts—are not the opposite of learning. They are the
engines or programs through which learning takes place. We learn only through
instincts—learning and reasoning instincts. There are instincts in songbirds for
learning songs, instincts in geese for learning which individual is one’s mother,
instincts in desert ants for learning how to return home, and instincts in humans
for learning a language or who our genetic relatives are. The greater the number of
specialized learning (or cognitive) programs we come equipped with, the more we can
learn from experience. Evolved programs do not constrain a “flexibility”that orga-
nisms otherwise would have; each additional program endows the organism with
competences it would not otherwise have. To take just one example, the evolved
language competence vastly multiplies the behavioral repertoire of humans. Humans
can respond with intricate contingency to the world because of these endowments.
This is why nature and nurture do not exist in a zero sum relationship, but in a positive
sum relationship. More nature (evolved systems of regulation and computation)
allows more nurture (exquisitely sensitive responsiveness to the world) (Boyer,
2001; Tooby & Cosmides, 1992).
Specialized or General Purpose? If the innate versus learned controversy is meaningless,
there is a genuine and illuminating question to be answered: What is the precise
structure of these evolved learning and regulatory programs? Are there many or just a
few? Which embody knowledge about enduring aspects of the world, and what
knowledge do their procedures reflect? To what extent is a program—regardless of
whether it governs learning—functionally specialized to produce the outcome that
you have observed?
What effect a given environmental factor will have on an organism depends
critically on the details of the designs of its evolved neurocomputational programs.
So the discovery of their structure is a pivotal question. Indeed, one of the few genuine
nature-nurture issues concerns the extent to which each evolved program is special-
ized for producing a given outcome (Cosmides & Tooby, 1987; Symons, 1987; Tooby &
Cosmides, 1992). Most nature-nurture issues disappear when more understanding is
gained about evolution, cognitive science, and developmental biology, but this one
does not.
Thus, the important question for any particular behavior is not, “Is it learned,”but,
“What kind of evolved programs produced it?”More specifically, “What is the
architecture of the evolved cognitive programs through which the organism learns
this particular type of behavior, acquires this kind of knowledge, or produces this form
of behavior?”
For any given (functional) outcome, there are three alternative possibilities: (1) It is
the product of domain-general programs, (2) it is the product of cognitive programs
that are specialized for producing that outcome (or a more inclusive set of which the
outcome in question is one instance), or (3) it is a byproduct of specialized cognitive
programs that evolved to solve a different problem.
36 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 37
The debate about language acquisition, which began in 1959 when Noam Chom-
sky reviewed B. F. Skinner’s book, Verbal Behavior, brings this issue into sharp focus,
because Chomsky and Skinner disagreed about precisely these issues (Chomsky,
1959; Skinner, 1957). Both sides in the ensuing controversy admit, as coherence
demands, that the human mind contains innate learning programs. But the two
camps differ in their answer to the question: Does a single set of general-purpose,
cognitive programs cause children to learn everything, with language as one
incidental example? Or is language learning caused, in part or in whole, by programs
that are specialized for performing this task: by what Chomsky called a language
acquisition device?
Questions about functional specialization cannot be answered a priori by theory or
logic alone. Each hypothesis about the computational architecture of a learning
mechanism—general, or specialized—must be evaluated on the basis of its coherence,
explanatory economy and power, retrodictive consistency with known phenomena,
and its ability to make successful, novel predictions. The theoretical tools and
empirical studies necessary will differ, depending on whether the proposal is about
language learning, inferring mental states, acquiring gender roles, developing friend-
ships, eliciting jealousy, or something else. For language, 55 years of research support
the hypothesis that humans have evolved programs specialized for various aspects of
language acquisition, although the debate remains heated (Pinker, 1994). With the
emergence of evolutionary psychology and under the weight of discoveries about
large numbers of diverse, specialized adaptive problems in many areas of biology, the
debate over adaptive specializations has now widened to include all human
competences.
Present at Birth? Sometimes people think that to show that a program is part of our
evolved architecture, researchers need to show that it is present from birth. Otherwise,
the behavior is “learned”(by which they implicitly mean learned through general-
purpose processes). But this assumes that all the evolved programs that cause
maturational development operate before birth and none after birth.
This assumption is clearly false. Teeth and milk-delivering breasts are uncontro-
versially standard parts of our evolved architecture, but they develop after birth, years
after in the case of breasts. Newborns lack teeth, but does this mean that infants and
toddlers acquire their first set through learning? Does cultural pressure lead them to
lose the first set in favor of the second?
Organs and design features can mature at any point of the life cycle, and this applies
to the adaptations in our brains just as much as it does to the features of our bodies.
Thus, the fact that a behavior emerges after birth tells us very little about how it was
acquired or why it has a certain organization. Organs can be disassembled on schedule
as well: Consider the placenta, umbilical cord, and fetal hemoglobin. Evolutionists
expect, and the evidence appears to bear them out, that many mechanisms will appear
and disappear on a condition-specific or life-history linked timetable based on when
they would have been needed, under ancestral conditions, to solve the challenges of
that life stage. Infants need the sucking reflex but not sexual desires; adolescents need
sexual desires but not the suckling reflex. For an example of a condition-specific
adaptation, consider pregnancy sickness. It does not manifest itself according to a
developmental schedule, but is triggered by a condition: Women during the first
trimester of pregnancy (that is, during fetal organogenesis) need a different set of
thresholds inhibiting the ingestion of substances that could cause birth defects than do
The Theoretical Foundations of Evolutionary Psychology 37

3GC01 06/09/2015 12:40:43 Page 38
nonpregnant women, so their disgust thresholds are adaptively adjusted by the
condition of pregnancy (Profet, 1992).
Presence at birth is only a function of what is needed at birth, not an indicator of
whether something is or is not part of our evolved architecture. Accordingly, much of
what is present in adult minds may have been put there by evolution and activated
through neural maturation, without depending on the accidents of personal experi-
ence. For example, infants who cannot crawl do not need a fear of heights, whereas
infants who can crawl do. But experiments have demonstrated that a fear of heights is
not learned by trial and error; rather, it is an evolved competence that is triggered
when the baby starts to self-locomote, even if researchers contrive the situation such
that the baby never experiences a fall (Campos, Bertenthal, & Kermoian, 1992).
Of course, the early presence of features is not completely irrelevant when
evaluating alternative hypotheses about our evolved design. For example, the early
emergence of a competence, before the social world could plausibly have acted, may
falsify or undermine a particular social constructionist hypothesis. But the early
absence of a competence does not by itself in any way undermine the claim that it
is part of our evolved design. We all start out as a single-celled zygote, so everything
develops.
The Twin Fallacies of Genetic Determinism and Environmental Determinism Traditional
researchers hold a series of beliefs that are widely accepted and that sound eminently
reasonable. Unfortunately, they are based on a series of fallacies about how develop-
ment works. The first belief is that some behaviors are genetically determined whereas
others are environmentally determined. The second is that evolutionary psychology
deals only with behavior that is genetically determined, not the much larger set of
behaviors that are environmentally determined. These beliefs are wrong for many
reasons (Tooby & Cosmides, 1990b, 1992; Tooby et al., 2003), of which we mention just
two (see also Hagen, Chapter 4, this volume).
First, genes are regulatory elements that use environments to construct organisms.
Thus, as discussed, every single component of an organism is co-determined by the
interaction of genes with environments. Moreover, some of those components are
computational mechanisms, designed to produce behavior on the basis of information
from the environment. Seen in this way, it is senseless to ask whether kin detection or
language acquisition or snake phobias are caused by the genes or the environment:
These phenomena are caused by evolved mechanisms that operate on information
from the environment in particular ways, and these evolved mechanisms were
themselves constructed by the interaction of genes with the environment.
Second, the view that evolutionary psychology deals only with “genetic”behaviors
(a nonexistent class) erroneously assumes that environmental causation is nonevolu-
tionary. In order to understand why environmental causation is every bit as “evolved”
as the genes, it is useful to distinguish “the environment”(in the sense of all properties
of the universe) from a given species’developmentally relevant environment. By
developmentally relevant environment we mean the subset of properties of the world that
evolution has made relevant to the development of the adaptations of a given species.
Evolution acts by selecting some genes over others, but in so doing it acts on the
relationship between the genes and the environment, choreographing their interaction
to cause evolved design. Genes are the so-called units of selection, which are inherited,
selected, or eliminated, so they are indeed something that evolves. But every time one
gene is selected over another, one design for a developmental program is selected as
38 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 39
well. (We all start as a single cell—brainless, limbless, gutless. Every cell and organ
system subsequently develops from that cell, nonrandomly climbing toward specific
organizational forms despite the onslaughts on entropy. For manifest organization to
emerge, there must be naturally selected processes that cause this to happen: devel-
opmental programs.)
Developmental programs, by virtue of their design, make some parts of the world
relevant to development and other parts irrelevant. Over evolutionary time, genetic
variation in developmental programs (with selective retention of advantageous
variants) explores the properties of the environment, discovering those that are useful
sources of information in the task of regulating development and behavior. Selection
tailors developmental programs to engage in organized interactions with facets of the
developmentally relevant environment to successfully produce highly ordered, func-
tional phenotypes. Selection also acts to render those features of the environment that
are unreliable or disruptive irrelevant to development. Step by step, as natural
selection constructs the species’gene set (chosen from the available mutations), it
selects in tandem which enduring properties of the world will be relevant to
development. Thus, a species’developmentally relevant environment—that set of features
of the world that a zygote and the subsequently developing organism depend on,
interact with, or use as inputs—is just as much the creation of the evolutionary process
as the genes are. Hence, natural selection can be said to store information necessary for
development both in the environment and the genes. Because for humans the amount
of information stored in the environment is much vaster than the quantity of genetic
information, one can think of the zygote, its genome, and its parentally supplied
cellular and uterine environment as analogous to a computer’s basic input/output
system (BIOS)—self-extracting kernels that bootstrap the single cell toward its highly
organized, realized set of adaptations (as expressed at a given point in life history). We
manifest species-typical (or population-tuned) evolved designs not because genes are
the only things that influence phenotypes, but because selection orchestrates the
interplay of gene-environment interactions through genes.
Hence, the developmentally relevant environment can be viewed as a second
system of inheritance comparable in some ways to genetic systems of inheritance. A
zygote in an environment can be seen as inheriting a set of genetic determinants
(including cellular machinery) and simultaneously a set of environmental determi-
nants. The environmental determinants are transmitted or inherited in a peculiar
fashion: They simply endure as physical arrangements in the world across generations
over the range in which the lineal series of zygotes appears. They must regularly recur
often enough that they select for developmental programs that interact with them to
cause reliable development every generation of the functional species-typical design.
From the point of view of any given subcomponent of the organism, other parts of the
organism are, of course, stable features of the environment, and so high levels of
functional interrelationship and developmental interdependence accumulate among a
body’s parts. In addition, some aspects of the environment outside the organism are
also enduring features of the ancestral world that interacted reliably with the orga-
nism’s design, and so subcomponents of the organism typically manifest highly
functional interrelationships with them (e.g., wings and air; eyes and light; digestive
enzymes and available diet), as well as developmentally interdependent relationships
with them (e.g., lung size and altitude during development). Some environmental
determinants are perfectly replicated across generations (e.g., the three-dimensional
nature of space, the properties of light, the properties of chemical compounds, the
The Theoretical Foundations of Evolutionary Psychology 39

3GC01 06/09/2015 12:40:43 Page 40
existence of two sexes, the presence of caretakers for infants that survive); others are
replicated reliably but imperfectly (e.g., mother smiling in response to an infant’s
smile, the presence of fathers during childhood, a correlation between duration of
childhood co-residence and genetic relatedness). In spite of omnipresent, order-
destroying entropy, organismic designs successfully develop the functional spe-
cies-typical design (and its locally tuned expressions) based on the degree to which
their genetic and environmental inheritances were functionally coordinated with each
other by selection adjusting them over evolutionary time so that they interactively
produced an adaptive phenotype. Gene–organism–environment webs have been
experimentally tested generation after generation; those interactions that led to
maladaptive development were discarded by selection. This evolutionarily orches-
trated coordination of genome and environment is how organisms are able to over-
come entropic processes that would otherwise preclude the existence of life (Tooby,
Cosmides & Barrett, 2003). Change in either of the two inheritances (either through
genetic mutation or change in the developmentally relevant environment) disrupts the
coordination, and the greater or more rapid the change, the greater is the disruption in
the always imperfect actual phenotype.
This view of development is not gene centered or a form of “genetic determinism”if
that is interpreted to mean that genes by themselves determine everything, immune
from environmental influence—or even that genes determine “more”than the
environment does. Although not gene centered, however, this view is very much
natural selection centered, because it is natural selection that chooses some genes
rather than others and, in so doing, orchestrates the interaction between the two
inheritances so that high degrees of recurrent functional order can emerge and persist,
such as eyes, kin-directed altruism, language, or maternal love.
Moreover, this view explains how reliable development both can and does
ordinarily occur—that is, it explains why a robust, species-typical design emerges
in almost all individuals (e.g., what can be seen in Gray’s Anatomy [Gray, 1918]).
The species-typical features of the genome interact with the features of evolutionarily
long-enduring, species-typical environments to produce the species-typical design
observable in organisms. Failures of reliable development are attributable to genetic
mutation, to environmental mutation (change), or to both.
The closest that the world comes to the fallacious distinction between biologically or
genetically determined traits versus environmentally or socially determined traits is in
the following real distinction: Some neural programs were designed by natural
selection to take in substantial amounts of environmental input (e.g., the language
acquisition device) whereas others were designed to take in less information and
express themselves less contingently (e.g., the typical form of the anger facial display
of emotion; Sell, Cosmides & Tooby, 2014). But in all cases, there is an underlying
regulatory or neural program designed by natural selection and a set of environmental
regularities necessary for that program’s reliable development. Indeed, as we discuss
later, there is not a zero-sum relationship between nature and nurture: More nature
(more evolved content specificity) allows more nurture (richer stores of ontogeneti-
cally elaborated data and locally contingent behavior). For example, the highly
organized language acquisition device allows marvelously rich and variable verbal
expression (Pinker, 1994).
From this perspective, successful development has to accomplish two tasks
(Tooby & Cosmides, 2001). The first is the reliable construction of the set of (largely
species-typical) adaptations required at each point in the organism’s life history (given
40 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 41
its sex). The second is to bring each adaptation into a state of readiness to perform its
evolved functions, given the organism’s situation. Accordingly, adaptations can be
conceptualized as operating in two different modes (Cosmides & Tooby, 2000a;
Tooby & Cosmides, 2001). The first is its functional mode, when it is performing
its evolved function (e.g., the incest avoidance system, calling up aversion at the
prospect of sex with a close relative). This is the aspect of the adaptation we normally
think about. The second is its organizational mode. This mode of operation is designed
to construct the adaptation, and in so doing, to furnish it with the information,
neuroendocrinological pathways, correct weightings in decision variables, proce-
dures, and representations it needs to behave adaptively when called upon to do
so. In general, the goal of the organizational mode of an adaptation is to cause it to
develop a better organization for carrying out its function so that, when it is called on
to operate, it discharges its function well (e.g., the kin detection front-end of the incest-
avoidance system, processing cues in its local environment into a kinship map of who
the individual’s genetic relatives are).
Although a natural first step for researchers is mapping adaptations operating in
their functional mode, it may be that solving the problem of correct assembly and
calibration of an adaptation is a much harder problem for the organism (given
entropy) than merely running the device, once it has been assembled. So, for example,
babbling, word learning, local syntax acquisition, the intrinsically entertaining nature
of verbal play, and so on, all seem to be the language system operating in its
organizational mode, so that when the individual needs to speak or understand,
the underlying adaptations are ready to perform their function. Rough and tumble
play are adaptations for fighting and defense operating in their organizational mode
(Symons, 1978). The organizational mode of an adaptation or set of adaptations will
generate different organizations (such as bodies of knowledge, habits, neuro-
endocrinological calibration, fear-sensitivity, etc.) in the minds of each individual,
given the individual’s unique experience or ontogenetic trajectory. This is the most
basic way that the evolved adaptations that compose our species-typical design lead to
large sets of functionally intelligible individual differences, without this outcome
being in any tension with an adaptationists perspective on human psychology and
behavior (Cosmides & Tooby, 2000a; Tooby & Cosmides, 2001).
Universal Architectural Design Versus Genetic Differences How are we to reconcile the
claim that there is a universal species-typical design—including a universal human
nature—with the existence of individual differences, especially those caused by
genetic differences among people?
At a certain level of abstraction, every species has a universal, species-typical
evolved architecture. For example, we humans all have a heart, two lungs, a stomach,
and so on. This is not to say there is no biochemical individuality, especially in
quantitative features. Stomachs, for example, vary in size, shape, and amount of
hydrochloric acid produced. Yet, all stomachs have the same basic functional design:
They are attached at one end to an esophagus and at the other to the small intestine,
they secrete the same chemicals necessary for digestion, they are made of the same cell
types, and so on. Indeed, when humans are described from the point of view of their
complex adaptations, differences tend to disappear, and a universal architecture
emerges. This universality is not only theoretically predicted, but is empirically
established (e.g., Gray’s Anatomy describes this architecture in minute detail). This
phenotypic universality is expected to be reflected at the genetic level through a
The Theoretical Foundations of Evolutionary Psychology 41

3GC01 06/09/2015 12:40:43 Page 42
largely universal and species-typical genetic architecture (“the”human genome) as
well.
The logic is as follows (see Tooby, 1982, and Tooby & Cosmides, 1990b, for a more
complete explanation, and a discussion of how to relate individual differences to
universal design):
1. Complex adaptations are intricate machines. Adaptations that consist of com-
plexly structured functional elements require, in turn, complex specification at the
genetic level. That is, they require coordinated gene expression, often involving
hundreds or even thousands of genes to regulate their development.
2. Like any other intricate machine, the parts of a complex adaptation must all be
present and fit together precisely if the adaptation is to work properly. Parts of
complex adaptations are functionally interdependent. All the genes necessary to
build each component part and assemble them correctly must be reliably brought
together in the same individual., Fitting together the parts specified by new genetic
combinations is not a problem for organisms that reproduce by cloning but it is for
sexual reproducers.
3. Each new human originates sexually. A randomly selected half of the mother’s
genes is recombined with a randomly selected half of the father’s genes. During
gamete and zygote formation, sexual reproduction automatically breaks apart
existing sets of genes and randomly generates in the offspring new combinations
at those loci that vary from individual to individual. This would not be a problem
if the mother and father were genetically identical at all loci. But it is a problem to
the extent that their genes differ at those loci underlying complex adaptations.
4. Hence, the successful assembly of a complex adaptation in a new individual
requires that all the genes necessary for that adaptation be supplied by the two
gametes, even though gametes are both randomly generated and consist of only
half of each parent’s DNA. Successful assembly would not be possible if only some
individuals in the population had the complex adaptation (and the suite of genes
that specified all its necessary component parts). If in a given generation, different
individuals had different complex adaptations, each of which was coded for by a
different suite of genes, then during the formation of the gametes for the next
generation the random sampling of subsets of the parental genes would break
apart each suite. During zygote formation, these incomplete specifications of
incompatible adaptations would be shuffled together. Consequently, the offspring
generation would be a handicapped jumble of fragments of functionally
incompatible adaptations. The simultaneous demand for functional compatibility
of complex adaptations and sexual reproduction places strong constraints on the
nature and distribution of functional variation.
5. Specifically, the only way that each generation can be supplied with the genetic
specification for complex adaptations is if the entire suite of genes necessary for
coding for each complex adaptation is effectively universal and hence reliably
supplied by each parent regardless of which genes are sampled. By analogy, if you
attempted to build a new car engine by randomly sampling parts from two parent
cars, you would fail if one parent were a Toyota and the other a Jaguar. To build a
new engine whose component parts fit together, you would have to salvage parts
from two parents that were of the same make and model.
6. By the same token, sexually reproducing populations of organisms freely tolerate
genetic variation to the extent that this variation does not impact the complex
42 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 43
adaptive organization shared across individuals. In the car engine example, the
color of the parts is functionally irrelevant to the operation of the car and thus can
vary arbitrarily and superficially among cars of the same make and model. But the
shapes of the parts are critical to functional performance and cannot vary if the
offspring design is to function successfully.
7. The constraint of functional universality applies to only adaptations whose genetic
basis is complex—that is, whose genetic basis involves multiple independently
segregating loci. This selection pressure starts when there are two independent
loci and becomes combinatorially more powerful with each additional locus.
However, if an adaptation can be coded for by a single gene in a way that is not
impacted by genes at other loci, then sexual recombination does not disassemble it,
and individuals may vary locally or regionally. Similarly, quantitative genetic
variation (e.g., height, arm length, how easily an individual is angered) is not
constrained by sexual reproduction and functional compatibility and thus may
also vary locally or regionally. Quantitative genetic variation is genetic variation
that shifts phenotypes quantitatively, but not outside the boundaries imposed by
the demand for functional compatibility.
8. Some evolved outcomes are the result of frequency-dependent selection. That is,
the population stabilizes at intermediate frequencies with two or more alternative
designs, such as male and female, because the relative reproductive advantage of
being one over the other decreases with increasing frequency (Fisher, 1930). If the
adaptation involves only a single locus, two or more alternative designs can persist
indefinitely in the species.
9. Finally, selection for genetic universality in complex adaptations does not rule out
the possibility that some individuals express complex adaptations that others do
not (as the two sexes and different life stages do, with, for example, the placenta,
fetal hemoglobin, teeth, the reproductively mature uterus, testes). Such expression,
however, must be based on a genetic architecture that is largely universal and
simply activated by an environmental trigger or a simple genetic switch such as a
single locus (e.g., the unrecombining regions of the Y chromosome). For example,
women express a different set of complex reproductive organs than men, but not
because men lack the genes necessary to code for ovaries and a uterus. If males and
females were different because each lacked the complex genetic specification of the
adaptations of the other sex, then, when they produced offspring, they would be
nonreproductive individuals of intermediate sex. In other words, functional aspects
of the architecture tend to be universal at the genetic level, even though their
expression may be typically limited to a particular sex or age or be contingent on
the presence of an eliciting condition (e.g., pregnancy adaptations) or at a single
nonrecombining stretch of DNA (e.g., biological sex in humans).
10. The living world sharply clusters into sets of organisms that share properties—
species—because of the demand for functional compatibility among sexual repro-
ducers. Indeed, it is striking the degree to which species are characterized by
complex, shared, and instantly recognizable designs (like different car models).
Still, the degree to which functional variation can be tolerated in a species is a
function of a number of variables, such as fecundity, migration rate, and popula-
tion density. In species in which successful parents have large numbers of
offspring, reproductive rates are high and migration rates are low between
populations, populations may diverge in some complex adaptations because local
mates are more likely to share functionally compatible genotypes even if there is
The Theoretical Foundations of Evolutionary Psychology 43

