![Sample item from the Paper-Folding Test. [Color figure can be seen in the online version of this article, available at wileyonlinelibrary.com]](https://www.researchgate.net/profile/Richard-Lamb-7/publication/271335894/figure/fig4/AS:272620092391425@1442009103145/Sample-item-from-the-Paper-Folding-Test-Color-figure-can-be-seen-in-the-online-version_Q320.jpg)
![Sample item from the Test of Mental Rotation. [Color figure can be seen in the online version of this article, available at wileyonlinelibrary.com]](https://www.researchgate.net/profile/Richard-Lamb-7/publication/271335894/figure/fig6/AS:272640967442436@1442014080051/Sample-item-from-the-Test-of-Mental-Rotation-Color-figure-can-be-seen-in-the-online_Q320.jpg)
Content uploaded by Richard L. Lamb
Author content
All content in this area was uploaded by Richard L. Lamb on Nov 20, 2017
Content may be subject to copyright.
JOURNAL OF RESEARCH IN SCIENCE TEACHING VOL. 52, NO. 3, PP. 410–437 (2015)
Research Article
Development of a Cognition-Priming Model Describing Learning in a
STEM Classroom
Richard Lamb, Tariq Akmal, and Kaylan Petrie
Department of Teaching and Learning, Washington State University, Pullman 99164, Washington
Received 21 September 2013; Accepted 1 December 2014
Abstract: Successful STEM learning depends on the interaction of affect, cognition, and application of
ideas. Simply put students who are unwilling to persist in STEM based endeavors do not suddenly develop
into scientists, mathematicians, engineers or computer scientists, nor do they seek out STEM related courses
or STEM based careers. The purpose of this study is to investigate content, cognitive, and affective outcomes
related to STEM integrated curriculum within the K-5 arena. Educational and psychological literature tends to
focus one aspect of the other when examining the role of affect and cognition on student outcomes. Current
trends in educational measurement and psychometrics have begun to address the artificial disconnect that
exists between affect, cognition, and content outcomes within the science education literature. The methods
used to develop the results within this study are a mixture of quantitative methods to develop a model of
learning occurring in a STEM school. Using ANOVA, structural equation modeling, and model analysis, an
understanding of the problems presented becomes clear. Analysis of model fit statistics suggests adequate
modelfit(x
2
(21) ¼30.91, p¼0.075, CFI ¼0.94, TLI ¼0.93, RMSEA ¼0.04, SRMR ¼0.05). The standard-
ized structural coefficients for the path from group to each of the constructs is statistically significant
(p<0.05) thus indicating that the two groups differ on the constructs of self-efficacy, science interest, spatial
visualization, and mental rotation. An estimate of effect size of the mean group difference across the
statistically significant constructs reveals self-efficacy (d¼1.27, large), science interest (d¼1.97, large),
spatial visualization (d¼1.30, large), and mental rotation (d¼1.42, large). There is considerable evidence
that the inclusion, STEM integrated learning at the earlier elementary level becomes critically important for
the students as they progress in school. #2015 Wiley Periodicals, Inc. J Res Sci Teach 52:410–437, 2015
Keywords: STEM education; elementary science; learning
Successful STEM (Science, Technology, Engineering, and/or Mathematics) learning
depends on the interaction of affect, cognition, and application of ideas by students in the
classroom (Dreyfus, Jungwirth & Eliovitch, 1990; Houseal, Abd-El-Khalick, & Destefano, 2014;
Oppezzo & Schwartz, 2013; Wright & Stone, 2004). Learning in general is defined as the
acquisition of knowledge and skills through experience and study (Drake, Land, & Tyminski,
2014; Rahm, 2014). While there is a clear definition of learning, the definition of STEM learning is
not as clear. One critical consideration when defining STEM learning is that STEM learning is a
broad area encompassing many disciplines and epistemological practices. As such, the definition
of STEM learning is not yet well developed and open to debate. The authors of this study
operationalize the definition of STEM learning by extending the definition of learning. STEM
learning is the acquisition of knowledge and skills through experience and study integrated though
Correspondence to: Richard Lamb; E-mail: richard.lamb@wsu.edu
DOI 10.1002/tea.21200
Published online 24 January 2015 in Wiley Online Library (wileyonlinelibrary.com).
#2015 Wiley Periodicals, Inc.
multiple lens allowing for the appreciation of the encompassing complexity and cross-cutting
ideas across the STEM disciplines as a whole. A critical differentiation of STEM learning from
other forms of learning (such as science learning) is the recognition of the interdependence and
integration of these multiple disciplines and epistemological practices. In this way, STEM
requires less specialization and more ability to see across areas of interaction and the resultant
complexity within the STEM disciplines. In particular, as the educational requirements of students
required greater and greater specialization those students who are more able to identify and
integrate disparate ideas will be in greater demand (van Eijck & Roth, 2011). Herein lies the
problem for researchers. How does one measure such complexity and cross-cutting ideas?
Measures such as those used in this study by necessity must consist of a battery of measures—
ideally qualitative and quantitative—both specific and general in nature in an attempt to address as
many facets of the STEM learning construct as possible. However, a gap still exists in that a
succinct and unified definition of STEM learning has yet to be fully developed.
Students’ aptitude with STEM concepts is in part attitudinal in nature; researchers
conceptualize attitudes within the larger educational and psychological literature as playing a
critical role in students’ success (Kimmons, Liu, Kang, & Santana, 2011; Sawtelle, Brewe, &
Kramer, 2012). Simply put, students who are unwilling to persist in STEM based classes and
learning do not suddenly develop into scientists, mathematicians, engineers, or computer
scientists, nor do they seek out STEM related courses and careers later in life. More importantly,
these students fail to become effective consumers of information and knowledge related to STEM
(McDonald, 2010). In this light, the interactions of the STEM learning facets of affect, cognition,
and application touch on each of the major perspectives within educational psychology as related
to learning. Educational psychologists generalize learning theory as the interaction between
behaviorist, cognitive, and social frameworks (Johri & Olds, 2011; van Gog & Rummel, 2010). It
is within the combination of these three frameworks that we hope to demonstrate an underst anding
of student learning, which extends beyond a typical one-dimensional approach looking simply at
affect, outcome, or cognition alone.
Purpose, Research Questions, and Hypothesis
The purpose of this study is to investigate content, cognitive, and affective outcomes related
to STEM integrated curriculum within the K-5 classroom. Exploring this relationship between
each of the facets of STEM is important to STEM educators for several reasons. First, an
understanding of student attitudes related to STEM is of vital importance when creating
meaningful dialogue between agents in the classrooms and schools and for developing future
STEM arena participants. This is not to suggest that self-reporting measures provide all of the
information necessary to understand the affect facet of attitude, but this approach will garner
information necessary for follow-up investigations. Second, exploring the interaction of student
self-referent beliefs and outcomes related to content and cognition yields important information
and insights guiding teacher instructional practices and curriculum development. Lastly,
understanding this interaction provides insight into areas of potential development when
cultivating student beliefs and attitudes. The authors suggest that there are significant gains within
the identified areas (cognition and affect) resulting from exposure to STEM integrated curriculum.
The research questions addressed in this study are as follows:
1 What is the effect of an integrated STEM curriculum on student affect, specifically Self-
Efficacy and Interest related to science and technology?
2 Does exposure to a STEM integrated curriculum generate change related to student
cognition related to Mental Rotation and Spatial Visualization?
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 411
3 What is the relationship between affect, cognition, and science content score outcomes?
Review of Literature
Two Learning Models
Educational and psychological literature tends to focus on examining the role of affect and
cognition as separate entities on student outcomes related to content within STEM areas (Pellegrino,
2012; Swarat, Ortony, & Revelle, 2012). Current trends in educational measurement and
psychometrics have begun to address the artificial disconnect that exists between affect, cognition,
and content outcomes within the science education literature (Klassen, 2006; Piety, 2013). Two
potential models that provide for the integration of these two areas of interest as a learning model are
the cognition-priming model of learning and the affect-priming model of learning.Thesemodelsare
of importance because they link attitudes in science, cognition in science, and orientation toward
science (Bohner & Dickel, 2011; Vedder-Weiss & Fortus, 2011). Reserachers differentiate the two
models though examination of the manner in which the first response to stimulus occurs.
Cognition priming occurs when an external stimulus triggers examination of the stimulus
using cognitive constructs known as cognitive attributes prior to the inclusion of affective
constructs (Lamb, Annetta, Vallett, & Sadler, 2014b). For example, a teacher leading an inquiry
investigation in which students must reason through observations activating cognitive attributes.