3GC01 06/09/2015 12:40:43 Page 44
variation elsewhere in the species. Compared with the great majority of other
species, however, ancestral humans had very low fecundity, had an open breeding
structure, and migrated across substantial distances. For these reasons, humans
are both expected to be, and are observed to be, characterized by a greater
tendency toward species typicality than many other species.
Thus, humans are free to vary genetically in their superficial, nonfunctional traits
but are constrained by natural selection to share a largely universal genetic design
for their complex, evolved functional architecture. Even relatively simple adaptive
programs must contain a large number of interdependent processing steps, limiting
the nature of the variation that can exist without violating the program’sfunctional
integrity. The psychic unity of humankind—that is, a universal and uniform human
nature—is necessarily imposed to the extent and along those dimensions that our
psychologies are collections of complex adaptations. In short, selection, interacting
with sexual recombination, tends to impose at the genetic level near uniformity in
the latent or potential functional design of our complex neurocomputational
machinery, and very high levels of expressed architecture uniformity at a given
sex and age.
Evolutionary Psychology and Behavior Genetics Ask Different Questions The preceding
discussion provides a framework for thinking about universal design and genetic
differences. Behavior geneticists, through twin studies and comparisons of kin raised
together and apart, explore the extent to which differences among individuals are
accounted for by differences in their genes. This difference is expressed as a heritability
statistic—h=Vg/Vg +Ve +Vge—which tells you the proportion of variance in a
population of individuals that is caused by differences in their genes (compared to
all causes: variance due to differences in environment, genes, and their interaction).
In contrast, evolutionary psychologists primarily explore the design of the universal,
evolved psychological and neural architecture that we all share by virtue of being
human.
Evolutionary psychologists are usually less interested in human characteristics that
vary due to arbitrary genetic differences because they recognize that these differences
are unlikely to be evolved adaptations central to human nature. Of the three kinds of
characteristics that are found in the design of organisms—adaptations, byproducts,
and noise—traits caused by genetic variants are predominantly (but not exclusively)
evolutionary noise, with little adaptive significance, while complex adaptations are
likely to be universal in the species.
Why is uniformity generally associated with functionality and variability typically
associated with lack of function? The first reason involves the constraints on organic
design imposed by sexual recombination, as explained earlier. Second, alternative
genes at the same locus (the same location in the human genome) are in a zero-sum
competition for relative frequency in the species: The more common one allele is, the
less common the others are. Natural selection tends to eliminate genetic differences
whenever two alternative alleles (genes) differ in their ability to promote reproduction
(except in the case of frequency-dependent selection). Usually, the better functioning
gene increases in frequency, squeezing out the less functional gene variant, until it
disappears from the species. When this happens, there is no longer genetic variability
at that locus: Natural selection has produced genetic uniformity instead. The more
important the function, the more natural selection tends to enforce genetic uniformity.
44 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 45
Thus, our important functional machinery tends to be universal at the genetic level,
and the heritability statistic associated with this machinery will be close to zero
(because there is little variation between individuals caused by genes). In contrast,
whenever a mutation fails to make a functional difference, selection will not act on it,
and such minor variants can build up at the locus until there is substantial genetic
variability for the trait, and its heritability statistic will be high (because most variation
between individuals in the trait will be caused by variation in genes). For this reason,
genetic variability tends to be predominantly nonadaptive or maladaptive evolu-
tionary noise: neutral or nearly neutral variants, negative mutations on their way to
being eliminated, and so on. Such variants may be, of course, of the greatest medical,
personal, or practical significance, as, for example, in the search for possible genetic
causes of schizophrenia, depression, and autism or the discovery that a formerly
neutral variant causes differential drug metabolism. The point is, however, genetic
variants causing medical vulnerabilities or personality differences are generally
unlikely to be adaptations designed to cause those effects. If something is highly
functional, selection usually acts to spread its genetic basis to the entire species.
Fundamentally, selection acts to decrease entropy in phenotypic design, while
mutation acts to increase it. Because of entropy, genomes are never close to flawless,
but are instead a balance between mutation and selection.
There is, nonetheless, a great deal of genetic variability within species, which is in
tension with the functional advantages of genetic uniformity. Aside from mutations
and neutral variants, there is a third reason for this genetic diversity. Genetic
variability, such as the ABO blood group system, is retained in the species because
genetically based, biochemical individuality interferes with the transmission of
infectious diseases from host to host (Tooby, 1982). Diseases that use or depend on
a protein found in their present host are thwarted when the next individual they jump
to has a different protein instead. Hence, natural selection sifts for genetic variants that
supply approximately the same functional properties to the adaptations they partici-
pate in but that taste different from the point of view of disease organisms. Because we
catch diseases from those we have contact with—such as our family, neighbors, and
other locals—selection favors maximizing genetically based protein diversity locally,
which requires pulling into every local population as many of the genetic variants
found anywhere in the species as possible. Thus, this explains why individuals are so
genetically different from one another, but different populations tend to be so
surprisingly genetically similar.
This large collection of genetic differences introduces minor perturbations into our
universal designs. The result is that each normal human expresses the universal
human design, but, simultaneously, each human is slightly different from every other
in personality, structure, temperament, health, anatomy, and appearance. These
differences tend to be quantitative in nature—a little more of this, a little less of
that—whereas the overall functional architecture remains the same.
Another category is the possibility of alternative, genetically based psychological
designs that are maintained through frequency-dependent selection. The existence of
male and female—two alternative designs—shows that such frequency-dependent
equilibria are not only possible but real for humans. Moreover, multiple behavioral
strategies often emerge in theoretical models through frequency-dependent selection
(e.g., cooperators and free-riders). Nevertheless, the constraints created by sexual
reproduction place strong limitations on the emergence of such systems in real species
(even the system of two sexes is based almost entirely on genetic uniformity). Indeed,
The Theoretical Foundations of Evolutionary Psychology 45

3GC01 06/09/2015 12:40:43 Page 46
as the case of the sexes shows, alternative phenotypic strategies can be based more
easily on substantial genetic uniformity and alternative developmental pathways than
on genetic differences encoding the alternative adaptations. It remains unclear the
extent to which humans exhibit allele-based frequency-dependent behavioral strate-
gies, and so far there are no well-established cases aside from the two sexes. For most
challenges, strategy selection might most advantageously take place when the
challenge is faced, so the strategy matches the challenge; this may be why genetic
commitments to strategies seem rare and would be generally disadvantageous.
However, the longer the period it will take to develop a good phenotype for a future
adaptive problem (as is arguably the case in mammals for developing a male or female
phenotypic design), the more it might pay to make an early commitment, undertaken
in greater ignorance of what future conditions will be like. Commitment by genetic
switch (e.g., XY sex determination) is the extreme case, where strategy commitment
occurs randomly at conception.The question of why there should be systems of
heritable dimensional personality variation will be addressed in the section on
epigenetics and parametric coordinative adaptations.
EVOLUTIONARY VERSUS TRADITIONAL APPROACHES TO PSYCHOLOGY:
HOW ARE THEY DIFFERENT?
If all psychologists are engineers working in reverse, if the goal of all psychologists is
to discover the design of the human mind, then how does evolutionary psychology
differ from traditional approaches?
Traditional approaches to psychology are not guided by any specific theory of what
the mind was designed to do. As animal species go, humans are startling in their
capabilities; from making lemon chiffon pies to writing waka to sending probes to
Titan, we are capable of solving many problems that no hunter-gatherers ever had to
solve (and that no other animal does solve). It, therefore, seemed obvious to many that
our minds are not designed to do anything in particular; rather, they must be designed
to reason and to learn, by virtue of mechanisms so general in function that they can be
applied to any domain of human activity. Reasoning and learning require certain
auxiliary processes: a memory to retain what is learned or inferred, perceptual systems
to bring sense data to the learning and reasoning mechanisms, and attention to
spotlight some aspects of perception for further analysis. But these auxiliary processes
were also thought to be domain-general. Noting the disconnection between assump-
tions in psychology and biology, Gallistel (2000, p. 1179) made the following obser-
vation about the study of learning:
Biological mechanisms are hierarchically nested adaptive specializations, each mecha-
nism constituting a particular solution to a particular problem . . . One cannot use a
hemoglobin molecule as the first stage in light transduction and one cannot use a
rhodopsin molecule as an oxygen carrier, any more than one can see with an ear or
hear with an eye. Adaptive specialization of mechanism is so ubiquitous and so obvious in
biology, at every level of analysis, and for every kind of function, that no one thinks it
necessary to call attention to it as a general principle about biological mechanisms. In this
light, it is odd but true that most past and contemporary theorizing about learning does
not assume that learning mechanisms are adaptively specialized for the solution of
particular kinds of problems. Most theorizing assumes that there is a general-purpose
46 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 47
learning process in the brain, a process adapted only to solving the problem of learning.
There is no attempt to formalize what the problem of learning is and thereby determine
whether it can in fact be conceived as a single or uniform problem. From a biological
perspective, this assumption is equivalent to assuming that there is a general-purpose
sensory organ, which solves the problem of sensing.
The same passage could have been written about reasoning, memory, or atten-
tion. The reigning assumption has been that the function of the mind is general—to
acquire information that is (roughly) true—which requires programs general
enough to handle content drawn from any and all domains. Thus, the study of
reasoning has concentrated on procedures that are content free. Examples include
logical procedures (which are designed to produce true conclusions from true
premises, no matter what the subject matter of the premises is); mathematical
procedures, such as Bayes’s theorem or multiple regression (which operate over
quantities of anything); and heuristics of judgment that use very general principles
such as similarity (the representativeness heuristic), frequency (the availability
heuristic), or what came first (anchoring and adjustment; e.g., Kahneman, Slovic, &
Tversky, 1982; Rips, 1994; but see Cosmides & Tooby, 1996a; Gigerenzer et al., 1999).
Memory has been conceived as a single system—after all, it had to be able to store
and retrieve information from all domains of human life. When multiple memory
systems are proposed, they are usually individuated by information modality or
source (a storage system for perceptual representations? motor skills? general
knowledge?) rather than by information content (Schacter & Tulving, 1994; but
see Caramazza & Shelton, 1998; Klein, 2005; Klein, Cosmides, Tooby, & Chance,
2002; Sherry & Schacter, 1987). Attention has primarily been seen as a content-free
mechanism that selects some information in an array for further processing. If true—
if attention contains no domain-specialized selection procedures—it should be safe
to study it using artificial stimuli that are easy to modify and manipulate in a
controlled fashion (Posner, 1978; Triesman, 2005). If true, principles derived from
experiments involving artificial stimuli should easily generalize to natural scenes
and stimuli—but they do not (Braun, 2003; Li, Van Rullen, Koch, & Perona, 2002;
New, Cosmides, & Tooby, 2007).
The traditional view of the mind is radically at variance with the view that emerges
from evolutionary psychology. Evolutionary psychologists expect a mind packed
with domain-specific, content-rich programs specialized for solving ancestral prob-
lems. For example, evolutionary psychologists would view attention not as a single
mechanism, but as an umbrella term for a whole suite of mechanisms, each designed to
select different information from a scene for different processing purposes. Some of
these may be relatively domain-general and deployed via volitional systems to any
task-relevant element in a scene—these are the attentional mechanisms that have been
studied most, using artificial stimuli. The mistake is not to think these exist, but to
think they are all that exist (Braun, 2003). For example, research with change detection
and attentional blink paradigms is uncovering attentional systems that are highly
domain-specific and deployed in the absence of any specific task demand. One system
preferentially attends to human faces (Ro, Russell, & Lavie, 2001). A similar system
snaps attention to the location at which a pair of eyes is gazing (Friesen & Kingstone,
2003) Yet another monitors animals for changes in their state and location: Changes to
animals are detected more quickly and reliably than changes to buildings, plants,
tools—even vehicles (New, Cosmides, & Tooby, 2007). Better change detection for
The Theoretical Foundations of Evolutionary Psychology 47

3GC01 06/09/2015 12:40:43 Page 48
animals than vehicles is significant because it shows a monitoring system tuned to
ancestral rather than modern priorities. Our ability to quickly detect changes in the
state and location of cars on the highway has life or death consequences and is a highly
trained ability in twenty-first century America, where the studies were done. Yet, we
are better at detecting changes in the states and locations of animals—an ability that
had foraging or sometimes predatory consequences for our hunter-gatherer ancestors
but is merely a distraction in modern cities and suburbs. By applying adaptationist
approaches, it is easy to predict and discover new principles of visual attention,
such as the evolved animacy bias, which would never have been discovered by a
metatheory that the brain consists primarily of general-purpose processes (New,
Cosmides, & Tooby, 2007).
The point is not just that attention will be composed of many different domain-
specific mechanisms, but that each domain-specialized attentional mechanism will be
part of a vertically integrated system linking the attended objects to domain-special-
ized inferential, learning, and memory systems. True, animals needed to be closely
monitored because they presented either dangers (e.g., predators) or opportunities for
hunting (prey), but once detected, other specialized processing is needed. Barrett has
shown that a predator-prey inference system develops early, regardless of relevant
experiences: 3- and 4-year-old children have a sophisticated understanding of preda-
tor-prey interactions, whether they grow up in urban Berlin or in a Shuar village in the
jaguar- and crocodile-infested Amazon, eating animals that their fathers hunted and
killed (Barrett, Chapter 9, this volume; Barrett, Tooby, & Cosmides, in press). Steen
and Owens (2001) have shown that chase play in toddlers and preschoolers has
features of special design as a system for practicing and perfecting escape from
predators (see also Marks, 1987).
Learning about animals is specialized as well. Mandler and McDonough (1998)
have shown that babies distinguish animals from vehicles by 7 months of age and
make different inferences about the two by 11 to 14 months. A detailed knowledge of
animal behavior is necessary for successful hunting (Blurton Jones & Konner, 1976;
Walker, Hill, Kaplan, & McMillan, 2002), and preschoolers as well as adults are
equipped with systems specialized for making inductive inferences about the prop-
erties of animals (Keil, 1994; Markman, 1989; Springer, 1992; and discussion thereof in
Boyer, 2001; Boyer & Barrett, Chapter 5, this volume; Barrett, Cosmides & Tooby, in
press). Atran and colleagues (Atran, 1998; López, Atran, Coley, Medin, & Smith, 1997)
provide cross-cultural evidence for a system specialized for sorting living kinds into
hierarchically organized, mutually exclusive taxonomic categories, which organize
inductive inferences: The closer two species are in this taxonomic structure, the more
likely someone is to assume that a trait of one is present in the other. Barrett Cosmides,
and Tooby (in press) have found a second parallel inductive system that uses
predatory role to guide inferences. This system assumes that two species are more
likely to share a trait if they are both predators than if one is a predator and the other an
herbivore. This system categorizes animals as predators or not on the basis of minimal
dietary information scattered amidst other facts about the species’natural history.
That is, the category predator is triggered by the information “eats animals”and guides
inductive learning; the effect on trait induction is strong—twice the size of the
taxonomic effect (Barrett, Chapter 9, this volume; Barrett, Cosmides, & Tooby, in
press). Animal-specialized memory systems appear to exist as well. For example,
Caramazza provides neuropsychological evidence that information about animals is
stored in a category-specific memory system, functionally and neurally separate from
48 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 49
that which stores information about artifacts (Caramazza, 2000; Caramazza & Shelton,
1998). From a traditional psychological perspective, content effects concerning ani-
mals are no more significant that hypothetical effects about door knobs, floorings, or
words that rhyme with Quetzalcoatl. From an evolutionary perspective, however,
animals were a selective agent of great magnitude and duration, and it would be a
surprise if our brains were not strongly shaped by their hundreds of millions of years
of interaction with other species.
We are emphasizing the content-specialized nature of processing about animals to
illustrate an important point. The benefit of an attentional system specialized for
monitoring animals is enhanced if its output is fed into inferential systems that infer
their mental states and use this information to predict their likely behavior. The
inferences and predictions generated by the mental state system are more useful if they
are reliably fed into decision rules that determine whether escape is necessary. The
monitoring system should also feed learning mechanisms that incidentally acquire
information about the animal’s properties; these, in turn, should feed memory systems
designed to encode, store, and retrieve information about the animals monitored,
according to ecologically relevant categories such as predator, taxonomically related, and
so on. Animal-specialized attentional, inferential, behavioral, learning, and memory
systems should be functionally integrated with one another, forming a distinct, category-
based system. The same should be true for other content domains. Distinct, content-
based information processing systems will exist to the extent that the computational
requirements for adaptive problem solving for one content area are functionally
incompatible with those for another (Sherry & Shacter, 1987; Tooby & Cosmides, 1992;
Tooby et al., 2005).
Seen from this perspective, the ordinary categories of psychology dissolve. To
have a textbook chapter on attention and a separate one on memory and then
learning and reasoning does not divide the mind in the most appropriate way.
Evolutionary psychologists suspect that there may be a domain-specialized system
for dealing with animals, with its own associated attentional, inferential, behavioral,
learning, and memory circuitry that are designed to work together as an integrated
system.
The organization of these specialized systems are expected to look nothing like
Fodor’s (1983, 2000) “pipelines”(for discussion, see Barrett, 2005, 2015; Boyer &
Barrett, Chapter 5, this volume). Some components of the system for making infer-
ences about animals will also be activated for plants and other living things as well
(e.g., taxonomic organization [Atran, 1990] or inferences that parts have functions
[Keil, 1994]). Other components of the animal system will be activated only in
response to animals—or, more precisely, to things manifesting those psychophysical
properties the system uses to detect animals, such as contingent reactivity or self-
propelled motion—whether the manifesting entity is a meerkat, a robot, or a cartoon.
Because many components of the animal system will be functionally specialized for
solving animal-specific adaptive problems, they will be composed of representations
and procedures that have little in common with those in a system for making
inferences about plants, artifacts, or cooperation between people (Boyer & Barrett,
Chapter 5, this volume). Nor will the boundaries between category-based systems be
clean. People may be attended by the animal monitoring system but also by the system
for monitoring social gestures; for inferences about growth and bodily functions,
people may be processed as animals but perhaps not for inferences about social
behavior. The organization of specializations will be complex and heterarchical, but
The Theoretical Foundations of Evolutionary Psychology 49