On the other hand, affective priming occurs when the stimulus elicits an affective response such as
anger or disengagement. This is not to say that the two priming systems are mutually exclusive just
that one activated before the other. In this understanding of activation order cognitive first then
affective, or affective first then cognitive, the biggest difference is the order in which affect and
cognition occur in response to the stimulus, in this case STEM content. This indicates the
importance of the recognition that the development of STEM competencies is not only ability
dependent but involves other aspects of the student psychological makeup (Osborne, 2010; Rivet
& Kastens, 2012). Berkowitz (1993) introduced the cognition and affect intersection as a learning
model. However, the logic and underlying theoretical components of Berkowtiz’ model were
developed from previous models focusing on automatic affect in learning by Epstein (1993).
At this point, it becomes important to distinguish between automatic associative affect (such
as self-efficacy or interest) and spontaneous affect (such as rage or surprise). Automatic
associative affect results from repeated contact with contexts not consciously under the
individual’s control, thus associating the affect with the context (Bandura, 1977a; Bleasdale,
1987; Fiske, 2013). For example, if a student has a negative interest in science due to repeated
failure, this negative interest is automatic associative affect. By contrast, spontaneous affect such
as rage is usually transient and general, meaning the response is not isolated to one response
context (Somerville, Wagner, Wig, Moran, & Whalen, 2013). These forms of affective
reactions (spontaneous affect) occur relatively quickly and give rise to low-order cognition;
because of the low order cognition, the resultant behaviors will be more simplistic (Cox & Klinger,
2011). For example, a person may exhibit happiness when presented with a gift, or when one has,
her/his needs met. Within the learning model described as the affective-priming model originally
developed by Berkowitz (1993), affective effects take place prior to cognitive processes. Thus
within this model, affect is precognitive (Bem, 2011). These low-level affective activations act as
primers for higher-level affective activations and arousal of cognition but do not always lead to
learning (Lowe & Ziemke, 2011).
A second mechanism of action related to learning and activation of affect and cognition is the
method in which higher-level cognition acts as a primer generating arousal related to affect.
Journal of Research in Science Teaching
412 LAMB, AKMAL, AND PETRIE
Within the context of this study, this model is referred to as the cognition-priming model and is of
interest to education in general but science education specifically as this model provides a window
into interactions seen within the science classroom (Paas & Sweller, 2012). Within the cognition-
priming model, the science content acts as the externalization of the cognitive process manifested
as behaviors such as responses on an assessment (Bogg & Finn, 2010; Zeineddin & Abd-El-
Khalick, 2010). Externalization of the content becomes important, as affect requires external
stimuli to trigger arousal of persistent affect as opposed to spontaneous affect. One additional
element that is missing within the cognition-priming model is the role of memory (Hardt,
Einarsson, & Nader, 2010). The cognition-priming model would arise within a controlled manner
and requires relatively more time, additional factors such as memory and precursors to inhibitory
behaviors such as self-regulation can insert within the process. Thus, there is a role for and ability
to recall previous experiences (memory) within the cognition-priming model due to reduced
allocation of processing resources (cognitive load) (Blanchette & Richards, 2010). The model
discussed above is applicable to multiple subject areas and domains. Given the multiple facets
associated with STEM learning and the intent to integrate the varied domains associated with
STEM learning, a flexible model of learning should be employed; thus the description of the
generalized learning model (such as the cognition priming model) is most appropriate for this
process (Cowell, Bussey, & Saksida, 2012). One manner in which researchers can identify which
model is active within the context of the STEM classroom is through examination of affective and
cognitive constructs such self-efficacy, interest, mental rotation, and spatial visualization.
Self-Efficacy
Psychologists see self-efficacy and interest governed under an area of psychology known as
self-referent thought (Kappes, Oettingen, & Pak, 2012). Psychometricians and measurement
experts suggest measurement of self-referent constructs occur most readily via self-reporting
measures as the constructs are internal and latent (Bandura, 1982; Berry, West, & Dennehey, 1989,
Lamb, Vallett, & Annetta, 2014). Self-efficacy related to addressing one’s environment is not fixed
but is semi-stable, nor is self-efficacy a matter of knowledge related to the environment (Bandura,
1982). The construct of self-efficacy is componential consisting of cognitive, social, and
behavioral skills integrated into coherent psychological heuristics for reactive approaches. In
addition to reactive approaches, people when confronted wit h novel problems use the components
of self-efficacy for more intensive exploratory behavior when heuristics fail (Bandura, 1982). The
four general components affecting self-efficacy are personal belief, verbal messages and social
encouragement, mastery experiences, and peer success (Britner & Pajares, 2006; Zeldin &
Pajares, 2000). These antecedents to behavior allow self-efficacy to develop one’s agency within
the context of action related to a specific domain. Self-efficacy theory suggests that differing
modes of influence change behavior through creation and strengthening self-precepts related to
efficacy (Bandura, 1977b). Specifically, one’s perceived self-efficacy concerns the ability of the
student to judge his/her capacity to complete an activity or more appropriately a task (Bandura,
1993). Within the STEM arena, this would correspond to perceptions and judgments of capacity
related to science, technology, engineering, and mathematics tasks and more importantly skill
development. This judgment of capacity creates influence in student choice of activities, efforts,
and persistence when meeting resistance (Annetta, Frazier, Folta, Holmes, &Lamb, 2013). It
is these aspects of self-efficacy (motivation) which become concerning when analyzing and
developing achievement behavior within elementary level children. Elementary children
exhibiting strong expressions of self-efficacy in given subjects would exhibit strong achievement,
subject selection in later years, and persistence within the domain of affect (Adedokun,
Bessenbaccher, Parkerm, Kirkham, & Burgess, 2013; Noguera, 2011). In contrast to students with
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 413
strong exhibition of self-efficacy, those students with low levels of self-efficacy would
demonstrate avoidance behavior related to the subject, apathy, and would not persevere when
presented with challenges (Zimmerman, 2011). Within this context, self-efficacy also is predictive
of levels of behavioral change, but also accounts for the variation in coping behaviors found within
children receiving the same treatment (Zimmerman & Schunk, 2012).
Self-Efficacy in STEM. Using Bandura’s (1994) sources of self-efficacy to examine the latent
factors related to self-efficacy, researchers developed a four-factor model of self-efficacy (Lent,
Lopez, Brown & Gore, 1996; Lent, Miller, Smith, Watford, & Lim, 2013). Results of the analysis
demonstrated that mastery experiences account for the greatest variance within self-efficacy
models related to mathematics, science, and engineering (Lopez & Lent, 1992; Lamb, Annetta,
Meldrum & Vallett, 2012). Other studies controlling for mastery experiences illustrate that the
remaining four factors did not significantly contribute to the variance accounting (Goddard, 2001;
Halbesleben, Wheeler, & Paustian-Underdahlm, 2013; Matsui, Matsui, & Ohnishi, 1990). These
results support Bandura’s position that mastery experiences provide the most explanatory value
for levels of self-efficacy (Zeldin & Pajares, 2000). Thus, it follows that mastery experiences
within STEM related fields play a large role in one’s level of self-efficacy (Vedder-Weiss & Fortus,
2013). Studies of student levels of self-efficacy and related constructs provide an illuminating
view of student self-efficacy development. Elementary students initially exhibit relatively high
levels of self-efficacy relating to STEM areas (Wang, 2013). Research evidence is indicative of a
continuous decline in self-efficacy related to STEM content areas culminating in lack of
participation and selection of STEM courses within secondary and postsecondary education along
with career choice (Anderman & Young, 1994; Bandura, 1997; Hackett, 1995; Klassen & Usher,
2010; Lent & Hacket, 1987). This decline in late elementary and early middle school self-efficacy
develops from the creation of a task mastery focused learning-environments coupled with
increasingly complex cognitive requirements for successful task completion (English &
Kitsantas, 2013). Lamb et al. (2014b) using a novel cognitive diagnostic modeling technique via
real-time STEM Serious Educational Game play, quantified the contribution of these cognitive
attributes to complex science task completion. A second related area that bears some discussion is
that of interest.