3GC01 06/09/2015 12:40:43 Page 50
with a functional logic that arose because of its excellence at solving ancestral
problems of survival and reproduction.
The old categories of psychological research have not led to robust models of the
human mind because they do not carve nature at the joints. Content specialization is
the rule, not the exception. The easiest way to make a domain-general model of
learning, reasoning, attention, or memory collapse is to introduce stimuli drawn from
different adaptive domains (e.g., Anderson & Phelps, 2001; Boyer & Barrett, Chapter 5,
this volume; Braun, 2003; Cosmides & Tooby, Chapter 25, this Handbook, Volume 2;
Gallistel, 2000). A more reasoned research strategy is to start developing some formal
(or even informal) analyses of specific adaptive problems and let these guide research.
If there are general systems or principles to be found, they will eventually emerge as
we gain a clear understanding of how each content-specialized system functions (for
an example, see Leslie, German, & Polizzi, 2005).
Biology is not split into evolutionary biology and nonevolutionary biology: All of
biology is organized by evolutionary principles. At some point, all psychology will be
evolutionary psychology, simply because it will make no sense to wall off the study of
humans from the rest of the natural world. When that happens, textbooks in
psychology will no longer be organized according to folk psychological categories,
such as attention, memory, reasoning, and learning. Their chapter headings will be
more like those found in textbooks in evolutionary biology and behavioral ecology,
which are organized according to adaptive problems animals must solve to survive
and reproduce: foraging (hunting, gathering), kinship, predator defense, resource
competition, cooperation, aggression, parental care, dominance and status, inbreeding
avoidance, courtship, mateship maintenance, trade-offs between mating effort and
parenting effort, mating system, sexual conflict, paternity uncertainty and sexual
jealousy, signaling and communication, navigation, habitat selection, and so on (e.g.,
see Buss, 1999). Future psychology textbooks will surely contain some additional
chapters that capture zoologically unusual aspects of human behavior, such as
language acquisition, coalition formation, deep engagement friendships, counter-
factual reasoning, metarepresentation, and autobiographical memory. But theories
of the computational mechanisms that make these unusual abilities possible will
include how they interact with and are supported by a wide variety of adaptive
specializations (e.g., Boyer, 2001; Cosmides & Tooby, 2000a; Klein, German, Cos-
mides, & Gabriel, 2004; Leslie et al., 2005; Sperber, 1994; Sperber & Wilson, 1995;
Tooby & Cosmides, 1996).
COMPUTATIONAL ADAPTATIONIST APPROACHES TO
MOTIVATION AND EMOTION
In principle, all modern behavioral scientists should understand that any mechanism
that processes information must have a computational description. This should
include psychological mechanisms that are responsible for motivation. For example,
mechanisms that cause fear, gratitude, sexual aversion to close relatives, romantic
love, guilt, anger, sexual jealousy, sexual attraction, the perception of beauty, or
disgust should all be describable in computational terms, which specify the relevant
inputs, representations, the procedures that act on them, and regulatory outputs. Yet,
until recently, most cognitive scientists, for example, would not even recognize these
topics as within their domain of study.
50 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 51
One reason for why even cognitive psychologists arbitrarily limit their scope is the
folk psychological distinction made between knowledge acquisition on the one hand
and motivation, emotion, feeling, and preferences on the other. Those who make this
distinction view cognition as the study of knowledge acquisition and leave motiva-
tion, emotion, and action to other research communities—a practice that presumes
that knowledge and motivation are separable rather than coevolved aspects of the
same unified systems of representation and action (see Fodor, 2000, for an example).
THE WEAKNESS OF CONTENT-FREE ARCHITECTURES
To some, it may seem as if an evolutionary perspective supports the case that our
species-typical psychological architecture consists primarily of powerful, general-
purposeproblemsolvers,inferenceenginesthatembodythecontent-freenormative
theories of mathematics and logic. After all, wouldn’t an organism be better
equipped and better adapted if it could solve a more general class of problems
over a narrower class? And won’t mathematical and logical inference engines
produce knowledge that is true, thereby providing a sound basis for choosing
themostadaptivecourseofaction?Thedifficulty with this intuition is that the more
general the problem-solving strategy is, the weaker and more nonfunctional it is.
What makes something a more general problem-solving strategy is that it can be
applied across a broader class of problems; to do this, it must be stripped of
strategies that yield correct answers on some subsets of problems and incorrect
answers on others. Domain-specific or content-sensitive architectures are not limited
in this way; if they can appropriately apply a program that evolved to solve a
specific subset of problems (e.g., kin detection), and others on other problem types
(optimal foraging; language acquisition), then it can solve a broader array of
problems than the one using content-independent general strategies. So our brains
should use the principle of preemptive specificity—use the program specialized for
thethecontent,ifthereisone,andifthereisnot,fallbacktostrategiesthatworkon
more inclusive problem types.
To be a plausible model of how the mind works, any hypothetical domain-general
neurocomputational architecture would have had to reliably generate solutions to all
of the problems that were necessary for survival and reproduction ancestrally. For
humans and most other species, this is a remarkably diverse, highly structured, and
very complex set of problems. If it can be shown that there are essential adaptive
problems that humans must have been able to solve to have propagated and that
domain-general mechanisms cannot solve them, the view of the mind as consisting
solely or primarily of domain-general programs fails. There appear to be a very large
number of such problems; at minimum, any kind of information-processing problem
that involves motivation and many others as well. This leads to the inference that the
human cognitive architecture contains many information-processing mechanisms that
are domain specific, content dependent, and specialized for solving particular adap-
tive problems (Cosmides, 1985; Cosmides & Tooby, 1987, 1994a, 1994b; Tooby, 1985;
Tooby & Cosmides, 1990a, 1992; Tooby et al., 2005).
Content-Free Is Content-Poor Some inferences are usefully applied to some domains
but not to others. For example, when predicting the behavior of people, it is useful to
assume they have beliefs and desires: invisible mental states that can be inferred but
The Theoretical Foundations of Evolutionary Psychology 51

3GC01 06/09/2015 12:40:43 Page 52
never observed. When predicting the behavior of rocks rolling down a hill, computing
their beliefs and desires is useless. Accordingly, the human psychological architecture
has evolved two separate inference systems for these two domains: a mind-reading
system for inferring the mental states of people (which can be selectively impaired in
autism; Baron-Cohen, 1995; Leslie & Thaiss, 1992) and an object mechanics system for
understanding the interactions of inanimate objects (Leslie, 1994; Spelke, 1990). Each
inference system is designed to be activated by cues particular to its domain of
applicability (e.g., human behavior for the mind-reading system, inanimate motion for
the object mechanics system). Because their domain of applicability is restricted,
specialized inferences appropriate for one domain can be made without producing
absurd inferences for another. This property allows domain-specific systems to
include rich, contentful inferential rules. For example, in content-free logics, “If P,
then Q”does not imply, “If Q, then P”because it would lead to absurd inferences (“If
you saw a horse, then you saw an animal”does not imply, “If you saw an animal, then
you saw a horse”). But a “logic”restricted to situations of social exchange, operating
over a more content-restricted set of representations (e.g., benefits, entitlement,
obligation, and so on), can usefully specify, “If you take the benefit, then you are
obligated to satisfy the requirement”implies, “If you satisfy the requirement, then you
are entitled to take the benefit”—an inference that is invalid for any content-free logic
(see Cosmides & Tooby, Chapter 25, this Handbook, Volume 2). Because they can have
content-restricted, specialized inference rules, domain-specific systems can arrive at
correct conclusions that more general rules are necessarily barred from making. As a
result, small inputs of information can generate many inductions or deductions.
Notice, however, that these powerful, content-rich inference systems are unavail-
able to a truly domain-general system. To maintain its domain generality, a general
system must be equipped only with rules that generate valid inferences across all
domains—people, rocks, plants, tools, nonhuman animals, and so on. It cannot take
advantage of any inference rules that are useful for one domain but misleading if
applied to another. It can have no mind-reading system, no object mechanics system,
no predator-prey inference system, or no specializations for tool use (e.g., Defeyter &
German, 2003; German & Barrett, 2005). The only kinds of inference rules that are left
are content-free ones, such as those found in logic and mathematics. Domain-general
systems are crippled by this constraint.
Combinatorial Explosion Combinatorial explosion paralyzes even moderately
domain-general systems when encountering real-world complexity. Imagine trying
to induce what caused your nausea in the absence of any privileged hypotheses. Your
entire life preceded the nausea, and a truly open-minded system would have to
consider every action, thought, sight, smell, taste, sound, and combination thereof as a
potential cause. In deciding how to respond, every possible action would have to be
considered singly and in combination. There would be nothing to privilege the
hypothesis that the cause was a recently consumed food and nothing to privilege
vomiting or future avoidance of that food as behavioral responses.
As the generality of a system is increased by adding new dimensions to a problem
space or new branch points to a decision tree, the computational load increases with
catastrophic rapidity. A content-free, specialization-free architecture contains no rules
of relevance, procedural knowledge, or privileged hypotheses and thus could not
solve any biological problem of routine complexity in the amount of time an organism
has to solve it (for further discussion, see, e.g., Carruthers, 2006; Gallistel, Brown,
52 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 53
Carey, Gelman, & Keil, 1991; Gigerenzer & Selten, 2002; Keil, 1989; Markman, 1989;
Tooby & Cosmides, 1992).
Acknowledging the necessity of a few “constraints”on learning will not solve this
problem. As Gallistel (2000, p. 1180) notes:
Early work focusing on the role of adaptive specialization in learning tended to formulate
the problem in terms of the constraints . . . or boundaries . . . that biological considera-
tions placed on the learning process ....[The contrasting argument] is that there is no
such thing as the learning process; rather there are many different learning processes.
While it is true that the structure of these processes constrain the outcome of learning in
interesting ways, the more important point is that it is the problem-specific structure of
these processes that makes learning possible.
Problem-specific learning specializations are necessary because the problem of
combinatorial explosion cannot be overcome by placing a few constraints on a single,
general learning process. Instead of asking, “How much specialization does a general-
purpose system require?”psychologists should be asking, “How many degrees of
freedom can a system tolerate—even a specialized, highly targeted one—and still
compute decisions in useful, real-world time.”Combinatorics guarantee that real
systems can tolerate only a small number. Without domain-specialized learning
mechanisms, we would learn nothing at all. Because the set of problems our ancestors
had to solve was not a random sample of the set of all logically possible information
relationships, the highly clustered relationships in real adaptive problems would have
selected, in many (perhaps all) cases, for networks of efficient specialization, along
with whatever strategies worked over broader sets of problems.
Clueless Environments Animals subsist on information. The single most limiting
resource to reproduction is not food or safety or access to mates, but what makes them
each possible: the information required for making adaptive behavioral choices. Many
important features of the world cannot be perceived directly, however. Content-free
architectures are limited to knowing what can be validly derived by general processes
from perceptual information, and this drastically limits the range of problems they can
solve. When the environment is clueless, the mechanism will be, too.
Domain-specific mechanisms are not limited in this way. When perceptual evi-
dence is lacking or difficult to obtain, they can fill in the blanks by using cues
(perceivable states or events) to infer the status of important, nonperceivable sets
of conditions, provided there was a predictable probabilistic relationship between the
cues and the unobservable states over evolutionary time. For example, it is difficult or
impossible to tell from experience that sex with siblings has a higher chance of
producing defective offspring—many conceptions are lost in utero, and whatever
problems exist in children born of such matings could have been caused by any
number of prior events. In contrast, a domain-specialized system can trigger disgust at
the prospect of sex with a sibling, drastically reducing the probability of inbreeding.
This will work, without individuals having to obtain any knowledge, conscious or
otherwise, about the pitfalls of inbreeding. Incestuous sex will simply seem disgusting
and wrong (Haidt, 2001; Lieberman et al., 2003, 2007). Similarly, ancestral hominins
had no method by which they could directly see another person’s genes to tell whether
they are genetic siblings or not. But a mind equipped with a domain-specific kin
detection system can estimate kinship on the basis of cues, such as maternal perinatal
The Theoretical Foundations of Evolutionary Psychology 53

3GC01 06/09/2015 12:40:43 Page 54
association, or co-residence during childhood, that were correlated with genetic
relatedness ancestrally. The person need not be aware of the cues used by this system,
the computational process employed, or even the concept of genetic relative.
What Counts as Adaptive Behavior Differs Markedly From Domain to Domain An
architecture equipped only with content-free mechanisms must succeed at survival
and reproduction by applying the same procedures to every adaptive problem. But
there is no domain-general criterion of success or failure that correlates with fitness
(for argument, see Cosmides & Tooby, 1987). For example, what counts as a “good”
mate has little in common with a “good”lunch or a “good”brother or a “good”person
to assault or a “good”place to set up camp. Designing a computational program to
choose foods based on their kindness or to choose friends based on their flavor and the
aggregate calories to be gained from consuming their flesh suggests the kind of
functional incompatibility issues that naturally sort human activities into
incommensurate motivational domains. Because what counts as the wrong thing
to do differs from one class of problems to the next, there must be as many domain-
specific subsystems as there are domains in which the definitions of successful
behavioral outcomes are incommensurate.
Amotivational domain is a set of represented inputs, contents, objects, outcomes, or
actions that a functionally specialized set of evaluative procedures was designed by
evolution to act over (e.g., representations of foods, contaminants, animate dangers,
people to emulate, potential retaliations to provocations). For a given species, there are
an irreducible number of these motivational domains; within each motivational
domain, there are an irreducible set of domain-specific criteria or value-assigning
procedures operating. For the domain of food in humans, for example, criteria and
value-assigning operations include salt, sweet, bitter, sour, savory, fat affordances,
putrefying smell avoidance, previous history with the aversion acquisition system,
temporal tracking of health consequences of specific foods by the immune system,
10
stage of pregnancy (because of the vulnerability of fetal organogenesis to chemical
disruption), boundaries on entities and properties considered by the system, perhaps
maggot-ridden food avoidance, and scores of other factors. When the required
assignments of value within a domain (e.g., food) cannot all be derived from a
common neurocomputational procedure, the number of motivational elements must
necessarily be multiplied to account for the data.
Thus, by evolved design, different content domains should activate different
evolved criteria of value, including different trade-offs between alternative criteria.
10
Humans and omnivorous nonhumans have a surprising ability to pick efficacious herbs to medicate
themselves with, to avoid foods with slow as well as fast acting toxins, to match nutritionally complemen-
tary foods, to identify effective nutrient releasing or detoxifying food processing practices, and to
differentially select foods with nutrients they are deficient in even with no obvious odor clues. To explain
these facts, we hypothesize that there is a set of adaptations that (a) exploits the immune system’s ability to
recognize alien proteins to construct recognition profiles of the digestive products of ingested substances; (b)
maps these recognition profiles to the sensory properties of foods ingested in temporal proximity to the
immune system’s exposure to the protein breakdown products; (c) identifies various components of health
(which specific detoxification pathways are overloaded, essential nutrient profile, immune categorization of
health, other short-term and long-term health consequences); (d) performs the matrix algebra of backward
inducing the temporal profiles of the health consequences of dietary substances onto the immune database of
recognized foods (plausibly using the Gallistelian time-series analysis component of conditioning); and (e)
maps the computational outputs of these analyses back to sensory food recognition templates, along with
valences that reweight how desirable or undesirable the food is to the organism.
54 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 55
Cases of motivational incommensurability are numerous and easily identified via
careful analyses of adaptive problems. Distinct and incommensurable evolved moti-
vational principles exist for food, sexual attraction, mate acquisition, parenting,
kinship, incest avoidance, coalitions, disease avoidance, friendship, predators, prov-
ocations, snakes, spiders, habitats, safety, competitors, being observed, behavior when
sick, motor skill acquisition, certain categories of moral transgression, and scores of
other entities, conditions, acts, and relationships.
There has been little progress over the past century toward constructing an
inventory of motivational domains. Without any proof or even an informal argument,
psychologists have presumed that most values are derived from the environment, by
computing contingencies between environmental conditions and a tiny set of
reinforcers (food, water, sex, pain; Herrnstein, 1977). As a field, we have been
shrugging off the issue of evolved motivations through the shell game of implying
that any given motivation is secondarily acquired, without obliging ourselves to
specify computationally how and from what. Yet, there are strong reasons to doubt
that a system of this kind would track fitness at all (Cosmides & Tooby, 1987; Tooby
et al., 2005).
Value and behavior cannot be induced from the environment alone. No environ-
mental stimulus intrinsically mandates any response or any value hierarchy of
responses. In the tangled bank of co-evolved organisms that Darwin memorably
contemplated at the end of the Origin of Species, naturally selected differences in the
brains of different species cause them to treat the same objects in a rich and conflicting
diversity of ways. The infant that is the object of caring attention by one organism is
the object of predatory ambition by another, an ectoparasitic home to a third, and a
barrier requiring effortful trajectory change to a fourth. It is the brains of these
organisms that introduce behavior-regulatory valuation into the causal stream and
natural selection that introduced into brains the neural subsystems that accomplish
valuation. The same stimulus set cannot, by itself, explain differences in the prefer-
ences and actions they provoke, nor indeed, the preferences themselves.
Value is not in the world even for members of the same species. Members of the
same species view the same objects differently. The very same object is one person’s
wife and another’s mother—an object of sexual preference in one case and sexual
aversion in the other. Moreover, because each evolved organism is by design the
center of its own unique valuer-centered web of valuations, evolved value, by its
nature, cannot have an objective character (Cosmides & Tooby, 1981; Hamilton, 1964).
Because of the structure of natural selection, social organisms are regularly in social
conflict, so that the objective states of the world that are preferred by some are aversive
or neutral to others (e.g., that this individual and not that should get the contested
food, mating opportunity, territory, parental effort, status, grooming, and so on). This
structure gives value for organisms an intrinsically indexical quality. Indeed, fitness
“interests”—the causal feedback conditions of gene frequency that value computation
evolved to track—cannot be properly assigned to such a high-level entity as a person
but are indexical to sets of genes inside the genome defined in terms of their tendency
to replicate under the same conditions (Cosmides & Tooby, 1981). Whatever else
might be attainable by sense data and content-free operations, value or its regulatory
equivalents must be added by our evolved architecture.
Values and Knowledge We can now address why knowledge acquisition cannot be
computationally divorced from motivation, valuation, and preferences.
The Theoretical Foundations of Evolutionary Psychology 55

3GC01 06/09/2015 12:40:43 Page 56
To behave adaptively, some actions, entities, or states of affairs must be valued
more than others, with a motivational system organized to pursue higher- over lower-
valued options. The computations whereby value is assigned typically involve many
of the same elements of conceptual structure that are the traditional objects of
cognitive science (representations of persons, foods, objects, animals, actions, events).
Thus, the evolution of motivational elements will mandate the evolution of an
irreducible set of conceptual elements as well. Why? A valuation is not meaningful
or causally efficacious for regulating behavior unless it includes some specification of
what is valued. That is, the specification of what the value applies to generally involves
conceptual structure.
For example, for natural selection to cause safe distances from snakes to be
preferred to closeness to snakes, it must build the recognition of snakelike entities
into our neurocomputational architecture. This system of recognition and tagging
operations is, for certain purposes, equivalent to having a snake concept, albeit a
skeletally specified one. Evidence supports the view that humans and related species
do indeed have a valuation system specialized to respond to snakes (e.g., Marks, 1987;
Mineka & Cook, 1993; Mineka, Davidson, Cook, & Keir, 1984; Yerkes & Yerkes, 1936).
This one consideration alone forces us to add a fourth “innate idea”to Kant’s trinity of
space, time, and causality. Yerkes’sfinding of evolved snake fear in chimpanzees
counts as empirically based philosophical progress and as straightforward progress in
the cognitive science of knowledge—derived (pace Fodor) from evolutionarily moti-
vated theories of function.
This argument not only establishes the necessity of evolved motivational elements
but also resurrects the argument for the necessity of “innate ideas,”that is, evolved
conceptual procedures within the cognitive architecture that embody knowledge
about the world and are triggered by evolved cue recognition systems that evolved
to be specifically responsive to stimuli with certain cues (however abstractly described
in the nervous system). It is the specificity of the coupling to the particular valuation
procedure (closer is negative) that individuates the concept with respect to the set of
motivational functions (e.g., beloved [your children], wary [snakes]).
Consider, for example, the series of interacting conceptual components necessary to
build a snake avoidance system. The system needs a psychophysical front-end: One of
its subcomponents assigns the evolved, internal tag snake through visual and bio-
mechanical motion cues to a perceptual representation of some entity in the world. It
has a second subcomponent that maps in a parameter, distance, between the snake and
the valued entity (e.g., self or child). The distance-representing component is used by
many systems. However, it also must have a component that assigns and updates
different specific valuation intensities for different distances, so that farther away is
better than closer for snakes (but not for food or other motivational domains). A
particular bad event (e.g., an imagined snake bite) need not be specifically represented
as a negative goal state in the snake avoidance system, with distance acquiring its
significance through backward induction and means-ends analysis. The distance-fear
relationship could fill the representation of space with a motivational manifold that
itself motivates avoidance (closeness is increasingly unpleasant). But such action-
inviting affordances are not the same, computationally, as a represented goal state.
The metric of valuation against distance (and its update rules) is proprietary to
snakes, but the output value parameter it produces must be accessible to other systems
(so that distance from snakes can be ranked against other goods, like getting closer to
extract your child from the python’s coils). Snake, distance, person, and the distance
56 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 57
(person, snake) valuation metric all necessarily operate together for this simple system
to work. Snakes, the entity to be protected, and distance cannot be assigned to one
computational process, with valuation assigned to another. Even in this simple
example, conceptual and valuation functions indivisibly interpenetrate each other,
with the representations necessarily coexisting within the same structure.
Learning, another clearly cognitive topic, is implicated in snake aversion as well,
but the learning process is domain-specific. It appears that the snake avoidance system
recalibrates based on individual experience, possibly slowly habituating in the
absence of negative experiences or observations and increasing sharply if snake
contact leads to injury. It also narrowly accepts inputs from the social world—a
conspecific expressing fear toward a snake (but not toward other stimuli such as
rabbits or flowers)—and uses this information to recalibrate the individual’s snake
valuation (Mineka & Cook, 1993; Mineka et al., 1984). Presumably, recalibration from
observing conspecifics evolved because the system operates more functionally by
upregulating or downregulating fear as a function of the local distribution of fear
intensities in others, which index to some degree the local rate at which venomous
snakes are encountered. (It is also worth pointing out that degrees of snake fear are,
therefore, “cultural”—weights in snake fear calibrate each other in interacting primate
communities.)
The key point is that even this apparently simple, one-function motivational system
involves a series of evolved content-specific conceptual elements, including snakes,
distance, conspecifics, that fear-faces have specific referents in the world, that snakes
are one of the privileged referents of a fear-face, and the output of fear itself. Not all
these elements are unique to the snake system (e.g., snake-recognition is; distance-to-
self, fear-faces, fear-output are not), but their pattern of distribution among motiva-
tional systems is heterarchical and itself not something that could be derived by
content-independent operations acting on unmediated experience.
As this form of analysis is applied to the other tasks humans perform, we think it
will be impossible to escape the general conclusion that cognitive science intrinsically
involves motivation and that the science of motivation intrinsically involves cognition.
The brain evolved as a control system (Weiner, 1948), designed to generate action.
From this perspective, there is not just a cognitive science of knowledge such as
language, intuitive physics, and number, but also a cognitive science of parenting,
eating, kinship, friendship, alliance, groups, mating, status, fighting, tools, minds,
foraging, threat, collective action, natural history, and scores of other ancient realms of
human action. Separating knowledge acquisition from motivation has placed the
study of motivation in cognitive eclipse and diverted cognitive scientists from
studying conceptual structure, motivation, and action as integrated systems (which
they will inevitably turn out to be). It ignores the many causal pathways whereby our
evolved architecture should have been designed to manufacture, store, communicate,
and act on the basis of representations that would not qualify as a rational architec-
ture’sefficient attempt at constructing true beliefs (Gigerenzer & Murray, 1987;
Haselton & Buss, 2000; Tooby & Cosmides, 1990a, in press). Evolved systems for
motivational computation use conceptual structure in targeted ways, so motivational
computation and knowledge computation cannot be isolated from each other into
separate systems, but instead evolves together. (For a more complete discussion, see
Tooby et al., 2005.) Indeed, many evolved concepts arguably exist so we can have
functional motivations about them (e.g., food, free rider, mother, child, predator,
snake, unclean, sexually attractive).
The Theoretical Foundations of Evolutionary Psychology 57