Interest
Interest, like self-efficacy, is also a self-referent, semi-fluid construct relating to psychologi-
cal functioning. Interest is the subject’s existing outlook and arousal triggers, which create
attendance to individually specific stimuli from objects or events (Hidi, 2006; Krapp, 2007; Lamb
et al., 2012; Patrick, Mantzicopiulos, & Samarapungavan, 2009). This psychological state is
associated with positive affect and persistent arousal. There are many parallels between interest
and self-efficacy as the two are interrelated under the larger more generalized self-referent
identified as motivation (van Dinther, Fochy,& Segers, 2011). Interest often acts as the antecedent
to behavior, providing a development of agency within the individual to act on the external stimuli
and encode the stimuli for further processing via cognitive attribute patterns (Decety, Michalska &
Kinzler,2012; Scherer, 1988; Seel & Winn, 2012). Within the STEM context, it is the behavior of
encoding for further cognitive processing which creates the desire for cognitive interaction with
content thus generating behavior. Interest itself is composed of multiple facets or specific unique
aspects of the affective trait (Lamb et al., 2012). The three facets composing the generalized
affective trait of interest are individual interest, situational interest, and topical interest. Situational
interest is similar to individual interest with the exception that the stimuli are environmental
inducted (Fredricks, Alfeld, & Eccles, 2010; Huang, 2006). More specifically, these
Journal of Research in Science Teaching
414 LAMB, AKMAL, AND PETRIE
environmentally induced stimuli are triggers linked to tasks such organization, presentation and
thematic displays (Edgar & Fox, 2006). The final area is topical interest and is a combination of
individual and situational interest as well as the arousal response to specific topics presented to the
individual (Schiefele, 1998). The capacity to initiate and maintain persistent arousal provides
the influence in student choice of activities, effort, and, to some degree, persistence. However, as
opposed to self-efficacy, interest does not manifest as enduring deep arousal present over time as
this corresponds to intrinsic motivation and not interest (Harackiewicz, Durik, & Barron, 2005).
Interest and STEM. Using the construct of interest articulated by Lamb et al. (2012), an
examination of the types of interest related to education suggests a large number of extrinsic
factors such as prior experience, family, community, and peer levels of interest accounting for the
largest sources of variation within interest related to STEM. Of the three sources of extrinsic
influence on interest, that of family on student STEM interest is most powerful (Azevedo, 2011;
Schoon, Ross, & Martin, 2007; Smetana, Campione-Barr, Metzger, 2006). Thus, parental
involvement in their children’s STEM education is proportional or related to student performance
in STEM environments (Archer, Dewitt, & Willis, 2014). Further, as the child matures from
Kindergarten through middle school, the influence of the family wanes and the influence of peers
waxes reaching a peak within mid-middle school to early high school (Gottfried, 2010; Johnston
& Viadero, 2000). Thus, peer group interest plays a large role in the individual student’s selection
and attention to STEM content. While affective interactions offer one view of the manner that
students interact with STEM content and classes, a second and equally important context to
explore STEM learning is through changes in cognitive attributes such as spatial visualization and
mental rotation.
Spatial Visualization
Spatial visualization is a more general construct composed of Spatial Relation and
Orientations (SR-O), Visualization (V), and Kinesthetic Imagery (K) (Michael, Guilford,
Fruchter, & Zimmerman, 1957). Spatial visualization is more substantively defined from a
cognitive perspective as a cognitive attribute (visuospatial) and subset of latent attributes related
to movement and positioning of stimulus within three-dimensional space (Lamb et al., 2014b;
Spence & Feng, 2010). One suggested difference between abilities of spatial relations, orientation,
and visualization is that spatial relations and orientations require an origin point and visualization
only requires the detachment of the observer from the stimulus to manipulate the components.
More to this point, visualization occurs within more complex stimulus patterns providing details
about the pattern to the viewer (Yassam, Mattfeld, Stark, & Stark, 2011). Essentially, between the
three factors, one is able to reference the place of the object in three-dimensional space, internalize
details for processing, and apply motion to the static stimuli. Researchers also describe spatial
visualization as the ability to manipulate, rotate, and twist presented visual stimuli from
the environment, though this would seem to be mental rotation a separate but related cognitive
attribute (Baki, Kosa, & Guven, 2011). Spatial reasoning is thought to develop during middle
childhood, which begins approximately between ages 5 and 7, and is usually fully developed by
the end of puberty (Piaget & Inhelder, 1963). When researching spatial visualization researchers
used two approaches. The first approach focuses on the cognitive processing need to mentally
manipulate and solve spatial tasks (Pezzulo, Barsalou, Cangelosi, Fisher, & McRae, 2012). The
second approach is predicated upon complicated multistep analytical processing to conceptualize
the presented stimuli as alternate rotations. While the two approaches provide a means to
conceptualize images, within each approach the cognitive attributes are not locally independent of
each other. It is the latter of the two approaches that is of interest to educational researchers.
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 415
Spatial Visualization and STEM. STEM fields are among the most visually intensive fields
(Cox, 1928; Feldon, Timmerman, Stowe, & Showman, 2010). Original studies examining spatial
visualization as a construct have their roots within the study of mechanical aptitude and educators
often overlook this area as a predictor of success in STEM (Bodzin, 2011; Mathewson, 1999). The
visual intensity of STEM fields arises from representations and models developed to describe
natural phenomena. Correlational and more recently predictive studies have solidly demonstrated
the relationship between STEM achievement and spatial visualization (Lubinski, 2010; Pruden,
Levine, & Huttenlocher, 2011; Wu & Shah, 2004). The increase in achievement may result from
undergoing cognitive reformulation and restructuring using the spatial domain, specifically
mental manipulation of quantities and numbers are a large component of all four STEM domains.
Neuroimaging studies illustrate activation within brain areas related to spatial tasks when
engaging in mathematics tasks, engineering tasks, and science tasks (Gonzalez-Castillo, Saad,
Handwerker, Inati, & Brenowitz, 2012). Thus, increased processing in the spatial visualization
locations within the brain allow for reduced cognitive load during processing through the addition
of cognitive channels to process data streams (Casey, Nuttall, Pezaris, & Benbow, 1995;
Konstantinou, Bahrami, Rees, & Lavie, 2012).
Mental Rotation
Cognitive processing of shapes consists of using fundamental representations, each
representation is developed from a small number of viewpoints transformed to appropriate size,
orientation, and location of the nearest similar shape held in stored memory. This allows one to
conceptualize the same object in different rotations (Hinkel, 2012). Mental rotation is the
incremental analogue transformation process undertaken when attempting to recognize three-
dimensional objects in space (Kirsh, 2009), and develops earlier than spatial visualization (Harris,
Newcombe, & Hirsh-Pasek, 2013). Mamor (1975) demonstrated that children starting at
approximately 4 years of age (and as young as 3) use mental rotation to solve tasks in their
environments. One should note that mental rotation is a component of spatial ability and relates to
spatial visualization through is not interchangeable with spatial visualization. When researching
mental rotation researchers use either a two-dimensional or three-dimensional stimulus referents.
The test then presents a rotated referent at a specific angle and distractors with similar visual
properties (also rotated). The subject then must select the correct object using the referent as a non-
rotated model. Generally, as the angle of rotation increases the amount of time required to identify
the object increased.
Mental Rotation and STEM. Much of the reasoning found in STEM fields concerns objects
existing within three-dimensional space (Fosnot, 1993). Mental rotation and spatial reasoning
are both contained within a larger cognitive construct of visual processing; the difference
between the two from a practical aspect is subtle. The connection of the two constructs to
the STEM contexts is not similar due to differences in brain activations during cognitive attribute
use. The divergence between the two attributes (mental rotation and spatial visualization) occurs
as a function of the purpose of the task. Mental rotation is required to understand object
similarities, which exist within three-dimensional space when rotated; otherwise, we would not
know a chair is a chair when it was turned. This additional application of processing power using
mental rotation opens new stimulus channels allowing additional data streams to process in
parallel allowing for reduction i n cognitive load when processing com ponents that require spatial
visualization (Ifenthaler & Eseryel, 2013). In similar contexts, spatial visualization and mental
rotation are predictive of science achievement when participants engage in visual tasks (Stieff,
2007).
Journal of Research in Science Teaching
416 LAMB, AKMAL, AND PETRIE
Attitudes
Attitudes are composed of three parts, affect, cognition, and behavior (Breckler, 1984;
Bouckenooghe, 2010). An example of thiswould be self-efficacy and interest (affective) interacting
with mental rotation and spatial visualization, interacting with the content outcomes (behavior) as
proposed in this study. The affect-behavior-cognition distinction better known as feeling, acting,
and knowing are the three major components (or facets) of human experience (Breckler, 1984;
McGuire, 1969; Revelle, Wilt, & Rosenthal, 2010). Within science education typical approaches to
access these components are surveys, interviews, and self-reporting for affective measures;
behavior measurement occurs through observation and outcomes, and cognition is measured
through task completion and talk-aloud protocols (Fowler, 2009). Despite the tripartite model of
learning’s acceptancewithin psychology and educational psychology, educational research related
to the interplay and role of these facets is minimal.Much of the literature within education related to
attitudinal research focuses on affect, thus missing two important components and resulting in lack
of clarity and significant development of theories of learning from a dynamical perspective
(Richetin, Connor, & Perugini, 2011). For science-education researchers, it is critical we
understand the role of all three facets within the classroom to effect change related to STEM
education and student-future careerselection (Henderson, Beach & Finkelstein, 2011).
Methods
The methods used in this study are a mixture of quantitative techniques used to reinforce and
triangulate one other. Results of each analysis inform later analysis and the results of subsequent
sections. ANOVA, structural equation modeling, and model analysis were all implemented in this
study to provide resolution on student learning.
Sample
The subjects selected for this study were students (n¼254) from grades, K, 2, and 5 in two
schools within the Mid-Atlantic region of the United States. Student ages range from 5- to 12-years
old. Grade selection allows for an analysis of subjects as they progressed through the STEM
program at the intervention school. Of the 254 students within the study, 143 students were a part of
the comparison group and 111 students were a part of the intervention group. The schools are
approximately 0.7miles apart from one another; the students often interacted with each other
within the context of their neighborhood play.
A summary of the intervention school statistics regarding Annual Yearly Progress (AYP) are
illustrative of the deep social and educational difficulties experienced by the subjects. The
intervention school’s AYP scores show low teens or single digit achievement in both reading and
mathematics on end of year tests. As the school has matured the enactment of the Improving
America’s Schools Act and the No Child Left Behind (NCLB) Act, has created a precipitous drop
in enrollment with students leaving the neighborhood school and attending charter schools in
other areas of the city; the parents of the students see the charter schools as an escape for the
students. The population as discussed in this study acts as a microcosm of the wider quadrant of the
city. Demographically, the student population is 100% African American, 99% free lunch and
approximately 16% have special needs. The control school statistics regarding AYP are strikingly
similar to the intervention school and illustrate mid-teens level of achievement in both reading and
mathematics on district and state standardized tests. The control school’s student population
demographics and enrollment issues mirror that of the intervention school. The similarity of the
two schools is further magnified as the school district does not offer public transportation to and
from the school and students are expected to walk to the school from home.
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 417
Design
The study design is a non-randomized, pretest–posttest, comparison intact group design. This
design has a primary benefit of allowing research within the natural ecology of the classroom with
minimal invasiveness. While the author of the study did not select participants randomly from the
target population, the lack of disruption to teacher planning and instruction is of benefit to the
study allowing for more natural and candid responses on measures. The treatment condition
consists of a whole school STEM program in which the teacher explicitly inserted STEM content
into and across the curriculum. By contrast, the comparison school did not explicitly use or apply
STEM curriculum nor was the use of STEM a priority. This school instead focused on English
Language Arts.
The pretest–posttest comparison design allows for examination of the treatment effects from
differing perspectives. The use of this design does have some threats to internal validity, the first
being carry-over during measure administration. Mitigation of this threat occurred via time,
meaning that time between administration of measures was 8-months, with the first measure
(pretesting) occurring 1-month into the school year and the posttest measure occurring near the
end of the year. Due to the age of the participants, particularly the kindergarten students,
the kindergarten teacher administered the battery of tests in small groups with the research
assisting. The second threat to internal validity is sensitization to concepts related to the STEM
program. Table 1 illustrates an overview of the study design.
Intervention
The intervention used in this study is a whole school STEM integration curriculum designed
by the STEM coordinator and district office. The school district implemented the STEM program
starting in 2009 and continued the program until 2012 through a grant from Google and several
other entities, including Siemens, Boeing and the Department of Education. Within this study, the
fifth graders were exposed to 3 years of the integrated STEM program, the second graders were
exposed to 2 years of the intervention and the kindergarteners were exposed to 1 year of the
intervention. Using this intervention, the district attempted to merge and integrate the boundaries
of the principal STEM disciplines while connecting these subjects to others such as English
Language Arts (ELA) and Social Studies using the Engineering Design process as the focal point.
The school STEM Coordinator’s primary charge wasto build a culture of science and mathematics
within the school through teacher professional development and curriculum development. From a
curriculum perspective, the intent of the STEM coordinator was to develop the complexity of
ideas around STEM curriculum. This was accomplished through cross-cutting approaches in each
of the STEM disciplines and developed in conjunction with a focus on challenging answering and
Table1
Summary of study design N ¼254
Treatment Condition Group Label Pretest Treatment Posttest
Intervention School (n¼111) E O
1
XO
2
Kindergarten (n¼37)
2nd Grade (n¼44)
5th Grade (n¼30)
Comparison School (n¼143) C O
1
.O
2
Kindergarten (n¼31)
2nd Grade (n¼64)
5th Grade (n¼48)
Journal of Research in Science Teaching
418 LAMB, AKMAL, AND PETRIE
working through authentic real-world problems requiring understanding and connections across
not just STEM disciplines but other disciplines as well. Each grade within the school received
45 minutes of STEM specific instruction 3 days a week for a total of 90 hours of direct STEM
related instruction. This instruction took the form of laboratory and computer work within the
classroom. Teachers within the STEM school tied concepts and curriculum to the district and state
standards for science and mathematics. The 90 hours did not include STEM instruction that
resulted from cross curriculum integration and instruction occurring by outside agencies. By
contrast, the control school received only 10 hours of direct STEM related instruction with little to
no integration across the curriculum and no outside agency interactions related to STEM. The
comparison school engaged in traditional elementary methods and approaches in the classroom.
In addition to the 90hours of direct STEM integrated instruction by teachers, students within
the intervention school also received instruction from varied sources. For example, the students as a
part of the Lego Robotics School Program used Lego Robotics once a week for 45 minutes, three
time a week visits by STEM professionals acting as mentors for 1.5hours per visit, and all students
took part in the Architects in Schools Program 1 hour per week. Students within the intervention
school also took part in a 2-hour per week visit by retired mathematicians, scientists, and engineers
as a part of the ReSET program. During these visits, students took part in hands-on laboratory and
experiential learning through visits and outdoor activities. The STEM program also had learning
partnerships with the National Air and Space Museum, the Smithsonian, AAAS, and 67 other
organizations providing material, learningopportunities, and student and teacher mentoring.During
the instructional time given by teachers and outside organizations, the students had the opportunity
to learn core ideas related to the three spheres of activities for scientists and engineers (NRC, 2012
p.45). Each one of the activities covered some aspect of the sphere of activities though not
necessarily for equal amounts of time during each session. In the broader picture, the students
engaged in investigation, evaluation and the development of explanations and solutions to problems
as a part of Science and Engineering Practices (NGSS, 2013).At each grade level, the students were
presentedwith big ideas established by the teacher. For example, one coreidea for the kindergartners
that acted as a unifying theme throughout the quarter was an examination of Human Impacts on
Earth Systems. Under this theme, the kindergarteners worked to demonstrate proficiency in asking
questions, interpreting data, and providing evidence. The core idea for the second graders was an
understanding of solutions. Second grade students worked to demonstrate proficiency in solution
development, comparing solution, and optimization of solutions. The overarching fifth grade core
idea was the development of models. In this example, the fifth grade students worked to show
proficiency in their understanding of the impacts of humans on the Earth’s systems. Throughout the
year, the themes rotated with the intention to drive continuity and increasing complexity as the
students moved through the school year and when the students transitioned from one grade to
another.
The comparison school, in contrast, did not receive these programs, nor additional support
other than what is mentioned previously. In particular, the classroom teachers were the primary
drivers of the core ideas in the classroom with little coherence and adherence to the NextGeneration
Science Standards. The classes, external programs, and materials at the intervention school were
approached from and taught using problem based learning, project based learning, and in using an
inquiry based approaches. All of the problems and resulting projects the students in the intervention
school engaged in were nested in the overarching process of Design Based Learning.
Measures
The researchers collected data using three instruments: (1) paper and pencil surveys related to
affect, specifically self-efficacy and science interest; (2) cognitive measures related to spatial
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 419
visualization and mental rotation; (3) science content knowledge measures as part of end of course
testing. Psychometric properties of the student measures are discussed at length below.
Pretest–posttest measures of student outcomes consisted of a combination of a content test,
affective measures, and cognitive measures. The content test consisted of state-created science
and mathematics end-of-course tests prepared by Pearson. Affective tests consisted of the Science
Interest Survey (Lamb et al., 2012) and the Self-Efficacy in Science and Technology Short Form
(Lamb et al., 2014a). Cognitive tests consisted of the Paper Folding Task (Ekstrom, French,
Harman, & Dermen 1976) and the Shepard Metzler Test of Mental Rotation Task (Shepard &
Metzler, 1971). The teacher or STEM coordinator individually administered the measures to
kindergarten students, allowing them to ask clarifying questions. Each of the measures was
selected due to the sensitivity of each construct to stimulation by STEM instruction. Meaning,
each of the constructs is thought to increase when measured as a part of the successful STEM
program. The focus of the measures varied from science, science and mathematics, to general
cognitive measures. The variable of the focus is partially due to a desire to capture the breadth and
complexity of STEM learning. By this, the authors attempted to both look at components of
STEM disciplines such as science and mathematics separately (using affective measures) and as
an integrated construct using the cognitive and content measures. The use of science and
mathematics specific measures in addition to the two selected cognitive measures more broadly
allows the examination of cross-cutting concepts such as pattern recognition, and cause and effect
as these cognitive attributes and components are the structural aspects of the brain that are
responsible for the students’ ability to engage in these activities. The affective measures within the
study allow for examination of the participant’s understanding of how science addresses questions
about the natural and material world. This is not to suggest that other measures looking at the
integrated construct of STEM, and each of the disciplines separately would not be appropriate.