3GC01 06/09/2015 12:40:43 Page 58
EMOTIONS AS A SOLUTION TO THE SHORT-TERM
PROBLEM OF MECHANISM COORDINATION
The preceding discussion leads us to view the mind as a crowded network of evolved,
domain-specific programs. Each is functionally specialized for solving a different
adaptive problem that arose during hominin evolutionary history, such as face
recognition, foraging, mate choice, heart-rate regulation, sleep management, or predator
vigilance, and each is activated by a different set of cues from the environment. But the
existence of all these microprograms itself creates an adaptive problem: Programs that
are individually designed to solve specific adaptive problems could, if simultaneously
activated, deliver outputs that conflict with one another, interfering with or nullifying
one another’s functional products (e.g., digest food versus devote maximum blood
resources to the cardiopulmonary system and muscles executing escape). They may also
make conflicting demands on common computational resources. The existence of
attention itself, where some things are selected to be processed with higher priority
than others, demonstrates this. For example, sleep and flight from a predator require
mutually inconsistent actions, computations, and physiological states. It is difficult to
sleep when your heart and mind are racing with fear, and this is no accident: Disastrous
consequences would ensue if proprioceptive cues were activating sleepprograms at the
same time that the sight of a stalking lion was activating ones designed for predator
evasion. To avoid such consequences, the mind must be equipped with superordinate
programs that override some programs when others are activated (e.g., a program that
deactivates sleep programs when predator evasion subroutines are activated). Further-
more, many adaptive problems are best solved by the simultaneous activation of many
different components ofthe neurocomputational architecture, such that each component
assumes one of several alternative states (e.g., predator avoidance may require simul-
taneous shifts in both heart rate and auditory acuity).Again, a superordinate program is
needed that coordinates these components, snapping each into the right configuration at
the right time given the array of challenges prioritized by likely fitness consequences.
We have proposed that emotions are such programs (Tooby, 1985; Tooby & Cos-
mides, 1990a, 2008). To behave functionally according to evolutionary standards, the
mind’s many subprograms need to be orchestrated so that their joint product at any
given time is functionally coordinated to produce a best-bet set of responses, rather than
clashing in a cacophonous and self-defeating fashion. This coordination is accomplished
by a set of superordinate programs, namely the emotions. On this view, emotions are
adaptations that have arisen in response to the adaptive problem of mechanism
orchestration. This view implies that the exploration of (a) the statistical structure of
ancestral situations (the EEA) and (b) their relationship to the mind’s battery of
functionally specialized programs is central to mapping the emotions because the
most useful (or least harmful) deployment of programs at any given time will depend
critically on the exact nature of the immediate situation being confronted.
How did emotions arise and assume their distinctive structures? Fighting, falling in
love, responding to mistreatment by another, escaping predators, seeing a potential
sexual or mate-recruitment opportunity, confronting sexual infidelity, experiencing a
failure-driven loss in status, responding to the death of a family member, and so on
each involved conditions, contingencies, situations, or event types that recurred
innumerable times in hominin evolutionary history. Repeated encounters with
each kind of situation selected for adaptations that guided information processing,
behavior, and the body adaptively through the clusters of conditions, demands, and
58 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 59
contingencies that characterized that particular class of situation. These functions
could be accomplished by engineering superordinate programs, each of which jointly
mobilizes a subset of the psychological architecture’s other programs in a particular
configuration. Each configuration would be selected to deploy computational and
physiological mechanisms in a way that, when averaged over individuals and
generations, would have led to the most fitness-promoting subsequent lifetime
outcome given that class of ancestral situation type. So those designs that responded
to large felid predators by approaching to better appreciate their beauty were selected
out; those designs that motivated avoidance but did not accelerate heart rate and
breathing were also selected out compared to designs that increased the maximum
possible speed of retreat by accelerating heart rate and breathing (and suspend
digestion, long-run somatic repair, attention to competing non-time-fused goals,
and so on). Step by step, design variants that more thoroughly coordinate effective
response sets become incorporated into the species design.
When we use the term emotions, we are linking these evolved programs to evolutio-
narily recurrent situations (whether challenges or opportunities)that have a short-term
or moderately extended duration. These situations may terminate (e.g., with a rescuer
killing the predator), gradually lose their structure (with the predator wandering away
from your arboreal refuge, so that the predator risk returns to baseline levels), or be
replaced by other situations that trigger new emotions (your child makes a misstep and
is struggling not to fall out of the tree). Moreover, there is not only an abstract structure of
a recurrent situation (to which we have evolved an organized response), but there will be
recurrent dimensions of variation in the abstract structure of the recurrent situation,
which are used to calibrate the response. That is, not only will there be predator-threat,
but predator-threat varied in terms of speed, surprise, number, distance to safety,
number of allies, and so on (Cosmides & Tooby, 2000b). Hence individual ontogeneti-
cally encountered situations will be responded to in terms of the long-term abstract
structure of a situation, as parameterized by psychological variables that serve to
meaningfully individuate the immediate situation in a way in which the architecture
can recognize and to which it can deploy appropriately.
Moreover, the world does not dichotomously chop itself into short-term situations
and long-term conditions. For convenience, we term programs that coordinate
responses to short-term conditions emotions; we term coordinated responses to
conditions of intermediate duration that recalibrate a constellation of decision-vari-
ables calibrational adaptations or (if related to traditionally recognized emotions) moods;
and we term coordinated responses to enduring conditions parametric coordinative
adaptations. As discussed later, the major dimensions of personality variation (includ-
ing perhaps what researchers sometimes call temperaments) may be constructed by
various parametric coordinative adaptations.
The coordinated adjustment and entrainment of mechanisms (emotions) functions as
a mode of operation for the entire neurophysiological architecture and serves as the
basis for a precise computational and functional definition of each emotion state. Each
emotion entrains various other adaptive programs—deactivating some, activating
others, and adjusting the modifiable parameters of still others—so that the whole
system operates in a particularly harmonious and efficacious way when the individual
is confronting certain kinds of triggering conditions or situations. The conditions or
situations relevant to the emotions are those that (a) recurred ancestrally, (b) could not be
negotiated successfully unless there was a superordinate level of program coordination
(i.e., circumstances in which the independent operation of programs caused no conflicts
The Theoretical Foundations of Evolutionary Psychology 59

3GC01 06/09/2015 12:40:43 Page 60
would not have selected for an emotion program and would lead to emotionally neutral
states of mind), (c) had a rich and reliable repeated structure, (d) had recognizable cues
or situation-representations signaling their presence,
11
and (e) an error would have
resulted in larger fitness costs than the remedy. When a condition or situation of an
evolutionarily recognizable kind is detected, a signal is sent out from the emotion
program that (a) activates the specific constellation of subprograms appropriate to
solving the type of adaptive problems that were regularly embedded in that situation,
and (b) deactivates programs whose operation might interfere with solving those types
of adaptive problems. Programs directed to remain active may be cued to enter
subroutines that are specific to that emotion mode and were tailored by natural selection
to solve the problems inherent in the triggering situation with special efficiency.
According to this theoretical framework, an emotion is a superordinate program
whose function is to direct the activities and interactions of many subprograms,
including those governing perception, attention, inference, learning, memory, motor
planning, goal choice, motivational priorities, categorization and conceptual frame-
works, physiological reactions (e.g., heart rate, endocrine function, immune function,
gamete release), reflexes, behavioral decision rules, motor systems, communication
processes, energy level and effort allocation, affective coloration of events and stimuli,
and the recalibration of probability estimates, situation assessments, values, and
regulatory variables (e.g., self-esteem, estimations of relative formidability, relative
value of alternativegoal states, efficacy discount rate). An emotionis not reducible to any
one category of effects, such as effects onphysiology (“arousal”),behavioralinclinations,
situation interpretations (“appraisals”), facial expressions, or consciously accessible
feeling states, becauseit involves evolved instructions for all of them together, as well as
other mechanisms distributed throughout the humanmental and physical architecture.
For example, some emotion researchers consider that definitional to a basic emotion
is an identifiable emotional expression, causing them to focus on a set of six or seven
(happiness, sadness, anger, fear, surprise, disgust, and perhaps contempt). However,
an evolutionary computational approach makes it plausible that emotions are far
more numerous, but for only seven of these (identified so far) did it pay to broadcast
the individual’s emotional state to others (Tooby & Cosmides, 2008). We would add
confusion to this list (since it is recognizable on the face), and because we think it is a
mode of operation.
12
All psychological programs—including superordinate programs
11
If there is no repeated structure or no cues to signal the presence of a repeated structure, selection cannot
build an adaptation to address the situation.
12
An evolutionary recurrent situation can be extremely abstract, provided that there is a deployment of the
architecture that improves performance given the detection of this abstract situation. To give a flavor of just
how strangely abstract a “situation”can be, consider the hypothesis that confusion as a mental state might
not be a failure of processing, as it is usually thought of, but rather itself an adaptation. Indeed, it seems likely
that humans even have adaptations for confusion—that is, that confusion as a detected situation selected for
a mode of operation (confusion) that improves resolution of the problem posed by confusion (the situation).
What is the recurrent situation that confusion (the mode of operation) is a response to? Confusion may be
defined as having insufficient information to decide on a single coherent representation of the organism’s
circumstances relevant to selecting a best response; this can involve feedback to behavior being highly
inconsistent with expectation; being exposed to conflicting cues that imply mutually inconsistent conditions,
or a situation requiring contradictory responses. Evolved best responses to the situation of confusion may be
such computational adjustments as a suspension of ongoing action; a broadening of attentional focus
beyond ongoing goal-pursuit; increasing the search for disambiguating cues; rapid shifts between different
interpretations of data to see which has the best fit; increasing uncertainty weightings on decision-relevant
variables; and dropping down the ladder of interpetations and responses to more conservative computa-
tional or behavioral strategies that yield positive returns over broader sets of conditions.
60 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 61
of this kind—are sometimes mistaken for “homunculi,”that is, entities endowed with
“free will.”A homunculus scans the environment and freely chooses successful
actions in a way that is not systematic enough to be implemented by a program. It
is the task of cognitive psychologists to replace theories that implicitly posit such a
computationally impossible entity with theories that can be implemented as informa-
tion processing architectures with open parameters. Emotion programs, for example,
have a front end that was designed to detect evolutionarily reliable cues that a
situation exists (regardless of whether these cues still reliably signal the presence
of that situation in the modern world); when triggered, they entrain a specific set of
subprograms: those that natural selection chose as most useful for solving the
problems that situation posed in ancestral environments. Just as a computer can
have a hierarchy of programs, some of which control the activation of others, the
human mind can as well. Far from being internal free agents, these programs have
execute their evolved code regardless of the needs and circumstances of the modern
individual; they were designed to create states (fury) and implement actions that
worked effectively in ancestral situations (e.g., murder a weaker rival), regardless of
their consequences in the present (e.g., prison).
FEAR (AN EXAMPLE)
The ancestrally recurrent situation is being alone at night and a situation-detector
circuit perceives cues that indicate the possible presence of a human or animal
predator. The emotion mode is a fear of being stalked. (In this conceptualization of
emotion, there might be several distinct emotion modes that are lumped together
under the folk category fear but that are computationally and empirically distinguish-
able by the different constellation of programs each entrains.) When the situation
detector signals that the individual has entered the situation “possible stalking and
ambush,”the following kinds of mental programs are entrained or modified:
•There are shifts in perception and attention. You may suddenly hear with far
greater clarity sounds that bear on the hypothesis that you are being stalked but
that ordinarily you would not perceive or attend to, such as creaks or rustling.
Are the creaks footsteps? Is the rustling caused by something moving stealthily
through the bushes? Signal detection thresholds shift: Less evidence is required
before you respond as if there were a threat, and more true positives will be
perceived at the cost of a higher rate of false alarms.
•Goals and motivational weightings change. Safety becomes a far higher priority.
Other goals and the computational systems that subserve them are deactivated.
You are no longer hungry; you cease to think about how to charm a potential
mate; or practicing a new skill no longer seems rewarding. Your planning focus
narrows to the present; worries about yesterday and tomorrow temporarily
vanish. Hunger, thirst, and pain are suppressed.
•Information-gathering programs are redirected. Where is my baby? Where are
others who can protect me? Is there somewhere I can go where I can see and hear
what is going on better?
•Conceptual frames shift, with the automatic imposition of categories such as
dangerous or safe. Walking a familiar and usually comfortable route may now
be mentally tagged as dangerous. Odd places that you normally would not
The Theoretical Foundations of Evolutionary Psychology 61

3GC01 06/09/2015 12:40:43 Page 62
occupy—a hallway closet, the branches of a tree—suddenly may become salient
as instances of the category safe or hiding place.
•Memory processes are directed to new retrieval tasks. Where was that tree I
climbed before? Did my adversary and his friend look at me furtively the last
time I saw them?
•Communication processes change. Depending on the circumstances, decision
rules might cause you to emit an alarm cry or be paralyzed and unable to speak.
Your face may automatically assume a species-typical fear expression.
•Specialized inference systems are activated. Information about a lion’s trajectory
or eye direction might be fed into systems for inferring whether the lion saw you.
If the inference is yes, a program automatically infers that the lion knows where
you are; if no, the lion does not know where you are (the seeing-is-knowing
circuit identified by Baron-Cohen, 1995, and inactive in people with autism). This
variable may automatically govern whether you freeze in terror or bolt (Barrett,
Chapter 9, this volume). Are there cues in the lion’s behavior that indicate
whether it has eaten recently and thus is unlikely to be predatory in the near
future? (Savanna ungulates, such as zebras and wildebeests, commonly make
this kind of judgment; Marks, 1987.)
•Specialized learning systems are activated, as the large literature on fear con-
ditioning indicates (e.g., LeDoux, 1995; Mineka & Cook, 1993; Öhman & Mineka,
2001; Pitman & Orr, 1995). If the threat is real and the ambush occurs, the victim
may experience an amygdala-mediated recalibration (as in posttraumatic stress
disorder) that can last for the remainder of his or her life (Pitman & Orr, 1995).
•Physiology changes and the immune system adjusts. Gastric mucosa turn white
as blood leaves the digestive tract (another concomitant of motivational priorit-
ies changing from feeding to safety); adrenalin spikes; heart rate may go up or
down (depending on whether the situation calls for flight or immobility), blood
rushes to the periphery, and so on (Cannon, 1929; Tomaka, Blascovich, Kibler, &
Ernst, 1997); instructions to the musculature (face and elsewhere) are sent
(Ekman, 1982). Indeed, the nature of the physiological response can depend
in detailed ways on the nature of the threat and the best response option (Marks,
1987).
•Behavioral decision rules are activated. Depending on the nature of the potential
threat, different courses of action will be potentiated: hiding, flight, self-defense,
or even tonic immobility (the latter is a common response to actual attacks, both
in other animals and in humans).
13
Some of these responses may be experienced
as automatic or involuntary.
13
Marks (1987) vividly conveys how many aspects of behavior and physiology may be entrained by certain
kinds of fear: “During extreme fear humans may be ‘scared stiff’or ‘frozen with fear’. A paralyzed conscious
state with abrupt onset and termination is reported by survivors of attacks by wild animals, by shell-shocked
soldiers, and by more than 50% of rape victims (Suarez & Gallup, 1979). Similarities between tonic
immobility and rape-induced paralysis were listed by Suarez & Gallup (features noted by rape victims
are in parentheses): (1) profound motor inhibition (inability to move); (2) Parkinsonian-like tremors (body-
shaking); (3) silence (inability to call out or scream); (4) no loss of consciousness testified by retention of
conditioned reactions acquired during the immobility (recall of details of the attack); (5) apparent analgesia
(numbness and insensitivity to pain); (6) reduced core temperature (sensation of feeling cold); (7) abrupt
onset and termination (sudden onset and remission of paralysis); (8) aggressive reactions at termination
(attack of the rapist after recovery); (9) frequent inhibition of attack by a predator . . .”(pp. 68–69).
62 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 63
From the point of view of avoiding danger, these computational changes are
crucial: They are what allowed the adaptive problem to be solved with high proba-
bility, on average, over evolutionary time. In any single case they may fail because
they are only the evolutionarily computed best bet, based on ancestrally summed
outcomes; they are not a sure bet, based on an unattainable perfect knowledge of the
present.
Whether individuals report consciously experiencing fear is a separate question
from whether their mechanisms assumed the characteristic configuration that, accord-
ing to this theoretical approach, defines the fear emotion state. Individuals often
behave as if they are in the grip of an emotion, while denying they are feeling that
emotion. It is perfectly possible that individuals sometimes remain unaware of their
emotion states, which is one reason subjective experience should not be considered the
sine qua non of emotion. At present, both the function of conscious awareness and the
principles that regulate conscious access to emotion states and other mental programs
are complex and unresolved questions. Mapping the design features of emotion
programs can proceed independently of their resolution, at least for the present.
This computational approach also allows testing for the presence of emotion programs
cross-culturally. The design features of an emotion mode should be present and
ascertainable experimentally, whether the language has a word for an emotion state or
not (pace Lutz, 1988).
THE FUNCTIONAL STRUCTURE OF AN EMOTION PROGRAM EVOLVED TO MATCH
THE EVOLUTIONARILY SUMMED STRUCTURE OF ITS TARGET SITUATION
According to this framework, the sets of human emotion programs assumed their
evolved designs through interacting with the statistically defined structure of human
environments of evolutionary adaptedness. Each emotion program was constructed
by a selective regime imposed by a particular evolutionarily recurrent situation—a
cluster of repeated probabilistic relationships among events, conditions, actions, and
choice payoffs. These would have had to have (a) endured over a sufficiently long
stretch of evolutionary time (and proportion of the species range) to have had selective
consequences on the design of the mind; and (b) be probabilistically associated with
cues detectable by humans. To the extent that situations exhibit such a structure, their
statistical properties are expected to have been used by selection to build an emotion
program whose detailed design features are favored given that recurrent situation.
That is, the architecture of the emotion program should manifest an advantageous
complementarity with the structure of the recurrent situation, so that their interaction
produces a better outcome (given ancestral conditions) than would have been
produced without the program.
Emotion programs have evolved to take features of the recurrent statistical and
causal structure into account, whether they could have been perceived ontogenetically
or not. This tailoring is accomplished by selection, acting over evolutionary time,
differentially incorporating program components that dovetail with individual items
on the list of properties probabilistically associated with the situation. Thus,
embedded in an emotion mode is a way of interpreting the world in terms of
parameters made meaningful by the recurrent structure, assuming causal connections
(even unobservable ones) that were typically present, and being motivated to take
action related to the ancestral cluster of probabilistically associated elements. So, for
The Theoretical Foundations of Evolutionary Psychology 63