The selected constructs have a large research base in the current literature and are linked to
individual components of STEM and to the construct as a whole (Penuel & Fishman, 2012; Tal &
Dierking, 2014; Tan, Calabrese Barton, Kang, & Oneill, 2013). Some additional areas of cognitive
measure future studies may want to address are critical reasoning and lateral thinking as they relate
to the construct of STEM. One other recommended are of research for future studies may also
be an examination of the degree to which STEM integration across disciplines is occurring at the
school.
Psychometric Propertied of the Measures
Science Efficacy in Technology and Science (SETS-SF). An affective measure originally
developed by Ketelhut (2006). The SETS-SF was redeveloped into a 16-item measure by Lamb
et al., (2014a). The measure is designed using a Likert like scale to ascertain student efficacy in
science and technology. The SETS-SF is a Rasch validated measure (M
Infit
¼0.97, M
Outfit
¼1.05,
M
Dif
¼0.1, Rasch Reliability ¼0.91). Researchers use this measure to quantifythe primary variable
of self-efficacy related to STEM, specifically technology and science. An example item from this
measure is “It is easyfor me to look at the results of an experiment and tell what they mean.”
Science Interest Survey (SIS-E). The SIS-E is an affective measure developed by Lamb et al.,
2012. It is an 18 item, forced choice dichotomous scale measure of the unidimensional construct
of interest related to science for elementary students. The SIS-E is a Rasch validated measure
(M
Infit
¼1.00, M
Outfit
¼1.01, M
Dif
¼0.2, Rasch Reliability ¼0.84). The primary variable
measured using this instrument is quantified interest related to science. An example item from this
measure is “My science teacher makes science interesting.”
Journal of Research in Science Teaching
420 LAMB, AKMAL, AND PETRIE
Paper-Folding Test (PFT). This is a cognitive measure developed by Ekstrom et al. (1976)
while at Educational Testing Services. The measure is a 20-item single factor cognitive test
measuring mental ability to perform complex spatial maneuvers and tasks. A single large
Eigenvalue accounts for the large majority of the variance within the measure thus indicating
assessment of a unidimensional construct spatial visualization. Each item on the test depicts two
or threefolds in a sheet of paper with a hole punched into it. Students select one of five drawings
showing how the punched paper would look when fully opened. The PFT was developed under a
classical test model with an estimated internal consistency reliability of a¼0.91 and 0.84 after
the use of a correction factor for guessing. The primary variable measured using this instrument is
quantified spatial visualization. An example item found on this measure is shown in Figure 1.
Scoring of the test was as correct (“1”) or incorrect (“0”). A correction factor for guessing was
used during scoring and calculated according to C-I/ (i-1) where Cis correct items, Iis incorrect
items, and iis the number of response options for each item.
Shepard and Metzler Test of Mental Rotation (MRT). The MRT is a cognitive measure
developed by Shepard and Metzler (1971) and is a 20-item, single factor test, broken into five sets
of a four items sensory-motor test for mental rotation ability. Each item consists of a criterion
figure and two correct alternative, and two distractors. The MRT development framework is under
a classical test model. Internal consistency reliability of the measure within this study is estimated
at a¼0.89 and 0.82 after the use of a correction factor for guessing. The primary variable
measured using this instrument is a quantification of Spatial Rotation ability. An example item is
found in Figure 2. As with the Paper-Folding Test, a correction factor for guessing was used during
scoring and calculated according to C-I/ (i-1).
Science and Mathematics Content Exam. This is a criterion-references yearly exam
consisting of 50 multiple-choice questions designed to test student knowledge related to
mathematics and science within specified grade levels. The questions asked of the students link
directly to state standards and state curriculum guides. Overall reliability of the instrument is
estimated based upon scoring reports presented by Pearson a¼0.83. Pearson developed the test at
the request of the district to assess yearly student content learning outcomes.
Table 2 provides an overview of reliability estimates using Cronbach’s alpha by measure, by
grade. Reliability estimates suggest the lowest level of reliability associated with the kindergarten
Figure 1. Sample item from the Paper-Folding Test. [Color figure can be seen in the online version of this article,
available at wileyonlinelibrary.com]
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 421
age group. This is expected given the varying ability of kindergarteners to interpret questions and
responses. The researchers attempted to mitigate this variability by allowing the students to ask
questions about work meaning as they participated (with the exception of the content test). All
measures are available for examination in Appendix Awith the exception of the content test due to
district testing policies and copyrights by Pearson. There is variation from the overall internal
consistency and reliability reported above due to differences between Rasch calculated reliability
and correction factors for guessing.
Data Analysis
To test the hypotheses presented earlier, researchers collected data from the comparison and
intervention schools. To develop a picture of how the STEM intervention affects the multiple
aspects of student learning within classrooms, the authors of the study conducted multiple
analyses. A Structural Equation Model (SEM) approach provides an overview of significant
differences between the treatment and comparison schools and provides a view of the
interrelationships found between constructs. The SEM also provides a means to verify the
presence of the cognition-priming model versus the affective-priming model of learning.
Specifically, due to measurement invariance across groups (demographically) within the
treatment and comparison groups, the authors used a structured means model, allowing for
estimation and comparison of group means on the constructs of interest to the researcher: efficacy,
interest, mental rotation, spatial visualization, and content. In addition to analyzing differences
across groups, the SEM provided confirmation of the measures’ structures via confirmatory factor
analysis. This provides additional evidence of each measures’ coherence.
Using the SEM to target specific relationships, a two-factor analysis of variance (ANOVA)
examining intervention status by grade is shown to be statistically significant allowing for
confirmation of the statistical significance and a deeper understanding of differences between
grades across the main effect of treatment (Lamb et al., 2012). Eta squared measured the omnibus
Table2
Internal consistency reliability measure by grade
Measure Kindergarten Second Grade Fifth Grade
SETS-SF .68 .86 .96
SIS-E .71 .74 .87
PFT .68 .88 .93
MRT .73 .84 .88
Content Exam .82 .84 .82
Figure 2. Sample item from the Test of Mental Rotation. [Color figure can be seen in the online version of this article,
available at wileyonlinelibrary.com]
Journal of Research in Science Teaching
422 LAMB, AKMAL, AND PETRIE
effect size by establishing the proportion of total variability in the dependent variable accounted
for by the effect of the intervention.
The authors of the study also conducted a path analysis to ascertain casual inferences related
to the measured constructs. It is important to note that intent of a path analysis is not to assign
cause, but to assist in the determination of viability of causal models. The researchers were
primarily interested in the effects of the measured constructs on the science content testing
outcomes. Path analysis differs from regression analysis in several key ways. First, path analysis
requires the formal specification of the model to be estimated and tested. Second, path analysis is a
multivariate technique specifying the relational aspects of two or more observed variables with
each equation solving for a parameter estimation. This effectively allows variables to exist as
dependent and independent variables simultaneously. Lastly, path analysis provides for evaluation
of model fit based upon modification indicators and not a significance test to explain differences or
variance accounting. Ordering of variables occurred within the model based upon theoretical and
empirical relationships found within the literature. In addition to model development from theory,
other alternative models were tested for data fit. Modification indices, parameterized at 3.84
(which is a conservative estimate), and equal to the x
2
critical value with one degree of freedom.
Results
Research Question 1: What is the effect of an integrated STEM curriculum on student affect,
specifically self-efficacy and interest related to science and technology? Research Question 2:
Does exposure to a STEM integrated curriculum generate change related to student cognition
related to mental rotation and spatial visualization? Both research questions were answered using
a combination of structural equation modeling, non-linear model fitting, and traditional ANOVA.
Using Mplus (Muthe
´n & Muthe
´n, 2007), and implementing a maximum likelihood method of
estimation, the authors propose a model of the difference across groups related to the measured
constructs. In Figure 3, the factor loadings are standardized, so the squared values indicate the
proportion of variance in the indicator accounted for by the variability in the construct. Analysis of
model fit statistics suggests adequate model fit (x
2
(21) ¼30.91, p¼0.075,CFI ¼0.94,
TLI ¼0.93, RMSEA ¼0.04, SRMR ¼0.05). The standardized structural coefficients for the path
from group to each of the constructs is statistically significant (p<0.05) thus indicating that the
two groups differ on the constructs of self-efficacy, science interest, spatial visualization, and
mental rotation. An estimate of effect size of the mean group difference across the statistically
significant constructs reveals self-efficacy (d¼1.27, large), science interest (d¼1.97, large),
spatial folding (d¼1.30, large), and mental rotation (d¼1.42, large). Thus, there is a relatively
large effect size between those students in the comparison group and the treatment group.