3GC01 06/09/2015 12:40:43 Page 64
example, if ancestrally a new group moving into one’s locale statistically foreshad-
owed eventual zero-sum conflict, competition, and potential expulsion from their
existing resource base by the new group at some nontrivial rate, then humans should
be designed to be more liable to experience intergroup fear, hostility, and rivalry.
Similarly, if anger is an emotion program that evolved to orchestrate negotiative
behaviors in conflicts of interest, and being perceived as stronger increases one’s
bargaining power (Sell, Tooby, & Cosmides, 2009), then evolution should have
incorporated elements into the facial display of anger that enhanced the appearance
of strength (as it appears to have done: Sell et al., 2014).
For example, the condition of having a mate plus the condition of your mate
copulating with someone else constitutes a situation of sexual infidelity—a situation
that has recurred over evolutionary time, even though it has not happened to every
individual. Associated with this situation were cues reliable enough to allow the
evolution of a “situation detector”(e.g., observing a sexual act, flirtation, or even
the repeated simultaneous absence of the suspected lovers are cues that could trigger
the categorization of a situation as one of infidelity). Even more importantly, there
were many necessarily or probabilistically associated elements that tended to be
present in the situation of infidelity as encountered among our hunter-gatherer
ancestors. Additional elements include: (a) a sexual rival with a capacity for social
action and violence, as well as allies of the rival; (b) a discrete probability that an
individual’s mate has conceived with the sexual rival; (c) changes in the net lifetime
reproductive returns of investing further in the mating relationship; (d) a probable
decrease in the degree to which the unfaithful mate’s mechanisms value the victim of
infidelity (the presence of an alternative mate lowers replacement costs); (e) a cue
that the victim of the infidelity will likely have been deceived about a range of past
events, leading the victim to confront the likelihood that his or her memory is
permeated with false information; and (7) the victim’s status and reputation for being
effective at defending his or her interests in general would be likely to plummet,
inviting challenges in other arenas. These are just a few of the many factors that
constitute a list of elements associated in a probabilistic cluster; they constitute the
evolutionary recurrent structure of a situation of sexual infidelity. The emotion of
sexual jealousy evolved in response to these properties of the world—this situation—
and there should be evidence of this in its computational design (Buss, 2000; Daly,
Wilson, & Weghorst, 1982).
For example, if in ancestral situations of sexual infidelity, there was a substantially
higher probability of a violent encounter than in its absence, the sexual jealousy
program will have been shaped by the distillation of those encounters, and the
jealousy subroutines will have been adjusted to prepare for violence (e.g., with
heart rate increase) in proportion to the raised probability in the ancestral world.
(Natural selection acts too slowly to have significantly updated the mind to post-
hunter-gatherer conditions.) Each of these subelements and the adaptive circuits they
require can be added to form a general theory of sexual jealousy (e.g., Buss, 2000).
The emotion of sexual jealousy constitutes an organized mode of operation
specifically designed to deploy the programs governing each psychological mecha-
nism so that each is poised to deal with the exposed infidelity. Physiological processes
are prepared for things such as violence, sperm competition, and the withdrawal of
investment; the goal of deterring, injuring, or murdering the rival emerges; the goal of
punishing, deterring, or deserting the mate appears; the desire to make yourself more
competitively attractive to alternative mates emerges; memory is activated to
64 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 65
reanalyze the past; confident assessments of the past are transformed into doubts; the
general estimate of the reliability and trustworthiness of the opposite sex (or indeed
everyone) may decline; associated shame programs may be triggered to search for
situations in which the individual can publicly demonstrate acts of violence or
punishment that work to counteract an imagined or real social perception of weak-
ness; and so on.
It is the relationship between the summed details of the ancestral condition and the
detailed structure of the resulting emotion program that makes this approach so useful
for emotion researchers. Each functionally distinct emotion state—fear of predators,
gratitude, guilt, sexual jealousy, anger, grief, and so on—corresponds to an integrated
mode of operation that functions as a solution designed to take advantage of the
particular structure of the recurrent situation or triggering condition to which that
emotion corresponds. This approach can be used to create theories of each individual
emotion, through three steps: (a) reconstructing the clusters of properties of ancestral
situations, (b) constructing engineering analyses about how each of the known or
suspected psychological mechanisms in the human mental architecture should be
designed to deal with each ancestral condition or cluster of conditions and integrating
these into a model of the emotion program, and (c) constructing or conducting
experiments and other investigations to test and revise the models of emotion
programs.
Evolutionarily recurrent situations can be arrayed along a spectrum in terms of how
rich or skeletal is the set of probabilistically associated elements that defines the
situation. A richly structured situation, such as sexual infidelity or predator ambush,
will support a richly substructured emotion program in response to the many
ancestrally correlated features. Many detailed adjustments will be made to many
psychological mechanisms as instructions for the mode of operation. In contrast, some
recurrent situations have less structure (i.e., they share fewer properties in common),
so the emotion mode makes fewer highly specialized adjustments, imposes fewer
specialized and compelling interpretations and behavioral inclinations, and so on. For
example, surges of happiness or joy are an emotion program that evolved to respond
to the recurrent situation of encountering unexpected positive events. The class of
events captured by “unexpectedly positive”is extremely broad and general and has
only a few additional properties in common, selecting for differential responses (e.g.,
adjusting the reserve price for taking action down or up, so that joy makes people
more energetic, whereas sadness deters action). Emotion programs at the most general
and skeletal end of this spectrum correspond to what some call mood (happiness,
sadness, excitement, anxiety, playfulness, homesickness, and so on).
MOTIVATIONAL SYSTEMS,INTERNAL REGULATORY VARIABLES,
AND RECALIBRATIONAL EMOTIONS
Although traditional theories of motivation have tended to be general-purpose or very
simple (e.g., motivation as goal seeking; motivation driven by a general-purpose
operant conditioning system shaped by histories of reinforcement, linked to a small
number of drives or reinforcers, such as food, water, sex, etc.). But evolutionary
research has identified a large and expanding number of adaptive problems for which
there exist no corresponding motivational theories in traditional psychology (e.g., kin-
directed altruism, incest avoidance, exchange partner management, power-based
The Theoretical Foundations of Evolutionary Psychology 65

3GC01 06/09/2015 12:40:43 Page 66
negotiation, habitat selection, prevention of partner infidelity, contagion avoidance,
child care, status-seeking, affiliation by association value, punitive deterrence of free-
riding, advancement of one’s coalition’s interests with respect to competing coalitions,
and so on). These do not involve freely chosen goals, nor do they appear to be well-
captured by any extant drive-reduction theory. These adaptive problems are each so
different from each other that they require distinct adaptive specializations to solve
them (how much sexual aversion should you feel toward this half-sibling—incest
avoidance; how much cost should you incur to struggle for a resource against this
adversary—anger; how much should you recalibrate your disposition to help some-
one who helped you more than you expected to consolidate a higher level of mutual
cooperation—gratitude; how determined should you be to punish a free rider—
punitiveness; how much effort should you devote to your group, given the likely
costs—loyalty).
Motivational adaptive problems are, abstractly, information processing problems
involving evaluating expected fitness payoffs to alternative courses of action, given
information available to the organism about its situation, in order to make decisions
that are best bet responses. Ancestrally recurrent situations that required choices (e.g.,
have sex with this person given cues that he might be your brother?; punish or ignore
free-riding?) can be organized into distinct sets or clusters with statistically recurrent
features, cues, invisible concomitants, outcomes, and payoff distributions. This in turn
led to selection for distinct motivational subsystems tailored to the special properties
of each motivational problem-type (incest-avoidance; child care; sacrifice for the
coalition; mateship maintenance; exploitation of opportunities for gain through
aggression; satisfaction of curiousity). To operate, each of these will generally be
associated with proprietary interpretive systems with reliably developing conceptual
primitives such as free-rider (Delton, Cosmides, Guerno, Robertson, & Tooby, 2012), so
that motivations such as punitive sentiment can be directed toward their functional
targets (e.g., the transgressing person; see also Price, Cosmides, & Tooby, 2002).
Conveniently, evolutionary biologists have developed a number of models of adap-
tive problems—that is, how selection acts in specific domains (such as kin selection,
inbreeding depression, sexual selection, the asymmetric war of attrition); these models
can be used to develop models of the computational architectures that specific
motivational subsystems should have in order to be able to solve their respective
adaptive problems.
In order to construct a theoretical framework capable of incorporating this new
range of cases, we need to introduce a new class of computational elements that have
no present counterpart in the cognitive sciences, traditional approaches to motivation,
or folk psychology. That is, they are not thoughts, or feelings, or desires as ordinarily
conceptualized. For sake of simplicity, we call these computational elements internal
regulatory variables (Tooby & Cosmides, 2008; Tooby, Cosmides, Sell, Lieberman, &
Sznycer, 2008). They are needed to register properties of persons, acts, and situations
that are needed to compute, implicitly or explicitly, the value and probability of an
outcome of a particular kind, given a course of action; to segregate elements in the
world into classes that can then be assigned motivationally relevant meanings (e.g.,
my child, a sexual opportunity, or a potential friend); or to store decision-making
thresholds that partition the set of possible actions in the immediate situation that are
fitness-promoting from those that are fitness-reducing (e.g., a welfare trade-off
threshold). They not only encode necessary precursors (e.g. co-residence as one input
into relatedness computation) necessary to specialized next-step input computations
66 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:43 Page 67
(e.g., update kinship index estimating how genetically related this person is); but they
also can encode values themselves (a high kinship index can then lead to a computa-
tion that this specific relative is valuable to this degree); this in turn can be used to
provide values (as parameters) to decision-making circuits (e.g., place a high weight
on this person’s welfare when making choices that affect the welfare of self versus
other—welfare trade-off magnitudes).
According to this view, internal regulatory variables evolved to track those narrow,
targeted properties of the body, the social environment, and the physical environment
whose computation provided inputs needed by evolved decision-making programs in
order to generate motivations relevant to choice and action. At their simplest, internal
regulatory variables have discrete parameter values (e.g., target person represented as
being male or female) or continuous magnitudes (target person is represented as
having a kinship index ranging between 0 and some evolutionarily set possible upper
bound). Final outputs of different motivational systems about the value of various
outcomes need also to be expressed in a common neural currency, so that trade-offs
and opportunity costs are incorporated into choice behavior for mutually exclusive
choices. That is, ultimately, you choose to deliver the gazelle haunch to your band or to
your sick brother in the neighboring village.
Therefore, we expect that the architecture of the human mind is full of evolved
variables, existing embedded in evolved circuits, whose function is to store propri-
etary parameters that are useful for regulating valuation, choice behavior, and
prospective computational preparation for future choice-forcing situations. Internal
regulatory variables are not explicit concepts, representations, or goal states, but
rather registers or indices that acquire their meaning by their location in the
architecture—for example from the situational cues that feed into them (e.g., co-
residence, perinatal association), and the evolved behavior-controlling and compu-
tation-controlling procedures that they in turn feed into (e.g., an estimated kinship
index between self and individual i, in turn leading to aversion at the prospect of sex
with i, a family member). Such regulatory variables may include measures of how
valuable to the individual a mate is, a child is, your own life is, and so on; how stable
or variable the food productivity of the habitat is; the distribution of condition-
independent mortality in the habitat; how long you have co-resided with an
individual; your expected future life span or period of efficacy; how good a friend
someone has been to you; the extent of your social support; how durable your social
partnerships are expected to be; your association value to others; your own and
others ability to inflict costs—aggressive formidabilities; your sexual attractiveness;
your status or self-esteem; the status of the coalition you belong to; present energy
stores; present health; how advantageous conception would be given your somatic
condition and circumstances; the degree to which subsistence requires collective
action, and so on.
Most evolutionarily recurrent situations and choice contexts that select for motiva-
tional subsystems and associated emotion programs involve the ongoing discovery of
information that allows and requires the recomputation of one or more of these
variables. Recalibration is, therefore, a major functional component of most emotion
programs. Recalibrational programs are components of emotion programs such as guilt,
gratitude, grief, depression, compassion and shame whose primary function is to carry
out such recomputations of internal regulatory variables (Cosmides & Tooby, 2013;
Tooby & Cosmides, 1990a, 2008; Tooby et al., 2008), rather than to orchestrate any
specific short-run behavioral response. Jealousy, for example, involves several sets of
The Theoretical Foundations of Evolutionary Psychology 67

3GC01 06/09/2015 12:40:44 Page 68
recalibrations (e.g., decrease in estimate of own mate value, decrease in trust in mate,
decrease in paternity confidence, increase in the benefit of eliminating the rival).
But information relevant to internal regulatory variables is not equally spread
throughout all points in time and throughout all situations. Some situations are
information dense, full of informative, ancestrally stable cues that reliably allowed
more accurate calibrations of what these variables should be set at (e.g., discovering
your child is dead; that your love is returned; your husband has not been faithful;
who your father really is; or that someone you know sacrificed a great deal on your
behalf). A well-designed architecture would exploit these information-dense situa-
tions to update the parameters in the system. This is particularly true since these
variables would logically exist in mutually interrelated networks. Among other
things, these networks need to exist to internally solve what microeconomists would
call pricing problems—computational problems that exist when there a large variety
of factors of production with different costs (e.g., different possible mutually
constraining courses of action), different possible products (outcomes with different
and mutually interacting payoffs), and so on. Externally caused changes in these
factors require extensive and spreading recomputation through the motivational
system. That is, new information relevant to opportunities, factors of production,
pay-offs, and uncertainties will necessarily have to ramify through the system that
governs the thousands of decisions a person makes each day. Accidentally spilling
your dinner in the dirt may require just a quick pang of annoyance or disappointment
toupdate;attheotherextreme,discoveringthatyourhusbandisdeadwillrequire
major changes in tens of thousands of decision-variables, trade-offs, and habit-
elements distributed throughout the architecture that have been calibrated in the past
to assume his presence (Who are you going to turn to when you are in trouble? Who
do you share food with? How vigilent do you have to be at night? Who will help care
for the children? How much food do you need to forage tomorrow?). For reasons that
are theoretically unclear, our brains are organized so that these recalibrational
processes often appear to require conscious attention to allow the appropriate
reweightings of the associated variables, and are associated with rich and distinct
affective feeling states that constitute a major dimension of human experience). These
emotions have often appeared puzzling from a functional perspective because the
feelings they engender interfere with short-term utilitarian action that an active
organism might otherwise be expected to engage in. For example, people voluntarily
or involuntarily take time out from obviously productive activities like foraging,
eating or sleeping in order to spend time feeling grief, depression, guilt, the onset of
romantic love, etc.). The suggestion here is that customary actions and stored
dispositions that were productive under one set of circumstances may no longer
pay off when the landscape suddenly changes, and people feel less motivated to act.
Indeed, people in grief or depression or infatuation show high levels of brain activity;
they want to be left alone, without outside demands on their attention. The brain
needs to revise large networks of regulatory and decision variables. The cognitive
sciences have devoted far more attention to cold cognition—perception, categoriza-
tion, language processing, object recognition—than to hot cognition. But we suspect
that far more of the brain may be organized to computationally implement feeling,
valuing, motivation, and emotion—hot cognition. Knowing what is in the world
(“objective”knowledge) is generally a far easier computational problem than
knowing what to do, and how much to value different courses of action (“subjective”
valuation that was fitness-enhacing).
68 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 69
The environment of evolutionary adaptedness was full of event relationships (e.g.,
mother is dead) and psychophysical regularities (e.g., blood indicates injury) that cued
reliable information about the functional meanings and properties of things, events,
persons, and regulatory variables to the psychological architecture. For example,
certain body proportions and motions indicated immaturity and need, activating
emotion programs for nurturing in response to “cuteness”releasers (see Eibl-Ebes-
feldt, 1970). Others indicated sexual attractiveness (Buss, 1994; Symons, 1979). To be
moved with gratitude, to be glad to be home, to see someone desperately pleading, to
hold your newborn baby in your arms for the first time, to see a family member leave
on a long trip, to encounter someone desperate with hunger, to hear your baby cry
with distress, to be warm while it is storming outside—these all mean something to us.
How does this happen?
In addition to the situation-detecting algorithms associated with major emotion
programs such as fear, anger, or jealousy, humans have a far larger set of evolved
specializations that we call recalibrational releasing engines that involve situation-
detecting algorithms and whose function is to provide inputs into internal regulatory
variables, and trigger appropriate recalibrations, including affective recalibrations,
when certain evolutionarily recognizable situations are encountered. Although these
pervasive microprograms construct a great deal of our world, investigations are only
beginning into adaptations of this nature.
WELFARE TRADE-OFF FUNCTIONS AND RECALIBRATIONAL EMOTIONS
Humans, like members of other social species, face a continuous flow of choices that
force them either to sacrifice another’s welfare to increase their own (selfish choices), or
to sacrifice their own welfare to increase the welfare of one or more others (altruistic
choices). Evolutionary biologists have identified a number of selection pressures for
which (under specified conditions) selection can favor trading off the immediate
welfare of the actor in favor of specific others. These include, among others, kin
selection (Hamilton, 1964; for evidence of adaptations in humans, see Lieberman, et al,
2007), reciprocation or exchange (Trivers, 1971; for adaptations in humans, see
Cosmides & Tooby, Chapter 25, this Handbook, Volume 2; Krasnow, Cosmides,
Pedersen, & Tooby, 2012), the asymmetric war of attrition (Hammerstein & Parker,
1982; for adaptations in humans, see Sell et al., 2009), and externality management and
partner choice (Tooby & Cosmides, 1996; see also Noë & Hammerstein, 1994).
For the human mind to solve the adaptive problem of motivating the actor to make
the fitness-promoting set of trade-offs between her own welfare and the welfare of
another under a given set of conditions, it must have adaptations designed to compute
regulatory variables that correspond to the relevant decision parameters (genetic
relatedness to this person—kinship index; did the person reciprocate previously?; how
much does the other person need this benefit? the formidability index of this person—
how much can this person injure me? the association value of this person, etc.) We and
our colleagues think these are organized through a human-universal motivational
subsystem in the mind which calculates, for each familiar individual, a welfare trade-
off function that sets thresholds (welfare trade-off thresholds) partitioning sacrifices
the individual is motivated to make on behalf of that familiar other from sacrifices the
individual is unwilling to make (Tooby & Cosmides, 2008; Tooby et al., 2008). These
thresholds should correspond, to the extent the system is well-engineered and
The Theoretical Foundations of Evolutionary Psychology 69

3GC01 06/09/2015 12:40:44 Page 70
operating under ancestral-like conditions, to sacrifices that were fitness-promoting,
and sacrifices that were fitness-reducing. This system also has to make estimates of
how valuable the act or resource is not only to the self, but also to the other party.
Each evolutionary theory of social interaction contains within it variables that help
specify how fitness-promoting a given welfare trade-off threshold from ito jwould be
(the kinship index, how reliable an exchange partner is, how much they value you, the
magnitude of benefits they can confer or withhold, etc.). This welfare trade-off
threshold (WTT) should be a quasi-stable variable—that is, it should be stable until
the system receives new information. When new information about these variables is
received, then the welfare trade-off threshold should be recalibrated to the magnitude
that is fitness-promoting under the new conditions.
When motivational problems are analyzed in terms of the internal regulatory
variables that would be needed to solve them, a pleasing finding is that many
hypothesized regulatory variables must be shared by a number of distinct motiva-
tional systems with different adaptive functions. For example, both of the independent
adaptive problems of how altruistic one should be toward relative iand how sexually
aversive one should find relative irequire the same regulatory variable: the kinship
index between self and I(Lieberman, et al 2007). One regulatory variable—the welfare
trade-off threshold—keeps reappearing in a broad variety of independent adaptive
motivational problems. For example, it is relevant to kin-directed altruism; to
exchange and reciprocation; to mateships and parenting; to aggression-based negoti-
ation; to benefit-based negotiation; to integrating externalities into social relationships;
to the management of social valuation, and so on.
Indeed, welfare trade-off thresholds and their recalibration appear to be deeply
embedded in the designs of a series of emotion programs: gratitude, anger, guilt,
compassion, shame, and contempt, to take leading examples. Anger appears to be
triggered when another person places too little weight on one’s own welfare (their
expressed WTT toward the self is too low), given the mind’s implicit estimation of
what welfare trade-off threshold it can plausibly enforce, given the person’s ability to
confer or withhold benefits, or inflict or withhold harms (Sell et al., 2009). That is, its
function is to bargain for a better WTT of the other to the self (or, if their WTT was
correct, but they did not understand how much you valued the service or resource, to
reeducate them). In cooperative relationships, the incentivization provided by the
angry individual to the other party is a threatened reduction in the angry individual’s
WTT towards the other: The other will no longer be able to expect the same delivery of
benefits through sacrifices unless their own welfare trade-off threshold towards the
angry individual is increased to acceptable levels. Guilt functions to recalibrate your
own welfare trade-off function towards a specific other when you get new information
indicating either that your previous welfare weighting on the other (as expressed
behaviorally) was too low, or that your estimation of the value a service or good to the
other person was too low—you did not know they cared that much (Tooby &
Cosmides, 1990a, 2008). Shame is the recalibrational emotion designed to deal
with the threat or actuality of negative information about you reaching others’minds,
so that they would devalue you—that is, the adaptive threat is others recalibrating
their welfare trade-off threshold toward you downwards in response to new infor-
mation about you (Sznycer et al., 2012).
Gratitude, correspondingly, is the recalibrational emotion program that is activated
in order to (1) increase the welfare trade-off threshold in the self toward another
person (2) upon discovery of new information that the association value of the other
70 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 71
person to the self is greater than previously estimated. For example, gratitude is
triggered when another person trades off their welfare for yours at a much higher level
than you had expected: They were unexpectedly kind to you, in a way not justified by
your previous treatment of them. Good cooperative relationships are rare, and the
higher the mutual welfare trade-off thresholds toward each other can become, (other
things being equal) the more efficiently your joint welfare can be promoted. In order to
stabilize this potential high level of mutual assistance, it is important to show the act
was noticed, attributed to the correct person, appreciated, and led to an increase in the
weight you place on your benefactor’s welfare. So the emotion program creates
communicative intent, and upregulates your welfare trade-off threshold toward the
other. This leads to a model of cooperation that is stabilized by the threat of the other’s
welfare trade-off threshold being downregulated (through the anger program) if your
WTT towards the other is too low; and your WTT toward the other being upregulated
(through the gratitude program) if their trade-off threshold toward you is higher than
yours presently justifies (Lim et al., forthcoming). A second kind of gratitude is not
based on exchange, but on association value and externalities (Tooby & Cosmides,
1996). Gratitude is triggered by high valuation toward the other party. The individual
may benefit by sacrificing for the welfare of a highly valued person, and the feeling of
valuation toward the person is also often called gratitude (i.e., you are grateful your
child lives; we are thankful for our blessings). Third, partner choice may be based on
the magnititude of positive externalities given off by the potential partner.
RECURRENT DIMENSIONS OF ENVIRONMENTAL
AND ORGANISMIC VARIATION SELECT FOR
PARAMETRIC COORDINATIVE ADAPTATIONS
When discussing the relationship between behavior genetics and human universal
design earlier, we postponed addressing one question: Why should some kinds of
individual differences in organisms be organized into a small number of dimensions of
variation? Over evolutionary time, many aspects of the world (including environ-
ments, organisms, and organism-environment interactions) shift within a evolutio-
narily recurrent covariant structure. That is, not only are there stable regularities in the
world (e.g., gravity, the properties of light, the proportion of oxygen in the atmo-
sphere), and stable regularities in the dimensions along which conditions and
phenotypes vary, but there are also higher order covariant relationships in conditions
and in phenotypes. For example, regional temperature may shift, but if the tempera-
ture increases then humidity increases in a coupled fashion. Moreover, aspects of the
environment and internal species organization may systematically co-vary as well. For
example, the environment may sometimes select for an increase in species size, and at
other times a decrease in species size. For a functionally scaled organism to be
maintained if (for example) head size increases, larger vertebrae are required, as
are greater neck and torso muscles as well. If all the size dimensions of the organism
were under independent genetic control, then for the species to grow (or shrink),
selection would have to independently occur in all functionally interrelated traits
throughout the organism, slowing down the rate at which the lineage can respond. A
mutation in one of these may not even be advantageous without others. The ability of
the lineage to shift in size in response to selection would be considerably impeded.
Alternating selection for larger and smaller size ought to therefore also select to for
The Theoretical Foundations of Evolutionary Psychology 71

3GC01 06/09/2015 12:40:44 Page 72
welding together formerly independent growth in various traits into a far smaller
number of developmental growth fields. This would be favored because then the
organism’s traits maintain functional allometric interrelationships while undergoing
directional selection. That is, mutations will be favored that developmentally link
formerly independent traits whose payoffs vary in response to a particular recurrently
varying environmental dimension. Thus, genetic correlation among traits is not, as is
often thought, a given, but is itself an evolved outcome. Given such genetic and
developmental adaptations, expressed phenotypes should track recurrently variable
environments with far less lag time, and remain somewhat closer to fitness optima.
The covariant structure of conditions, and the covariant structure of the developmen-
tal system should evolve to complement each other over evolutionary time.
In the case of allometry, the number of (for example) growth fields ought to be
reduced to the number of phenotypic dimensions that benefit by being tuned by
selection independently. Quantitative genetic variation that moves designs along
these dimensions is expected to have accumulated because adding numerous loci to
the determination of a quantitative trait slows loss of variants, maintaining the ability
of the lineage to respond more rapidly to reversals in directional selection. This
reduces the risk of being stuck at fixation at a ceiling or floor when directional selection
reverses again. (Surprisingly, Gould and Lewontin (1979) considered observed allo-
metric relationships to be “constraints”on development and hence constraints on
adaptive design itself, rather than adaptations themselves. Not only does simple
mechanics support the functional nature of many of these scaling relationships, but
selection must be actively maintaining these allometric relationships, because there are
always outliers in populations and species that deviate from these relationships that
selection could act on; and related species deviate from each other as well.)
Of course, a better design would be to have developmental adaptations that
facultatively regulate trait expression so that the expressed phenotype matches the
demands of the specific environment the organism matures in (rather than, if the the
phenotype was determined by inherited genetic differences, a random sample out of a
cross-generationally lagging quantitative genetic distribution). Indeed, many individ-
ual differences appear to be due to the operation of calibrational adaptations (muscle
increase due to increased exercise; the development of calluses in response to abrasion;
increased storage of fat in response to a history of calorie flow variance). Regardless
of whether the phenotype is facultatively calibrated or just determined by alleles, the
dimensions of variation would be the product of selection. For developmental
adaptations for matching local demands, both the dimensions of variation and the
regulation of the individual outcome would be an expression of the adaptative system.
That is, adaptations would take environmental or organismic condition as input, and
produce a facultatively calibrated phenotypic outcome. In contrast, for systems of
individual differences caused directly by quantitative genetic variation, the system
would be the product of selection, but the particular outcome for an individual would
be a random but beneficially biased outcome. A third (and likely) possibility is that
given there is genetic noise throughout the system, there should be phenotypic
differences (like differences in strength) which species-typical adaptations would
respond functionally to—what we have called reactive heritability (Tooby &
Cosmides, 1990b)—see discussion of anger, strength, and heritable factors leading
to differences in strength below. More generally, if over evolutionary time there is
covariation in the independent adaptive demands placed by the environment on the
organism (called here a selective regime), and a set of independent traits with shifting
72 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 73
parameter values that are best responses to these demands, then one expects the
evolutionary emergence of parameterized dimensions of covariation in the develop-
mental expression for these formerly independent adaptations. There need be no
logical coherence or functional necessity to the set of properties that are phenotypically
scaled together: Just that the organism did better when these properties were increased
(or decreased) together when an environmental variable (or covariant set of variables)
moved together.
Dimensions of personality variation are potential candidates for adaptationist
explanations of this kind. Although there ought to be individual differences caused
by genetic noise permeating the human neurocomputational architecture, the consist-
ent emergence (for some sets of individual differences) of a far smaller number of
robust dimensions in personality suggest that patterns like the five factor model or the
HEXACO model (Ashton et al., 2004) might be the product of adaptations to the
covariant structure of selective regimes. In exploring this hypothesis, it is always
preferable to begin with theoretically well-motivated theories of adaptive function,
rather than simply constructing explanations after the fact.
Consider, for example, the hypothesis that the human anger program is a species-
typical adaptation that evolved to orchestrate an individual’s bargaining behavior in
conflicts of interest so that they secure for themselves an advantageous resolution of
the conflict (Sell et al., 2009). Power in bargaining comes from the ability to confer or
withhold benefits, and the ability to inflict or refrain from inflicting harm. This theory
predicts that individual differences in the ability to inflict harm (for example, by upper
body strength) and the ability to confer or withhold benefits (for example, by
attractiveness) should calibrate how successfully the individual will be able to
incentivize better treatment for him- or herself using these advantages. Therefore,
upper body strength and attractiveness were predicted to calibrate how readily the
individual angers; how entitled they feel to better treatment; how successful they are
in resolving conflicts of interest in their favor; (for strength) how useful they think
force is in resolving disputes, and so on (these predictions were supported; Sell et al.,
2009). This provides a case study of a theoretically derived, empirically supported
adaptationist theory of some types of individual differences: Individual differences in
inputs (strength, attractiveness, being male, being female) fed into the species-typical
negotiative system then outputs advantageously calibrated behaviors and motiva-
tional settings. The species-typical adaptation creates a systematic and adaptively
calibrated functional relationship between the magnitudes of some individual differ-
ences and the magnitudes of others. This is simultaneously consistent with the
possibility of high heritability in anger proneness and entitlement (for example),
because there is likely to be genetic variation in the factors that produce upper body
strength and beauty (e.g., reactive heritability; Tooby & Cosmides, 1990b). These are
processed by the organism just as if they were environmentally caused individual
differences: The organism must respond adaptively to its own condition, however
caused. Hence, a human-universal adaptation (the anger program) can, by taking in
heritable inputs (individual differences in strength caused by individual differences in
genes), produce functionally calibrated individual differences in anger proneness.
These results only scratch the surface of the potential ramifications of the evolved
bargaining system on individual differences. Stronger and more attractive people will
have less to fear from interacting with larger numbers of less familiar others, and
because of the nature of social markets, will have more to gain. This predicts that there
should be a functionally calibrated relationship between strength and attractiveness
The Theoretical Foundations of Evolutionary Psychology 73

3GC01 06/09/2015 12:40:44 Page 74
and the extraversion-introversion dimension. Lukaszewski and Roney (2011) did
superb work testing this hypothesis in two studies, along with simultaneously
investigating the contribution of specific genetic polymorphisms (the AR CAG repeat
polymorphism) to extraversion. They found the relationship between strength and
attractiveness and extraversion to be, as they predicted, high (and varying by sex in the
expected directions). They also found that the AR CAG polymorphism accounted for
some of the variance in extraversion.
So we know that at least some dimensions of personality variation are the product
of parametric coordinative adaptations similar to, but more slowly changing than
emotional states (e.g., strength persists far longer than a snake appearing in the path).
Just as emotion programs involve adaptively coordinating multiple mechanisms
within the architecture to the adaptive demands posed by an evolutionarily recurrent
short term adaptive problem, personality factor (and especially subfactor) phenotypes
are proposed to be parameterized coordinative adaptations to evolutionarily recurrent
and longer-lasting selective regimes. That is, they are hypothesized to be best-bet
deployments of the mechanisms in the psychological architecture and body, given the
developmental adaptations’reading of the individual’s location within the covariant
structure of adaptive demands posed by the environment and the individual’s own
condition. These conditions may disappear over the lifespan (e.g., with loss of
strength), but might last several generations (e.g., a high-warfare social ecology).
Consider dimensions of the biotic and social ecology: Some environments will have
higher rates of predation; some will have higher rates of warfare and/or within-group
exploitation; some biotic and social ecologies impose zero-sum relationships between
individuals and groups (where resource extraction by some intrinsically decreased
resource extraction by others; some environments will be less abundant, or have
periodic famines; some will have higher rates of disease; some individuals will be
weaker or less attractive or have fewer kin). Optimal settings on anxiety, fear,
thresholds for project abandonment in the face of risks or setbacks (i.e., proneness
to discouragement or depression), vulnerability by gender, willingness to defer
gratification, willingness to trust, rivalrousness, anger and so on should all shift
depending on predictable features of the self, the social ecology, and the biotic ecology.
These would provide a straightforward functional interpretation for the dimension of
neuroticism, and for subfactors in agreeableness or HEXACO’s honesty-humility.
This suggests an entirely different framework for research into personality. Instead
of starting with empirical relationships of unknown functional significance and
unknown ecological validity, it might be useful to (1) make or adopt models of
adaptations that (2) need to take as inputs—in order to perform their function—
locations along adaptively salient dimensions in ancestral environments and individ-
ual conditions, and (3) attempt to identify which adaptations should facultatively shift
in response to movement along the same dimensions. By starting with specific
adaptations and adaptive problems, and considering how sets of them should jointly
vary by individual condition and ecology, one might be able to derive a principled
series of empirically validated theories of personality variation. One might work
upward, from specific adaptive problems and associated subscales, to larger sets of
adaptations with more weakly associated responses to dimensions of ecological or
phenotypic variation.
Indeed, there is no reason to attempt to force dimensions of personality variation to
be statistically independent. On the contrary, one would expect from first principles
that dimensions would derive from how the structure of variation in the environment
74 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 75
drives demands for coordinated adaptive responses. There is no reason whatsoever to
expect these dimensions of variation to all be orthogonal to each other (e.g., the degree
to which social interactions in the social ecology are highly positive-sum might impact
both agreeableness and honesty-humility, but neuroticism to a lesser degree). Also,
researchers ought to be open to the discovery of major, previously unknown person-
ality dimensions, since existing dimensions were empirically derived overwhelmingly
in developed mass societies and abundant environments, rather than by (impossibly)
censusing the range of fitness regimes that characterized the ancestral world.
For example, ancestrally, the fitness of individuals or sets of individuals might have
been inversely related, unrelated, or positively related. Our normal intuitive expect-
ations of rationality (characterized by a set of social orientations, emotional calibra-
tions, ways of interpreting events, and motivated appetites forged in cooperative and
positive sum social ecologies) we suspect is just one parameterization of a coordinative
adaptive system capable of creating very different rationalities, including what might
be called predatory rationality. In a fitness regime where those who socially interact
are typically in intense negative sum or zero sum relationships with each other
(because competition is local), win-win strategies are not seen to be best-bet responses;
strength and aggressive formidability are highly prized and cultivated; there are no
inhibitions on preying on the weaker; audacious predatory attacks and the extermi-
nation or humiliation of the antagonist is more attractive to the predatory-minded
than to whose who have a cooperative orientation; cooperativeness, paranoia, gener-
osity, revenge-proneness, envy, sensitivity to cues of fitness differentials and status
differentials, propensity to exploit—all these are set at surprisingly different levels
(Sznycer, et al, forthcoming; Tooby et al, forthcoming).
WHY MIGHT SOME COORDINATIVE CALIBRATIONS BE
PARAMETERIZED BY ONTOGENETIC INPUTS, SOME
BY QUANTITATIVE GENETIC INHERITANCE, AND
SOME BY INHERITED EPIGENETIC INFORMATION?
For aspects of the world where the variance in the situation distribution is small (e.g.,
the geometry and physics of light), then a single design can uniformly develop (e.g.,
the visual system) to reliably improve the behavioral output of members of the species.
In contrast, where variance in the situation distribution is large, a uniform expressed
phenotype will rarely be the best solution. In such cases, fitness is enhanced to the
extent that regulatory designs match their phenotypic outputs (e.g., mature early;
invest in larger musculature; extend less credit in cultivating cooperative relation-
ships) to the demands of actual conditions (e.g., greater extrinsic mortality; a social
ecology of greater competition; a social ecology of lower payoffs to cooperation). In
this case, the underlying uniform adaptation lies in the design of the regulatory
machinery that parameterizes the expressed phenotypes to the particular situations
that it will be facing.
What is key is that there be a principled guidance system whose evolved architec-
ture decides on and then implements those targeted phenotypic modifications that
correctly close the gap between the needed phenotype (in a given situation) and the
realized phenotype, over the range of situations the species typically faces (Tooby &
Cosmides, 1992). Success in the game of matching phenotype to circumstances would
be impossible for the architecture unless there existed (1) information that (with
The Theoretical Foundations of Evolutionary Psychology 75

3GC01 06/09/2015 12:40:44 Page 76
computation) systematically predicted which circumstances the organism was going
to face, (2) a repertoire of phenotypic alternatives that encoded the phenotypic
modifications that would be needed for the organism to develop a high-performing
response for those circumstances, and (3) a function that mapped the information
about circumstances to the best bet phenotypic alternative. Fortunately for organisms,
there are many such systematic relationships that natural selection has exploited to
build developmental or facultative adaptations that successfully solve these problems
(e.g., skin darkening in response to sun exposure).
The two key questions governing how these systems evolve are (Tooby, 1976;
Tooby & Cosmides, 2003; Tooby, Cosmides, & Barrett, 2003):
1. When does the information become available that is needed to decide on the
best-bet phenotype?; and,
2. How much lead time does the organism need to generate or construct the best-
bet phenotype in time for it to discharge its function?
For the startle reflex, the information needed to protect the eyes does not become
available until a few hundred milliseconds before impact; but the system needs
around 50 milliseconds to begin to respond. So flinching is linked to rapid looming
(and if it is faster than that, you are out of luck). It would make no sense to determine
the time of the flinch minutes, hours, days, or years in advance, because time the flinch
is needed could not be known to sufficient precision earlier. The best design (where
possible) is one in which response selection can be cost-effectively postponed until the
environmental demand can be assayed with high reliability (e.g., through perception);
and only then is the phenotypic response selected and implemented. To take a more
interesting example, some fish species change sex based on their relative size and the
death of the dominant local male (Warner, 1989). Humans and other mammals, in
contrast, use a genetic sex determination system, presumably because you can build a
better woman or man if you start very early in development differentiating the
adaptations of the two phenotypes, long before there any is useful information about
the adult sex ratio will be at maturation. We have to place our gender bets before we
have information (so genetic sex determination flips a coin). For systems involving
major tissue differentiation, intricate wiring, and/or long-term nutrient flow (like
becoming female or male), construction needs to begin very early. Similarly, to acquire
large databases of intricately patterned information (as in acquiring a skill given a
sexual division of labor), the human child also may not be able to afford to wait too
long. It seems likely that there exist developmental adaptations whose function is to
make predictive inferences about the adult fitness regime by sampling self and world
early in life, and then using these predictions to calibrate life history (see, e.g.,
Griskevicius et al., 2011). However, for many traits, early life does not predict the
best bet later in life, either because the correlation is too low, or the sample is too brief
and unrepresentative to be useful.
What then? It must often be the case that there is a correlation of conditions among
adjacent generations in certain respects (e.g., if a mother faces an exceptionally
competitive, predatory, or food limited environment, then there is an increased
probability that offspring will too—and with some decay function, that subsequent
generations will as well). If this information exists, it is available long before
development even begins. Such cases would select for coopting non-DNA-based
systems of inheritance that could transmit regulatory signals from one or more
76 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 77
generations to the next. Although several of these systems have been suspected or
known for decades (Cosmides & Tooby, 1981; Tooby, 1976; Tooby et al., 2003), only
recently have they become a focus of interest (see Jablonka & Lamb, 2005, for review).
We believe that cross-generational epigenetic effects are not simply accidental
byproducts, but that they are evolved adaptations with functions. Most generally, the
function of these signals is to parameterize individual development so it goes along
pathways that better suit it to the conditions it is likely to face across its life. Given the
operation of such systems, individual differences in phenotypes would be partly
calibrated from environmental cues during ontogeny (conditions less far away in time
should be more diagnostic); partly epigenetically inherited (i.e., frequencies of events
summed over multiple generations are going to provide an independent method of
improving predictive validity); these systems should cause heightened parent–off-
spring phenotypic similarity in a way not attributable to DNA-sequence differences.
Indeed, these additional systems of inheritance would be selected to use nonDNA-
based mechanisms, because DNA sequence transmission is too high fidelity to be
useful for tracking rapid changes across multiple generations.
To take a hypothetical example, if the mother is made repeatedly fearful by
exposure to predators in an enduringly predator-rich environment, then signals
transmitted by methylation, in utero, or in early maternal care to the offspring could
be designed to cause it to develop a predator-cautious phenotype usefully in advance
of experiencing attacks by local predators. Depending on the temporal structure of the
environmental change, these systems could be designed to be passed on signals
according to a multigenerational decay function to subsequent generations. That is, by
including (say) three generations of information gathering on the frequency of
droughts, the system could make better predictions than if it simply used one.
Similarly, if the parents (and/or other close lineal ancestors) are food limited, and
such a condition often persists across generations, then the offspring would benefitby
developing a more frugal metabolism, selecting for an inheritance system that
regulates metabolism and life-history across generations. To take a third case, if
the parents are in an exceptionally competitive environment, then offspring would
benefit by developing a more aggressive, territorial, competitive phenotype, with a
greater tendency to emigrate, delayed maturation, and a greater tendency to bias
uterine sex ratio toward the more dispersing sex. Not only have many of these
empirical relationships been observed (Clark & Galef, 1995; Clark, Karpiuk, & Galef,
1993; Francis, Diorio, Liu, & Meaney, 1999), but they fit elegantly into an evolutionary
psychological theory of functional development.
From this theoretical vantage point, cross-generational inheritance effects are not
only unsurprising but are instead predicted for traits whose value depends on
conditions (a) that frequently endure across more than one generation; or (b) whose
probability of occurrence in the upcoming generation can be better estimated using
their incidence over multiple recent generations; and (c) that repeatedly cycle along
dimensions of variation across generations (Tooby, 1976; Tooby et al., 2003). We
predict that such inheritance systems should be be especially prominent in regulating
traits that are used starting early in the life cycle (e.g., frugal metabolism, predator-
evasion tactics, physiology tuned to local conditions; affiliativeness to coalitions;
ecological incidence of positive vs. zero-sum/negative sum interactions) or that are
less costly or more effective if the organism begins to develop them prior to directly
detecting the conditions it will be facing (e.g., life history trajectory, competitive
ability, size reduction and heightened fat stores for better survive food interruptions).
The Theoretical Foundations of Evolutionary Psychology 77