The ANOVA F-test results show that there are statistically significant differences on the
measured constructs across the main effect of group F (2,253) ¼6.97, p¼0.001,h
2
¼0.13). The
omnibus effect size suggests that 13% of the variance within measure scores is accounted for by
score differences between the treatment and comparison group. By Cohen’s (1988) guidelines for
h
2
, the effect size is considered large. The results of the Tukey post hoc multiple comparisons
indicate that there are statistically significant differences between groups within grade across the
constructs. Table 3 illustrates the difference.
Figure 4 illustrates a graphical comparison of the results of group scores by grade. The curve
lines represent actual best-fit “lines” for the data. The straight lines represent lines tangent to the
curve. The tangent line is a line, which intersects at a point equaling the function that identifies
the domain variables for which the derivatives exist. Less formally, the tangent line intersects the
curve at a point where the slope of the curve equals the slope of the line. The appearance of a
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 423
Figure3. Model for group difference on five constructs.
Table3
Sample statistics, F-values, and effect size
Grade F-value n
E
n
C
p-value h
2
Effect DM
Science Interest K 1.70 37 31 0.093 0.001 None 1
Science Interest
*
2 4.63 44 64 0.001 0.15 Large 3
Science Interest
*
5 12.06 30 48 <0.001 0.18 Large 6
Mental Rotation K 1.87 37 31 0.10 0.002 None 1
Mental Rotation
*
2 2.01 44 64 0.059 0.01 Small 4
Mental Rotation
*
5 3.49 30 48 <0.001 0.16 Large 5
Spatial Folding K 0.19 37 31 0.94 0.004 None 1
Spatial Folding 2 1.94 44 64 0.98 0.009 Small 3
Spatial Folding
*
5 2.21 30 48 0.05 0.13 Large 3
Content K 0.05 37 31 0.99 0.02 Small 1
Content 2 2.62 44 64 0.08 0.09 None 1
Content 5 N/A 30 48 N/A 0.00 None 0
Self-Efficacy K 1.09 37 31 0.99 0.004 None 1
Self-Efficacy
*
2 3.27 44 64 0.03 0.21 Large 5
Self-Efficacy
*
5 4.16 30 48 0.04 0.17 Large 8
*
indicates a statistically significant difference between the control and experimental groups.
Journal of Research in Science Teaching
424 LAMB, AKMAL, AND PETRIE
difference at grade “0”(kindergarten) is the difference after students had participated in the STEM
program for one year.
Comparison of affect results via Figure 2 illustrate how self-efficacy remained relatively
stable over time for the treatment group, while the comparison group demonstrated a rapid decline
indicated by the negative slope of the curve related to self-efficacy between grades kindergarten
(0) and grade five (5). While there is a similar trend found with the treatment group experiencing a
slower decline in interest, both groups demonstrated a decline moving from kindergarten to fifth
grade.
Figure 5 illustrates change in cognitive attributes related to grade. Given the similarity between
the comparison and treatment groups and the statistically significant differences of group means at
the kindergarten grade level (0), differences that appear prior to grade one are considered statistical
noise. Comparison of each graphic reveals that initial developmental rates of the students are equal.
However, as students increased exposure to the STEM intervention, differences begin to appear as
early as second gradewith significant differences appearing in fifth grade. Analysis of tangent line
slopes reveals a more rapid increasein cognitive development within the treatmentgroup. Variance
accounting specifically for each of the cognitive attributes suggests that 19.4% of the difference in
mental rotation is accounted for via the intervention. Variance accounting for spatial folding
suggests that 24.1% of the difference in this measure is accounted for via the intervention. These
calculations do not account for the non-linear nature of the curves. Review of the model fit statistics
for both cognitive attributes suggests that treatment condition is predictive of increases in mental
rotation (x
2
(2) ¼8.97, p¼0.011) and spatial folding (x
2
(2) ¼7.31, p¼0.025). There is also good
data fit to the model, as indicated by a non-significant Hosmer–Lemeshow statistic (x
2
(3) ¼2.74,
p¼0.43).
Three models of interaction between the variables were examined, two of the models are
based within the literature, and the third was developed stepwise purely based upon model
convergence with data using modification indices. The first model examined was the affective-
priming model. The second model examined was the cognition-priming model and the third
model was guided using modification indices. Examination of model comparisons occurred using
the Bayesian Information Criterion (BIC). The BIC takes into account measure fit and model
complexity. BIC also increases the penalty to increases in sample size. Comparatively, the model
with the smallest BIC is considered to have the greatest fit. When comparing fit between models,
greater than a 5-point difference is considered most likely different while, greater than a 10-point
difference is considered almost certainly different (Dimitrov, 2006; Lee, Song, Skevington, &
Hao, 2005). Results of BIC analysis illustrates the cognition-priming model (BIC ¼10394.26)
provides the lowest BIC when compared with the affect-priming model (BIC ¼10487.42).
Stepwise modifications of an exploratory model ultimately and independently resulted in the
cognition-priming model as proposed for the path analysis. Thus, this provides independent
confirmation of the cognition-priming learning model.
Research Question 3: What is the relationship between affective and cognitive measures and
science content score outcomes? was answered using a path an alysis relating cognitive attribut es,
affective traits, and science content (behavior). Results of the path analysis shown in Figure 6
provide insight into the constructs that have a statistically significant relative effect on science
content learning. Review of model fit statistics suggested by Dimitrov (2006) indicate adequate
model fit (x
2
(21) ¼38.147, p¼0.012, CFI ¼0.94, TLI ¼0.93, RMSEA ¼0.04, SRMR ¼0.05)
(Bentler & Bonett, 1980). This is despite a statistically significant chi square value as chi square
is sensitive to sample size and the other indicators are at an acceptable level of fit (Raykov &
Marcoulides, 2010). In the case of the cognitive attributes, the path diagrams are non-recursive;
however in the case of the affective traits, the path diagrams are recursive indicating that the
Journal of Research in Science Teaching
426 LAMB, AKMAL, AND PETRIE
relationships are bidirectional and not unidirectional. Positive path coefficients are suggestive of
a positive linear relationship between the variables. Direct effects are shown between the
affective traits, cognitive attributes, and science content scores. Indirect effects are observed
within this model from cognition to content (behavior), to affect. In addition, positive path
coefficients resulting from the regression of science content scores on the cognitive traits are
found to be causal (within the model) in increases in self-efficacy and interest. Significant
positive relationship between interest and efficacy and between spatial visualization and mental
rotation exist, though these relationships are not indicative of causal relationships. Taken as a
whole, the four co nstructs (efficacy, interest, visualization, and rotation) account for 90.7% of t he
variance within the m odel.
Discussion
The goal of this study was to examine the cognitive, content, and affective relationships and
outcomes associated with the STEM based education within urban elementary schools. Results of
the structural means model-a form of structural equation modeling- suggest statistically
significant differences across the measured construct with the exception of the content measure.
More detailed examination of the individual grade differences provide a clearer picture of the
grade level at which these differences start to manifest, particularly as a function of the amount of
time related to participation in the intervention. While much of this discussion has focused on the
developmental aspects of the results from the study,a second and equally important consideration
is the “time in intervention.” The interaction of maturation and time in the intervention provides
explanatory effects but is difficult to disentangle. Examination of the outcome curves suggests that
the total time in the intervention can act as an explanatory factor for creating the increased gap
between the comparison and treatment schools increased. These increases in cognition and affect
are potentially due to time in intervention and in tandem or addition to developmental increases.
Finally, path analysis also confirms the cognitive-priming model found within the literature and
Science Self-
Efficacy
Science Interest
Spatial-
Visualization
Mental Rotation
Science Content
p
51
=.41
p
54
=.33
p
53
=.28
p
52
=.51
r
12
=.68
r
34
=.47
p
15
=.69
p
25
=.84
Figure6. Path analysis for the cognition-priming model.
Journal of Research in Science Teaching
428 LAMB, AKMAL, AND PETRIE
context suggests this model plays a key role in student outcomes and provides a mechanistic
means to understand why interventions later within student development are not as successful.
Research Question 1: what is the effect of an integrated STEM curriculum on student affect,
specifically self-efficacy and interest related to STEM? and Research Question 2: does exposure to
a STEM integrated curriculum generate change related to student cognition related to mental
rotation and spatial visualization? are both answered though analysis of each set of model data
and visualization of data. There is evidence of differences between the comparison and treatment
schools in both affect and cognition. These differences seem to arise over time and not necessarily
within a single year of implementation. ANOVA results provide evidence of differences for affect
starting in second grade and changes in cognition occurring between second and fifth grade. This
may arise from a divergence in developmental timelines for affect and cognition. Affect related to
interests is evidenced in newborn infants and affect related to more complex affect (such as
efficacy) is evidenced in children around two years of age (Monk, Geirgueff, & Osterholm, 2013).