3GC01 06/09/2015 12:40:44 Page 78
One might also expect that the sex that disperses less would be selected to engage in
more epigenetic inheritance, since the correlation of environments of parent and
offspring would be higher (Tooby, 1976).
In sum, these systems should evolve and regulate development cross generation-
ally when (1) the dimensional covariation they track is autocorrelated on time scales
longer than a generation—i.e., conditions persist long enough in the environment (or
in the lineage); (2) when the optimal developmental coordinative response or strategy
is dependent on being parameterized by information about the position of the system
in its dimensions of variation; (3) for developmental processes such as physiological
adjustment or expertise acquisition that need to begin early in development (or at least
before the task environment that must be prepared for can be directly perceived). It is
important to recognize that not just physical features of the environment (e.g. climate)
fit these criteria. Local social ecologies (cooperative hunting; warfare; intensity of
individual competition) and biotic ecologies (disease, food abundance, predation) fit
these criteria, as do individual heritable somatic features such as strength, reflexes, or
genetic impairment.Finally, it bears noting that the methods used by behavior
geneticists would tend to misattribute systems of epigenetically regulated individual
differences in behavior to genetic differences. Epigenetic states, like the genes they
adhere to, are passed down from parents to offspring, and make family members more
similar to each other that they would be to nonrelatives. Since these are the sources of
data that behavior geneticists use to compute heritability it is indeed possible that a
great proportion of the variance in individual behavioral (and somatic) phenotypes
that has been attributed to DNA sequences are instead due to epigenetic systems. This
would make particular sense for dimensional systems of personality variation which
may well be the product of parameterized coordinative adaptations. This would
explain why it has been so difficult to track down and identify many DNA sequences
that can be shown to explain observed behavioral differences. If epigenetic systems
can process information more rapidly that selection acting on quantitative genetic
variation (which they can), and if they can efficiently parameterize coordinative
adaptations so phenotypes are better matched to ontogenetic conditions (which seems
likely), then behavior genetics findings may mostly be behavior epigenetics findings.
An examination of the subtly discordant empirical models of heritability across
different familial pathways might be more consistent with more rapidly mutating
epigenetic transmission. One might predict that traits that turn out to be determined
by quantitative genetic variation will be ones where the temporal structure of
successive environments cannot be better predicted by sampling over small numbers
of generations (that is, environmental autocorrelation is too low to be useful), and the
best that can be done is to sample randomly quantitative genetic variation from the
broader population.
How could this make sense of dimensions of personality variation? In the first
place, one would expect that—just as selection has merged independent traits into
allometric growth fields so that the species or population responds to selection more
rapidly—selection would have done the same to quantitative settings in sets of
psychological adaptations that would benefit from being adjusted together. If,
cross-generationally, ecologies shift periodically into conditions where male-male
competition is more intense, then all of the parameterizations of psychological
adaptations that improve performance in male competition might be linked together
so the dial can be turned up or down (leading to parameterizations of, e.g., the shame-
honor system). Initially, this would select for systems of quantitative genetic variation
78 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 79
in such a dimension. Similarly, where extrinsic mortality is high, this might link fear-
anxiety sensitivity to a lower propensity to defer gratification (e.g., showing up in
personality psychology as psychoneuroticism). Second, where reliable information
exists during development, then cues may parameterize the coactivation of adapta-
tions functionally. Third, where the parameterization bet can be improved by
information transmitted from previous generations, epigenetic systems would be
one adaptive calibrational system that natural selection could construct. So personality
variation would be the functional product of the program architecture of psychologi-
cal adaptations, as calibrated by cues during development; heritable differences in
other parts of the organism (e.g., strength) being functionally responded to by these
adaptations (e.g., bargaining power, anger); epigenetic signals sent from past genera-
tions, which improve the ability of the developmental system to bet on the best
parameterizations of adaptations for the organism; or, failing that, falling back of
quantitative genetic variation responding slowly to recent selection on the population.
THE FUTURE OF EVOLUTIONARY PSYCHOLOGY
AND A UNIFIED SOCIAL SCIENCE
The amazingly high levels of functional order that are found in evolved systems is so
dazzlingly intricate, that originally the machinery of life clearly seemed to be the
handiwork of an omniscient transcendental craftsman, who built physical systems far
beyond human comprehension. Since Darwin’s discovery of how blind causality can
push replicating systems uphill against the physical tendency of ordered systems to
deteriorate, and the other natural sciences made rapid strides in understanding micro-
scale causation, we now have an emerging skeletal framework around which to
organize our understanding of life forms. Yet it remains important to recognize that at
every scale and level of organization, the structure of biological systems is so
labyrinthine and sophisticated that what that we so far understand is merely the
nearest edge of a vast space of unseen and uncharted evolved organization.
Therefore, we are only at the beginning of an age of extraordinary discovery, and
we should be open to surprising transformations and additions to our knowledge. To
judge by the systems (like the visual system) we understand to some limited degree,
natural selection produces exquisitely subtle and sophisticated functional complexity
that may be likened to a high technology developed by extraterrestrials millions of
years ahead of us. The key idea is that natural selection tends to build subtle
adaptations out of the enduring structure of the world, the information ecology
provided by that structure, and by the computational or regulatory power provided
wherever biological structure can be hijacked to provide it. So, we can expect many
unexpected and major discoveries about how these are woven together functionally.
For example, the DNA and RNA machinery inside individual cells provide all the
elements necessary for each cell to function as an assemblage of Turing machines. It
seems unlikely that selection would have left this vast computational power
untapped, which means that a great deal of computation might be taking place
within and not just between neurons. So we expect that the neuron—once regarded as
a mere on-off switch—will eventually be found to be something much more like an
integrated circuit. Similarly, we expect that the epigenetic machinery underlying
cellular differentiation in ordinary development has also been hijacked to transmit and
process information across generations through the genetic machinery in gametes and
The Theoretical Foundations of Evolutionary Psychology 79

3GC01 06/09/2015 12:40:44 Page 80
other biologically active molecules provided by parents to offspring. Similarly, the
immune system is both capable of recognizing immense numbers of proteins, and
simultaneously monitoring various components of health, and so we consider it likely
that this has been conscripted as a powerful organ of perception and dietary
regulation.
Finally, we briefly return to an earlier question. We began our discussion of
traditional versus evolutionary approaches to psychology by noting that humans
are able to solve a wide array of problems that were no part of their evolutionary
history, and that this observation lent appeal to the view that the mind is a general-
purpose machine. But this is to confuse the range of problems solved with the
architecture that solves it. One could get breadth not only by having a general
purpose architecture (an unspecified, hypothetical and arguably incoherent entity),
but alternatively by bundling an increasing number of specializations together, each
capable of solving an additional class of problems. Moreover, it leaves open the
possibility of evolved architectures that include numerous specializations, plus
additional components designed to exploit the specializations by integrating infor-
mation from across these systems to manufacture a flexibly deployable array of tools
to attack novel problems (e.g., the concept of causation in the object mechanics system
provided the core concept that was used to develop modern science).
The evolved architecture of the mind includes specialized mechanisms that permit
offline, decoupled cognition. These include metarepresentations, imagery, and a scope
syntax, which together can interact with the outputs of domain-specific mechanisms to
allow counterfactual and suppositional thinking that is basic to human evaluation,
decision-making, and causal reasoning (Cosmides & Tooby, 2000a; Leslie, 1987;
Sperber, 1994). Decoupled cognition may have evolved to help calibrate or recalibrate
mechanisms through experiencing evaluative feedback from imagined or planned
outcomes, infer other people’s mental contents, or imagine solutions to social, tool use,
or other ancestral problems. But it seems likely that, whether as byproducts or not,
decoupled cognition also permits the kind of thinking that underlies scientific
discovery, religious ideas, and other uniquely human preoccupations (Boyer, 2001;
Cosmides & Tooby, 2000a, 2001; Sperber, 1994; Tooby & Cosmides, 2001).
In sum, the century long scientific program that assumed that the human psycho-
logical architecture consisted predominantly of general purpose, content-indepen-
dent, equipotential mechanisms has failed to explain much of human behavior.
Indeed, it has failed even to develop a set of persuasive models about what the
computational architecture of putatively general purpose learning, rationality, or
intelligence would look like, and cannot account for any significant kind of human
activity. In contrast, evolutionary theory when joined with a computational approach
to the mind leads to the conclusion that the human psychological architecture is very
likely to include a large array of adaptive specializations. Evolutionary psychologists,
and others, have found detailed empirical confirmation of a large series of narrow,
deductive predictions derived from models of evolutionarily specialized computa-
tional adaptations.
Accordingly, we think that, over the next four or five decades, as a large scale
collaborative program by the scientific community, it may be possible to turn human
nature from a vague idea into a set of precise, high resolution models of our evolved
computational architecture—models that can be cashed out genetically, at the cellular
level, developmentally, physiologically, and neurally. These in turn can then inform
models of social interactions and culture, providing a foundation for a more rigorous
80 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY
no

3GC01 06/09/2015 12:40:44 Page 81
and integrated social science. It will be a fundamental advance for our species once we
have constructed a true, natural science of humanity.
REFERENCES
Alcock, J. (2001). The triumph of sociobiology. Oxford, England: Oxford University Press.
Anderson, A., & Phelps, E. (2001). Lesions of the human amygdala impair enhanced perception of
emotionally salient events. Nature,411, 305–309.
Ashton, M. C., Lee, K., Perugini, M., Szarota, P., de Vries, R. E., Di Blas, L., . . . De Raad, B. (2004). A six-
factor structure of personality-descriptive adjectives: Solutions from psycholexical studies in seven
languages. Journal of Personality and Social Psychology,86(2), 356–366.
Atran, S. (1990). Cognitive foundations of natural history. Cambridge, England: Cambridge University Press.
Atran, S. (1998). Folk biology and the anthropology of science: Cognitive universals and cultural particulars.
Behavioral and Brain Sciences,21, 547–611.
Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press.
Barrett, H. C. (1999). From predator-prey adaptations to a general theory of understanding behavior (Doctoral
dissertation, Department of Anthropology, University of California, Santa Barbara).
Barrett, H. C. (2005). Enzymatic computation and cognitive modularity. Mind and Language,20(3), 259–287.
Barrett, H. C. (2015). The shape of thought: How mental adaptations evolve (Evolution and cognition) New York,
NY: Oxford University Press.
Barrett, H. C., Cosmides, L., & Tooby, J. (in press). By descent or by design? Evidence for two modes of biological
reasoning.
Barrett, H. C., Tooby, J., & Cosmides, L. (in press). Children’s understanding of predator-prey interactions:
Cultural dissociations as tests of the impact of experience on evolved inference systems.
Blurton Jones, N., & Konner, M. (1976). !Kung knowledge of animal behavior (or The proper study of
mankind is animals) In R. Lee& I. Devore (Eds.), Kalahari hunter-gatherers: Studies of the !Kung San and their
neighbors (pp. 325–348). Cambridge, MA: Harvard University Press.
Boole, G. (1848). The calculus of logic. Cambridge and Dublin Mathematical Journal,III, 183–98.
Boyer, P. (2001). Religion explained: The evolutionary roots of religious thought. New York, NY: Basic Books.
Brase, G., Cosmides, L., & Tooby, J. (1998). Individuation, counting, and statistical inference: The role of
frequency and whole object representations in judgment under uncertainty. Journal of Experimental
Psychology: General,127,1–19.
Braun, J. (2003). Natural scenes upset the visual applecart. Trends in Cognitive Sciences,7(1), 7–9.
Brown, D. (1991). Human universals. New York, NY: McGraw-Hill.
Buss, D. (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures.
Behavioral and Brain Sciences,12,1–49. doi: 10.1017/S0140525X00023992
Buss, D. M. (1994). The evolution of desire. New York, NY: Basic Books.
Buss, D. (1999). Evolutionary psychology: The new science of the mind. Boston, MA: Allyn & Bacon.
Buss, D. M. (2000). The dangerous passion. London, England: Bloomsbury.
Campos, J., Bertenthal, B., & Kermoian, R. (1992). Early experience and emotional development: The
emergence of wariness of heights. Psychological Science,3,61–64.
Cannon, W. (1929). Bodily changes in pain, hunger, fear and rage. Researches into the function of emotional
excitement. New York, NY: Harper & Row.
Caramazza, A. (2000). The organization of conceptual knowledge in the brain. In M. S. Gazzaniga (Ed.),
The new cognitive neurosciences (2nd ed., pp. 1037–1046). Cambridge, MA: MIT Press.
Caramazza, A., & Shelton, J. (1998). Domain-specific knowledge systems in the brain: The animate-
inanimate distinction. Journal of Cognitive Neuroscience,10,1–34.
Carruthers, P. (2006). The case for massively modular models of mind. In R. Stainton (Ed.), Contemporary
debates in cognitive science (pp. 3–21). Oxford, England: Blackwell.
Cheney, D., Seyfarth, R., Smuts, R., & Wrangham, R. (Eds.). (1987). Primate societies. Chicago, IL: University
of Chicago Press.
Chomsky, N. (1959). A review of B. F. Skinner’s Verbal Behavior. Language,35(1), 26–58.
Chomsky, N. (1965). Aspects of a theory of syntax. Cambridge, MA: MIT Press.
Cosmides, L. (1985). Deduction or Darwinian Algorithms? An explanation of the “elusive”content effect on the
Wason selection task (Doctoral dissertation, Harvard University). (UMI No. #86–02206)
Cosmides, L., & Tooby, J. (1981). Cytoplasmic inheritance and intragenomic conflict. Journal of Theoretical
Biology,89,83–129.
The Theoretical Foundations of Evolutionary Psychology 81

3GC01 06/09/2015 12:40:44 Page 82
Cosmides, L., & Tooby, J. (1987). From evolution to behavior: Evolutionary psychology as the missing link.
In J. Dupre (Ed.), The latest on the best: Essays on evolution and optimality. Cambridge, MA: MIT Press.
Cosmides, L., & Tooby, J. (1994a). Beyond intuition and instinct blindness: The case for an evolutionarily
rigorous cognitive science. Cognition,50,41–77.
Cosmides, L., & Tooby, J. (1994b). Origins of domain-specificity: The evolution of functional organization.
In L. Hirschfeld& S. Gelman (Eds.), Mapping the mind: Domain-specificity in cognition and culture
(pp. 85–116). New York, NY: Cambridge University Press.
Cosmides, L., & Tooby, J. (1996a). Are humans good intuitive statisticians after all? Rethinking some
conclusions of the literature on judgment under uncertainty. Cognition,58,1–73.
Cosmides, L., & Tooby, J. (1996b). A logical design for the mind? (Review of The psychology of proof, by Lance
J. Rips, 1994 MIT Press.) Contemporary Psychology,41, 448–450.
Cosmides, L., & Tooby, J. (2000a). Consider the source: The evolution of adaptations for decoupling and
metarepresentation. In D. Sperber (Ed.), Metarepresentations: A multidisciplinary perspective (pp. 53–115).
New York, NY: Oxford University Press.
Cosmides, L., & Tooby, J. (2000b). Evolutionary psychology and the emotions. In M. Lewis& J. M. Haviland-
Jones (Eds.), Handbook of emotions (2nd ed., pp. 91–115). New York, NY: Guilford Press.
Cosmides, L., & Tooby, J. (2001). Unraveling the enigma of human intelligence: Evolutionary psychology
and the multimodular mind. In R. J. Sternberg& J. C. Kaufman (Eds.), The evolution of intelligence
(pp. 145–198). Hillsdale, NJ: Erlbaum.
Cosmides, L., & Tooby, J. (2013). Evolutionary psychology: New perspectives on cognition and motivation.
Annual Review of Psychology,64, 201–229.
Daly, M., & Wilson, M. (1988). Homicide. New York, NY: Aldine.
Daly, M., Wilson, M., & Weghorst, S. J. (1982). Male sexual jealousy. Ethology and Sociobiology,3,11–27.
Darwin, C. (1859). On the origin of species. London, England: John Murray.
Dawkins, R. (1982). The extended phenotype: The long reach of the gene. New York, NY: Oxford University Press.
Dawkins, R. (1986). The blind watchmaker. New York, NY: Norton.
Dawkins, R. (1996). Climbing Mount Improbable. New York, NY: Norton.
Defeyter, M. A., & German, T. (2003). Acquiring an understanding of design: Evidence from children’s
insight problem solving. Cognition,89, 133–155.
DeSteno, D., Bartlett, M., Braverman, J., & Salovey, P. (2002). Sex differences in jealousy: Evolutionary
mechanism or artifact of measurement. Journal of Personality & Social Psychology,83(5), 1103–1116.
Delton, A. W., Cosmides, L., Guemo, M., Robertson, T. E., & Tooby, J. (2012). The psychosemantics of free
riding: Dissecting the architecture of a moral concept. Journal of Personality and Social Psychology,102,
1252–1270. doi: 10.1037/a0027026
Delton, A., Krasnow, M., Cosmides, L., & Tooby, J. (2011). The evolution of direct reciprocity under
uncertainty can explain human generosity in one-shot encounters. Proceedings of the National Academy of
Sciences, USA,108(32), 13335–13340. doi: 10.1073/pnas.1102131108
Dennett, D. (1987). The intentional stance. Cambridge, MA: MIT Press/Bradford.
DeVore, I. (1962). The social behavior and organization of baboon troops. Doctoral dissertation, University of
Chicago.
DeVore, I. (1965). Primate behavior: Field studies of monkeys and apes. New York, NY: Holt, Rinehart & Winston.
Eaton, S. B., Shostak, M., & Konner, M. (1988). The Paleolithic prescription: A program of diet, exercise and a design
for living. New York, NY: Harper & Row.
Eibl-Ebesfeldt, I. (1970). Ethology: The biology of behavior. New York, NY: Holt, Reinhart & Winston.
Ekman, P. (Ed.). (1982). Emotion in the human face (2nd ed.). Cambridge, England: Cambridge University
Press.
Fiddick, L., Cosmides, L., & Tooby, J. (2000). No interpretation without representation: The role of domain-
specific representations and inferences in the Wason selection task. Cognition,77,1–79.
Fisher, R. A. (1930). The genetical theory of natural selection. Oxford, England: Clarendon Press.
Fodor, J. (1983). The modularity of mind. Cambridge, MA: MIT Press.
Fodor, J. (2000). The mind doesn’t work that way. Cambridge, MA: MIT Press.
Frege, G. (1879). Begriffsschrift (“Concept Notation”), eine der arithmetischen nachgebildete Formelsprache des
reinen Denkens. Halle a. S.
Friesen, C., & Kingstone, A. (2003). Abrupt onsets and gaze direction cues trigger independent reflexive
attentional effects. Cognition,87,B1–B10.
Gallistel, C. R. (2000). The replacement of general-purpose learning models with adaptively specialized
learning modules. In M. S. Gazzaniga (Ed.), The new cognitive neurosciences (pp. 1179–1191). Cambridge,
MA: MIT Press.
82 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 83
Gallistel, C. R., & Gibbon, J. (2000). Time, rate and conditioning. Psychological Review,107, 289–344.
Gallistel, C. R., Brown, A., Carey, S., Gelman, R., & Keil, F. (1991). Lessons from animal learning for the study
of cognitive development. In S. Carey& R. Gelman (Eds.), The epigenesis of mind (pp. 3–36). Hillsdale, NJ:
Erlbaum.
Gaulin, S. (1995). Does evolutionary theory predict sex differences in the brain? In M. S. Gazzaniga (Ed.), The
cognitive neurosciences (pp. 1211–1225). Cambridge, MA: MIT Press.
Gaulin, S., & Schlegel, A. (1980). Paternal confidence and paternal investment: A cross cultural test of a
sociobiological hypothesis. Ethology and Sociobiology,1, 301–309.
Geertz, C. (1973). The interpretation of cultures. New York, NY: Basic Books.
German, T. P., & Barrett, H. C. (2005). Functional fixedness in a technologically sparse culture. Psychological
Science,16(1), 1–5.
Gigerenzer, G. (1991). How to make cognitive illusions disappear: Beyond heuristics and biases. European
Review of Social Psychology,2,83–115.
Gigerenzer, G., & Murray, D. (1987). Cognition as intuitive statistics. Hillsdale, NJ: Erlbaum.
Gigerenzer, G., & Selten, R. (Eds.). (2002). Bounded rationality: The adaptive toolbox. Cambridge, MA: MIT
Press.
Gigerenzer, G., Todd, P., and the ABC Research Group . (1999). Simple heuristics that make us smart. New
York, NY: Oxford University Press.
Gould, S. J., & Lewontin, R. C. (1979). The Spandrels of San Marco and the Panglossian Paradigm: A critique
of the adaptationist programme. Proceedings of the Royal Society B: Biological Sciences,205, 581–598.
Gray, H. (1918). Gray’s anatomy, Twentieth Edition. W. Lewis (Ed.). Philadelphia, PA: Lea & Febiger.
Griskevicius, V., Delton, A. W., & Robertson, T. E., Tybur, J. M. (2011). Environmental contingency in life-
history strategies: Influence of mortality and socioeconomic status on reproductive timing. Journal of
Personality and Social Psychology,100, 241–254.
Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach to moral judgment.
Psychological Review,108(4), 814–834.
Hamilton, W. D. (1964). The genetical evolution of social behavior. I. II. Journal of Theoretical Biology,7,1–52.
Hammerstein, P., & Parker, G. A. (1982). The asymmetric war of attrition. Journal of Theoretical Biology,96(4),
647–682.
Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex
mind reading. Journal of Personality and Social Psychology,78,81–91.
Herrnstein, R. J. (1977). The evolution of behaviorism. American Psychologist,32, 593–603.
Hirschfeld, L. A., & Gelman, S. A. (Eds.). (1994). Mapping the mind: Domain specificity in cognition and culture.
Cambridge, England: Cambridge University Press.
Isaac, G. (1989). The archaeology of human origins: Papers by Glynn Isaac (Barbara Isaac, Ed.). Cambridge,
England: Cambridge University Press.
Jablonka, E., and Lamb, M. J. (2005). Evolution in four dimensions. Cambridge, MA: MIT Press.
Kahneman, D., Slovic, P., & Tversky, A. (Eds.). (1982). Judgment under uncertainty: Heuristics and biases.
Cambridge, England: Cambridge University Press.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica,47(2),
263.
Kaplan, H., & Hill, K. (1985). Food sharing among Ache Foragers: Tests of explanatory hypotheses. Current
Anthropology,26(2), 223–246.
Keil, F. (1989). Concepts, kinds, and cognitive development. Cambridge, MA: MIT Press.
Keil, F. C. (1994). The birth and nurturance of concepts by domains: The origins of concepts of living things.
In L. A. Hirschfeld& S. A. Gelman (Eds.), Mapping the mind: Domain specificity in cognition and culture.
Cambridge, England: Cambridge University Press.
Klein, S. (2005). The cognitive neuroscience of knowing one’s self. In M. S. Gazzaniga (Ed.), The cognitive
neurosciences, III (pp. 1077–1089). Cambridge, MA: MIT Press.
Klein, S., Cosmides, L., Tooby, J., & Chance, S. (2002). Decisions and the evolution of memory: Multiple
systems, multiple functions. Psychological Review,109, 306–329.
Klein, S., German, T., Cosmides, L., & Gabriel, R. (2004). A theory of autobiolographical memory: Necessary
components and disorders resulting from their loss. Social Cognition,22(5), 460–490.
Krasnow, M., Cosmides, L., Pedersen, E., & Tooby, J. (2012). What are punishment and reputation for? PLOS
One,7(9), e45662 (pp. 1–9).
Krasnow, M. M., Delton, A. W., Cosmides, L., & Tooby, J. (2013). Meeting now suggests we will meet again:
Implications for debates on the evolution of cooperation. Nature Scientific Reports,3, 1747. doi: 10.1038/
srep1747
The Theoretical Foundations of Evolutionary Psychology 83