Thus, changes in the affect are more likely to become viable earlier on than cognitive difference,
which require more time to both develop and indicate differences. Moreover, cognitive attributes
also tend to become far more stable than affective traits and require greater time in intervention to
see change. The most complex of the outcomes to measure is science content knowledge. The
difficulty in measuring this component of the study arises from the need to integrate both cognitive
and affective components associated with the construct of science content knowledge.
Research Question 3: what is the relationship between affective and cognitive measures and
science content score outcomes? is answered using a structural equation model developed from
theoretical and empirical evidence. The relationship between the affective, cognitive, and content
seems to develop from the cognition-priming model. To examine this model it is first important to
generalize content as an externalized subsequent behavior. Reconceptualization of content as a
behavior provides the externalized stimuli needed to generate affect and allows application of the
cognition-priming model.
Through visualization of the data shown in Figures 2 and 3, one can start to develop a
characterization of the changes seen across grade level. Starting with Figure 2, the affective curves
take into account the relationship seen within the cognitive priming-model and the link between
both self-efficacy and interest. This relationship is recursive and creates a feedback loop between
content and affect. More specifically, as a student engages with content positively or negatively,
the change within the affect occurs and magnifies with interaction. Qualitatively examining the
slopes of the tangent lines provides evidence that the STEM education within the intervention
school develops higher levels of self-efficacy and interest. These increased levels may help to
eliminate the movement of students away from STEM courses due to reduced self-efficacy and
interest over time. Within the comparison school, the slope of the tangent line is negative. This in
part seems to be due to the nature of the construct of self-efficacy. Essentially, self-efficacy levels
are collectively falling within the age cohort as the members move from elementary to middle
school this may be due to a realization of their level of understanding (Plass, Milne, Homer,
Schwartz, & Hayward, 2012). The influence of the age cohort members on the individuals within
the cohort is increasing, creating downward affect pressures on the individual students within the
cohort. This group norming pressure results in less selection of STEM based courses and careers
and creating a cycle of decline that becomes increasingly difficult to change over time (Tal &
Morag, 2013). This difficulty in raising interest and efficacy results from self-efficacy and interest
traits stabilizing as the student ages. In part, the stability within the affect traits arises due increased
need for higher-level cognition to arouse affect as seen within the cognition-priming model. These
findings point to a potentially important area of consideration when developing curriculum and
lessons within the STEM classroom.
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 429
Review of the data within Figure 3 provides an overview of cognitive development within the
students from kindergarten to fifth grade. Review of the figures provides evidence that early
exposure to STEM curriculum increases development of cognitive attributes related to STEM
tasks. Tangent lines provide suggestions that the rate of development for mental rotation and
spatial visualization cognitive attributes increase more quickly. For mental rotation, developmen-
tal psychology literature suggests this attribute starts its development between 3 and 5 years of
age, while spatial visualization starts its development between 5 and 7 years of age. While it is
possible to “retrain” attributes post-development, it becomes increasingly difficult to do so as the
cognitive attributes are less plastic as one gets older. This does not suggest that neuroplasticity is
not present later in life just there is greater opportunity earlier on. The interesting connection
between affect and cognition occurs through the content as evidenced in the cognition-priming
model. Based upon the processing role that spatial visualization and mental rotation play within
cognition-priming theory, one can suggest a connection between the cognitive and affective
through internalization of external antecedent stimuli in the form of STEM content. Changes in
each of these cognitive attributes may have arisen through attributional retraining associated with
components of the classes such as three-dimensional model/robot building, projects involving
architecture, and experiments with the mentor scientists.
Implications for Practice
When considering the cognition-priming model, it becomes necessary to use higher-level cognition
to arouse changes in affect. When developing lessons for students within the science classroom the
practitioner must consider student cognitive level of development and amount of prior exposure to
content if the interest of the educator is to change stable affect orientations toward STEM courses and
career selection. Older students (fifth grade) will require much more cognitively complex interactions
often seen within problem-based learning, and more broadly, inquiry. Thus, the natural tendency of the
educator to reduce cognitive difficulty when encountering students with less prior experience or
knowledge actually results in less positive affect when considering the cognition-priming model. By
this, what is meant is that the teacher may seek to reduce cognitive aspects of engagement as a way to
increase student participation in the classroom. This in fact may reduce student participation, as
intellectual engagement does not occur leaving the student open to affective priming, which is much less
predictable and may arise from negative experiences in the science classroom. Evidence for this exists in
the development of the STEM intervention for the treatment school. The treatment school—in an
intentional way—sought to include complex real-life examples and problems within the classroom to
assist in driving learning forward. This is because the cognitive priming is sufficient to first engage the
student using cognitive tools that then create arousal of higher-level affect such as positive efficacy and
interest. This is not to suggest that students having difficulty within a STEM course require more difficult
problems to solve but that the approach should be sufficiently cognitively stimulating to create the
change in affect and ultimately STEM orientation. Within the science classroom, cognition-priming
would take the form of an initial discrepant event demonstration followed by scaffolding through a
related real-world problem leading to the development of ideas, questions, experimentation,
observations, claims, evidence, consultation with others, and ultimately reflection and products.
Educational and cognitive psychology suggest that rich and complex experiences tied to prior
understanding provide a means to engage students who might otherwise be unwilling to participate in
the classroom. More importantly, this suggests from a policy and curriculum point of view that it is far
easier to create and maintain affect towards STEM the earlier the complex exposure occurs. Complex
exposure simply means exposure in the form of inquiry learning or problem-based learning, which
provide greater activation of cognition. With the increased difficulty of changing stable affect later in the
student’s life and the difficulty in retraining stable cognitive attributes later in life there is considerable
Journal of Research in Science Teaching
430 LAMB, AKMAL, AND PETRIE
evidence that the inclusion of STEM integrated learning at the earlier elementary level becomes
critically important for the students.
Taking the developmental aspects of the cognition-priming model, one can begin to see why
STEM interventions within the higher-grade levels may not be as effective. This is a critical
challenge for science educators to overcome. Considering the national movement to create more
STEM high schools at the expense of middle and elementary schools, policy makers may not see
the gains they are hoping to see at the high school level. In this light curriculum and assessment
developers can facilitate situations for younger learners by moving away from rote and predefined
inquiry involving causality and patterns and allowing students to generate their own understandings
of the world though argumentation, presentation of evidence, and evaluation of evidence.
Curriculum developers may alsowant to focus around the continuity in core ideas found within the
Next Generation Science Standards. This is not to say that there are no other effective means to
engage students in the classroom around STEM, just that from a cost-benefit approach, outcomes
seem more promising when schools implement STEM interventions earlier and continue the
exposure throughout the students’ educational career, seemingly increasing opportunities for
success. This success may in part be due to the development of suitable cognitive, affective, and
ultimately attitudinal infrastructurethat is developed for the student to capitalizeupon later in life.
Limitations
Several important limitations should be considered when reviewing the results of this study.
First, data collection skipped grade levels for those grades not explicitly identified. Second, the
results are also limited due to the use of self-reporting measures for affect. Specifically, additional
resultant research is needed to verify the consistency and accuracy of findings. The addition of
more measures related to cognition and affect can provide greater resolution related to effects on
content and with content’s effects on these components. Lastly, given the homogeneity of the
sample, additional follow-up studies are required for increased generalizability. Despite these
limitations, this study provides important relational information regarding the interplay of
content, affect, and cognition related to STEM curriculum. This study also provides concrete
suggestions for exposure to STEM curriculum. In addition, the findings in this study provide a
novel approach to examining student learning within the STEM classroom.
Conclusion
Taken as a whole, the results of this study identify the importance of addressing each of the
three areas psychologically related to learning. The results also highlight the need to understand
the interplay of cognition, affect, and content within the context of developing a sophisticated
understanding of learning within STEM education. The cognition-priming model proposed in this
study confirms the relationship between interest and self-efficacy, mental rotation, and spatial
visualization, and the construct’s relationship to content. Overall, student change in each of the
areas is present pointing to positive effects of STEM education. The study shows that there is
potential in STEM education for elementary students and helps student to develop cognitive and
affective aspects related to STEM beyond content alone.
References
Adedokun, O. A, Bessenbacher, A. B., Parker, L. C.,Kirkham, L. L., & Burgess, W. D. (2013). Research
skills and STEM undergraduate research students’ aspirations for research careers: Mediating effects of
research self-efficacy. Journal of Research in Science Teaching, 50(8), 940–951.