3GC01 06/09/2015 12:40:44 Page 84
Krasnow, M. M., Delton, A. W., Cosmides, L., & Tooby, J. (2015). Group cooperation without group
selection: Modest punishment can recruit much cooperation. PLoS ONE,10(4), e0124561. doi: 10.1371/
journal.pone.0124561
Kurzban, R., Tooby, J., & Cosmides, L. (2001). Can race be erased?: Coalitional computation and social
categorization. Proceedings of the National Academy of Sciences, USA,98(26), 15387–15392.
LeDoux, J. (1995). In search of an emotional system in the brain: Leaping from fear to emotion to
consciousness. In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. 1049–1061). Cambridge, MA:
MIT Press.
Lee, R., & DeVore, I. (Eds.). (1968). Man the hunter. Chicago, IL: Aldine.
Lee, R., & DeVore, I. (Eds.). (1976). Kalahari hunter-gatherers: Studies of the !Kung San and their neighbors.
Cambridge, MA: Harvard University Press.
Lenneberg, E. (1967). Biological foundations of language. New York, NY: Wiley.
Leslie, A. (1987). Pretense and representation: The origins of “theory of mind.”Psychological Review,94,
412–426.
Leslie, A. M. (1994). ToMM, ToBy, and agency: Core architecture and domain specificity. In L. A.
Hirschfeld& S. A. Gelman (Eds.), Mapping the mind: Domain specificity in cognition and culture
(pp. 119–148). Cambridge, England: Cambridge University Press.
Leslie, A. M., & Thaiss, L. (1992). Domain specificity in conceptual development: Neuropsychological
evidence from autism. Cognition,43, 225–251.
Leslie, A. M., German, T. P., & Polizzi, P. (2005). Belief-desire reasoning as a process of selection. Cognitive
Psychology,50,45–85.
Li, F. F., Van Rullen, R., Koch, C., & Perona, P. (2002). Rapid natural scene categorization in the near absence
of attention. Proceedings of the National Academy of Sciences, USA,99, 9596–9601.
Lieberman, D., Tooby, J., & Cosmides, L. (2003). Does morality have a biological basis? An empirical test of
the factors governing moral sentiments relating to incest. Proceedings of the Royal Society B: Biological
Sciences,270(1517), 819–826.
Lieberman, D., Tooby, J., & Cosmides, L. (2007). The architecture of human kin detection. Nature,445(7129),
727–731. doi: 10.1038/nature05510
López, A., Atran, S., Coley, J., Medin, D., & Smith, E. (1997). The tree of life: Universals of folkbiological
taxonomies and inductions. Cognitive Psychology,32, 251–295.
Lutz, C. A. (1988). Unnatural emotions: Everyday sentiments on a Micronesian Atoll and their challenge to western
theory. Chicago, IL: University of Chicago Press.
Lukaszewski, A. W., & Roney, J. R. (2011). The origins of extraversion: Joint effects of facultative calibration
and genetic polymorphism. Personality and Social Psychology Bulletin,37(3), 409–421.
Mandler, J., & McDonough, L. (1998). Studies in inductive inference in infancy. Cognitive Psychology,37(1),
60–96.
Markman, E. (1989). Categorization and naming in children. Cambridge, MA: MIT Press.
Marks, I. (1987). Fears, phobias, and rituals. New York, NY: Oxford University Press.
Maynard Smith, J. (1982). Evolution and the theory of games. Cambridge, England: Cambridge University
Press.
Mineka, S., & Cook, M. (1993). Mechanisms involved in the observational conditioning of fear. Journal of
Experimental Psychology: General,122,23–38.
Mineka, S., Davidson, M., Cook, M., & Keir, R. (1984). Observational conditioning of snake fear in rhesus
monkeys. Journal of Abnormal Psychology,93, 355–372.
New, J., Cosmides, L., & Tooby, J. (2007). Category-specific attention to animals reflects ancestral priorities,
not expertise. Proceedings of the National Academy of Sciences, USA,104(42), 16598–16603. doi: 10.1073/
pnas.0703913104
Noë, R., & Hammerstein, P. (1994). Biological markets: supply and demand determine the effect of partner
choice in cooperation, mutualism and mating. Behavioral Ecology and Sociobiology,35(1) 1–11.
Öhman, A., & Mineka, S. (2001). Fear, phobias and preparedness: Toward an evolved module of fear and
fear learning. Psychological Review,108, 483–522.
Petersen, M. B., Sznycer, D., Sell, A., Cosmides, L., & Tooby, J. (2013). The ancestral logic of politics: Upper-
body strength regulates men’s assertion of self-interest over economic redistribution. Psychological
Science. doi: 10.1177/0956797612466415
Pietraszewski, D., Cosmides, L., & Tooby, J. (2014). The content of our cooperation, not the color of our skin:
Alliance detection regulates categorization by coalition and race, but not sex. PLoS ONE,9(2), e88534. doi:
10.1371/journal.pone.0088534
Pinker, S. (1994). The language instinct. New York, NY: Morrow.
84 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 85
Pinker, S. (1997). How the mind works. New York, NY: Norton.
Pinker, S., & Bloom, P. (1990). Natural language and natural selection. Behavioral and Brain Sciences,13(4),
707–784.
Pinker, S. (2002). The blank slate. New York, NY: Viking Press.
Pitman, R., & Orr, S. (1995). Psychophysiology of emotional and memory networks in posttraumatic stress
disorder. In J. McGaugh, N. Weinberger, & G. Lynch (Eds.), Brain and memory: Modulation and mediation of
neuroplasticity (pp. 75–83). New York, NY: Oxford University Press.
Posner, M. (1978). Chronometric explorations of mind. New York, NY: Oxford University Press.
Price, M. E., Cosmides, L., & Tooby, J. (2002). Punitive sentiment as an anti-free rider psychological device.
Evolution and Human Behavior,23, 203–231.
Profet, M. (1992). Pregnancy sickness as adaptation: A deterrent to maternal ingestion of teratogens. In J. H.
Barkow, L. Cosmides, & J. Tooby. The adapted mind: evolutionary psychology and the generation of culture
(pp. 327–366). New York, NY: Oxford University Press.
Rips, L. (1994). The psychology of proof. Cambridge, MA: MIT Press.
Ro, T., Russell, C., & Lavie, N. (2001). Changing faces: A detection advantage in the flicker paradigm.
Psychological Science,12(1), 94–99.
Rode, C., Cosmides, L., Hell, W., & Tooby, J. (1999). When and why do people avoid unknown probabilities
in decisions under uncertainty? Testing some predictions from optimal foraging theory. Cognition,72,
269–304.
Sahlins, M. (1976). The use and abuse of biology: An anthropological critique of sociobiology. Ann Arbor: University
of Michigan Press.
Schacter, D., & Tulving, E. (Eds.). (1994). Memory systems 1994. Cambridge, MA: MIT Press.
Schmitt, D. P. (2005). Sociosexuality from Argentina to Zimbabwe: A 48-nation study of sex, culture, and
strategies of human mating. Behavioral and Brain Sciences,28, 247–275.
Sell, A., Tooby, J., & Cosmides, L. (2009). Formidability and the logic of human anger. Proceedings of the
National Academy of Sciences, USA,106(35), 15073–15078. doi: 10.1073/pnas.0904312106
Sell, A., Cosmides, L., & Tooby, J. (2014). The human anger face evolved to enhance cues of strength.
Evolution & Human Behavior. doi: 10.1016/j.evolhumbehav.2014.05.008
Shannon, C. E. (1948). A mathematical theory of communication. Bell System Technical Journal,27, 379–423 &
623–656.
Shepard, R. N. (1984). Ecological constraints on internal representation: Resonant kinematics of perceiving,
imagining, thinking, and dreaming. Psychological Review,91, 417–447.
Shepard, R. N. (1987). Evolution of a mesh between principles of the mind and regularities of the world.
In J. Dupre (Ed.), The latest on the best: Essays on evolution and optimality (pp. 251–275). Cambridge, MA:
MIT Press.
Sherry, D., & Schacter, D. (1987). The evolution of multiple memory systems. Psychological Review,94,
439–454.
Shostak, M. (1981). Nisa: The life and words of a !Kung woman. Cambridge, MA: Harvard University Press.
Skinner, B. F. (1957). Verbal behavior. New York, NY: Appleton-Century-Crofts.
Spelke, E. S. (1990). Principles of object perception. Cognitive Science,14,29–56.
Sperber, D. (1994). The modularity of thought and the epidemiology of representations. In L. A. Hirschfeld&
S. A. Gelman (Eds.), Mapping the mind: Domain specificity in cognition and culture. Cambridge, England:
Cambridge University Press.
Sperber, D. (1996). Explaining culture: A naturalistic approach. Oxford, England: Blackwell.
Sperber, D., & Wilson, D. (1995). Relevance: Communication and cognition (2nd ed.). Oxford, England:
Blackwell.
Springer, K. (1992). Children’s awareness of the implications of biological kinship. Child Development,63,
950–959.
Steen, F., & Owens, S. (2001). Evolution’s pedagogy: An adaptationist model of pretense and entertainment.
Journal of Cognition and Culture,1(4), 289–321.
Suarez, S. D., & Gallup, G. G. (1979). Tonic immobility as a response to rage in humans: A theoretical note.
Psychological Record,29, 315–320.
Symons, D. (1978). Play and aggression: A study of rhesus monkeys. New York, NY: Columbia University
Press.
Symons, D. (1979). The evolution of human sexuality. New York, NY: Oxford University Press.
Symons, D. (1987). If we’re all Darwinians, what’s the fuss about. In C. B. Crawford, M. F. Smith, & D. L.
Krebs (Eds.), Sociobiology and psychology (pp. 121–146). Hillsdale, NJ: Erlbaum.
Symons, D. (1989). A critique of Darwinian anthropology. Ethology and Sociobiology,10, 131–144.
The Theoretical Foundations of Evolutionary Psychology 85

3GC01 06/09/2015 12:40:44 Page 86
Symons, D. (1992). On the use and misuse of Darwinism in the study of human behavior. In J. Barkow,
L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture
(pp. 137–159). New York, NY: Oxford University Press.
Sznycer, D., Takemura, K., Delton, A. W., Sato, K., Robertson, T., Cosmides, L., & Tooby, J. (2012). Cross-
cultural differences and similarities in proneness to shame: An adaptationist and ecological approach.
Evolutionary Psychology,10(2), 352–370.
Tinbergen, Niko. (1963). On aims and methods of ethology. Zeitschrift für Tierpsychologie,20, 410–433.
Thornhill, R. (1997). The concept of an evolved adaptation. In G. Bock& G. Cardew (Eds.), Characterizing
human psychological adaptations (pp. 4–13). London, England: CIBA Foundation.
Tomaka, J., Blascovich, J., Kibler, J., & Ernst, J. (1997). Cognitive and physiological antecedents of threat and
challenge appraisal. Journal of Personality and Social Psychology,73,63–72.
Tooby, J. (1976). The evolution by natural selection of systems of adaptive multigenerational nongenetic
inheritance for parameterizing development: Not your father’s Lamarck. Institute for Evolutionary Studies
Technical Report 76(1)
Tooby J. (1982). Pathogens, polymorphism, and the evolution of sex. Journal of Theoretical Biology,97, 557–576.
Tooby, J. (1985). The emergence of evolutionary psychology. In D. Pines (Ed.), Emerging syntheses in science
(pp. 124–137). Santa Fe, NM: The Santa Fe Institute.
Tooby, J. (2014). Learning and culture: Scientific ideas ripe for retirement. Contribution to Edge Annual
Question: What scientific ideas are ready for retirement? Retrieved from https://edge.org/response-
detail/25343
Tooby, J., & Cosmides, L. (1990a). The past explains the present: Emotional adaptations and the structure of
ancestral environments. Ethology and Sociobiology,11, 375–424.
Tooby, J., & Cosmides, L. (1990b). On the universality of human nature and the uniqueness of the individual:
The role of genetics and adaptation. Journal of Personality,58,17–67.
Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. Barkow, L. Cosmides,
& J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New
York, NY: Oxford University Press.
Tooby, J., & Cosmides, L. (1996). Friendship and the banker’s paradox: Other pathways to the evolution of
adaptations for altruism. Proceedings of the British Academy,88, 119–143.
Tooby, J., & Cosmides, L. (2001). Does beauty build adapted minds? Toward an evolutionary theory of
aesthetics, fiction and the arts. SubStance,94/95(1), 6–27.
Tooby, J., & Cosmides, L. (2003). Elements of a Darwinian theory of Lamarckian inheritance. Paper presented at
the annual meeting of the Human Behavior and Evolution Society, Lincoln, NE.
Tooby, J., & Cosmides, L. (2008). The evolutionary psychology of the emotions and their relationship to
internal regulatory variables. In M. Lewis, J. M. Haviland-Jones, & L. Feldman Barrett (Eds.), Handbook of
emotions (3rd ed.). New York, NY: Guilford Press.
Tooby, J., & Cosmides, L. (2010). Groups in mind: Coalitional psychology and the roots of war and morality.
In H. Høgh-Olesen (Ed.), Human morality and sociality: Evolutionary and comparative perspectives
(pp. 191–234). New York: Palgrave Macmillan.
Tooby, J., & Cosmides, L. (in press). Ecological rationality in a multimodular mind. In Evolutionary
psychology: Foundational papers. Cambridge, MA: MIT Press.
Tooby, J., Cosmides, L., & Barrett, H. C. (2003). The second law of thermodynamics is the first law of
psychology: Evolutionary developmental psychology and the theory of tandem, coordinated inheri-
tances. Psychological Bulletin,129(6), 858–865.
Tooby, J., Cosmides, L., & Barrett, H. C. (2005). Resolving the debate on innate ideas: Learnability constraints
and the evolved interpenetration of motivational and conceptual functions. In P. Carruthers, S.
Laurence, & S. Stich (Eds.), The innate mind: Structure and content. New York, NY: Oxford University
Press.
Tooby, J., Cosmides, L., Sell, A., Lieberman, D., & Sznycer, D. (2008). Internal regulatory variables and the
design of human motivation: A computational and evolutionary approach. In Andrew J. Elliot (Ed.)
Handbook of approach and avoidance motivation (pp. 251–271). Mahwah, NJ: Erlbaum.
Tooby, J., Cosmides, L., & Price, M. (2006). Cognitive adaptations for n-person exchange: The evolutionary
roots of organizational behavior. Managerial and Decision Economics,27, 103–129. doi: 10.1002/mde.1287
Tooby J., & DeVore, I. (1987). The reconstruction of hominid behavioral evolution through strategic
modeling. In W. Kinzey (Ed.), Primate models of hominid behavior (pp. 183–237). New York, NY:
SUNY Press.
Trivers, R. L. (1971). The evolution of reciprocal altruism. Quarterly Review of Biology,46,35–57.
86 FOUNDATIONS OF EVOLUTIONARY PSYCHOLOGY

3GC01 06/09/2015 12:40:44 Page 87
Triesman, A. (2005). Psychological issues in selective attention. In M. S. Gazzaniga (Ed.), The cognitive
neurosciences III (pp. 529–544). Cambridge, MA: MIT Press.
Vining, D.R. (1986). Social versus reproductive success: The central theoretical problem of human socio-
biology. Behavioral and Brain Sciences,9, 167–216.
von Neumann, J. (1945). First Draft of a Report on the EDVAC. Contract between the United States Army
Ordnance Department and the University of Pennsylvania.
von Neumann, J., & Morgenstern, O. (1944). Theory of games and economic behavior. Princeton, NJ: Princeton
University Press.
Walker, R., Hill, K., Kaplan, H., & McMillan, G. (2002). Age dependency of hunting ability among the Ache
of eastern Paraguay. Journal of Human Evolution,42, 639–657.
Wang, X. T. (2002). Risk as reproductive variance. Evolution and Human Behavior,23,35–57.
Weiner, N. (1948). Cybernetics or control and communication in the animal and the machine. Cambridge, MA: MIT
Press.
Williams, G. C. (1966). Adaptation and natural selection: A critique of some current evolutionary thought.
Princeton, NJ: Princeton University Press.
Williams, G.C., & Williams, D.C. (1957). Natural selection of individually harmful social adaptations among
sibs with special reference to social incests. Evolution,11,32–39.
Wilson, E.O. (1975). Sociobiology: The new synthesis. Cambridge, MA: Belknap Press.
Wynn, K. (1998). Psychological foundations of number: Numerical competence in human infants. Trends in
Cognitive Sciences,2, 296–303.
Yerkes, R. M., & Yerkes, A. W. (1936). Nature and conditions of avoidance (fear) response in chimpanzee.
Journal of Comparative Psychology,21,53–66.
The Theoretical Foundations of Evolutionary Psychology 87