Anderman, E., & Young, A. (1994). Motivation and strategy use in science: Individual difference and
classroom effects. Journal of Research in Science Teaching, 31(8), 811–831.
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 431
Annetta, L. A., Frazier, W. M., Folta, E., Holmes, S., Lamb, R., & Cheng, M. T. (2013). Science teacher
efficacy and extrinsic factors toward professional development using video games in a design-based research
model: The next generation of STEM learning. Journal of Science Education and Technology, 22(1), 47–61.
Archer, L., DeWitt, J., & Willis, B. (2014). Adolescent boys’ science aspirations: Masculinity, capital,
and power.Journal of Research in Science Teaching, 51(1), 1–30.
Azevedo, F. S. (2011). Lines of practice: A practice-centered theory of interest relationships. Cognition
and Instruction, 29(2), 147–184.
Baki, A., Kosa, T., & Guven, B. (2011). A comparative study of the effects of using dynamic geometry
software and physical manipulatives on the spatial visualization skills of pre-service mathematics teachers.
British Journal of Educational Technology, 42(2), 291–310.
Bandura, A. (1997). Self-efficacy: The exerciseof control. New York: Freeman.
Bandura, A. (1994). Self-efficacy.John Wiley & Sons, Inc.
Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning. Educational
Psychologist, 28(2), 117–148.
Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37(2), 122.
Bandura, A. (1977a). Social learning theory.
Bandura, A. (1977b). Self-efficacy: Toward a unifying theory of behavioral change. Psychological
Review, 84(2), 191.
Bem, D. J. (2011). Feeling the future: Experimental evidence for anomalous retroactive influences on
cognition and affect. Journal of Personality and Social Psychology, 100(3), 407.
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance
structures. Psychological Bulletin, 88(3), 588.
Berkowitz, L. (1993). Aggression: Its causes, consequences, and control. Mcgraw-Hill Book Company.
Berry, J. M., West, R. L., & Dennehey, D. M. (1989). Reliability and validity of the memory self-efficacy
questionnaire. Developmental Psychology, 25(5), 701.
Blanchette, I., & Richards, A. (2010). The influence of affect on higher-level cognition: A review of
research on interpretation, judgment, decision making and reasoning. Cognition & Emotion, 24(4), 561–595.
Bleasdale, F. A. (1987). Concreteness-dependent associative priming: Separate lexical organization for
concrete and abstract words. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13(4), 582.
Bodzin, A. M. (2011). The implementation of a geospatial information technology (GIT)-supported
land use change curriculum with urban middle school learners to promote spatial thinking. Journal of
Research in Science Teaching, 48(3), 281–300.
Bogg, T., & Finn, P. R. (2010). A self-regulatory model of behavioral disinhibition in late adolescence:
Integrating personality traits, externalizing psychopathology, and cognitive capacity. Journal of Personality,
78(2), 441–470.
Bohner, G., & Dickel, N. (2011). Attitudes and attitude change. Annual Review of Psychology,62, 391–417.
Bouckenooghe, D. (2010). Positioning change recipients’ attitudes toward change in the organizational
change literature. The Journal of Applied Behavioral Science, 46(4), 500–531.
Breckler, S. J. (1984). Empirical validation of affect, behavior, and cognition as distinct components of
attitude. Journal of Personality and Social Psychology, 47(6), 1191.
Britner, S. L., & Pajares, F. (2006). Sources of science self-efficacy beliefs of middle school students.
Journal of Research in Science Teaching, 43(5), 485–499.
Carrion, R. E., Goldberg, T. E., McLaughlin, D., Auther, A. M., Correll, C. U., & Cornblatt, B. A.
(2011). Impact of neurocognition on social and role functioning in individuals at clinical high risk for
psychosis. The American Journal of Psychiatry, 168(8), 806.
Casey, M. B., Nuttall, R., Pezaris, E., & Benbow, C. P. (1995). The influence of spatial ability on gender
differences in mathematics college entrance test scores across diverse samples. Developmental Psychology,
31(4), 697.
Cohen, J. (1988). Statistical power analysis for the behavioral sciencies. Routledge.
Cowell, R. A., Bussey, T. J., & Saksida, L. M. (2012). Empiricists are from Venus, modelers are from
Mars: Reconciling experimental and computational approaches in cognitive neuroscience. Neuroscience &
Biobehavioral Reviews, 36(10), 2371–2379.
Journal of Research in Science Teaching
432 LAMB, AKMAL, AND PETRIE
Cox, W. M. & Klinger, E. (Eds.). (2011). Handbook of motivational counseling: Goal-based approaches
to assessment and intervention with addiction and other problems. West Sussex, UK: John Wiley & Sons
Cox, J. W.(1928). Mechanical aptitude. London: Methuen.
Decety, J., Michalska, K. J., & Kinzler, K. D. (2012). The contribution of emotion and cognition to
moral sensitivity: A neurodevelopmental study. Cerebral Cortex, 22(1), 209–220.
Drake, C., Land, T. J., & Tyminski, A. M. (2014). Using educative curriculum materials to support the
development of prospective teachers’ knowledge. Educational Researcher,0013189–14528039.
Dreyfus, A., Jungwirth, E., & Eliovitch, R. (1990). Applying the “cognitive conflict” strategy for
conceptual change—some implications, difficulties, and problems. Science Education, 74(5), 555–569.
Dimitrov, D. M. (2006). Comparing groups on latent variables: A structural equation modeling
approach. Work:A Journal of Prevention, Assessment and Rehabilitation, 26(4), 429–436.
Edgar, K., & Fox, N. (2006). Temperamental contributions to children’s performance in an emotion-
word processing task: A behavioral and electrophysiological study. Brain and Cognition, 65(1), 22–35.
Ekstrom, R. B., French, J. W., Harman, H. H., & Dermen, D. (1976). Kif offuctor-referenced cognitive
tests. Princeton, NJ: Educational Testing. Service.
Epstein, S. (1993). Implications of cognitive-experiential self-theory for personality and developmental
psychology.
English, M. C., & Kitsantas, A. (2013). Supporting student self-regulated learning in problem-and
project-based learning. Interdisciplinary Journal of Problem-based Learning, 7(2), 6.
Feldon, D. F., Timmerman, B. C., Stowe, K. A., & Showman, R. (2010). Translating expertise into
effective instruction: The impacts of cognitive task analysis (CTA) on lab report quality and student retention
in the biological sciences. Journal of Research in Science Teaching, 47(10), 1165–1185.
Fiske, S. T.(2013). Social cognition: From brains to culture. Sage.
Fosnot, C. T. (1993). Rethinking science education: A defense of Piagetian constructivism. Journal of
Research in Science Teaching, 30(9), 1189–1201.
Fowler, F. J. (Ed.). (2009). In: Survey research methods. Vol. 1. Sage.
Fredricks, J., Alfeld, C., & Eccles, J. (2010). Developing and fostering passion in academic and
nonacademic domains. Gifted Child Quarterly, 54(1), 18–30.
Goddard, R. D. (2001). Collective efficacy: A neglected construct in the study of schools and student
achievement. Journal of Educational Psychology, 93(3), 467.
Gonzalez-Castillo, J., Saad, Z. S., Handwerker, D. A., Inati, S. J., Brenowitz, N., & Bandettini, P. A.
(2012). Whole-brain, time-locked activation with simple tasks revealed using massive averaging and model-
free analysis. Proceedings of the National Academy of Sciences, 109(14), 5487–5492.
Gottfried, M. A. (2010). Evaluating the relationship between student attendance and achievement in
urban elementary and middle schools an instrumental variables approach. American Educational Research
Journal, 47(2), 434–465.
Hackett, G. (1995). Self-efficacy in career choice and development. In: A. Bandura (Ed.), Self-efficacy
and changing societies. (pp. 232–258). New York: Cambridge University Press.
Halbesleben, J. R., Wheeler, A. R., & Paustian-Underdahl, S. C. (2013). The impact of furloughs on
emotional exhaustion, self-rated performance, and recovery experiences. Journal of Applied Psychology,
98(3), 492.
Harackiewicz, J. M., Durik, A. M., & Barron, K. E. (2005). Multiple goals, optimal motivation, and the
development of interest. Social motivation: Conscious and unconscious processes, 21–39.
Hardt, O., Einarsson, E., & Nader, K. (2010). A bridge over troubled water: Reconsolidation as a link
between cognitive and neuroscientific memory research traditions. Annual Review of Psychology, 61, 141–167.
Harris, J., Newcombe, N. S., & Hirsh-Pasek, K. (2013). A new twist on studying the development of
dynamic spatial transformations: Mental paper folding in young children. Mind, Brain, and Education, 7(1),
49–55.
Henderson, C., Beach, A., & Finkelstein, N. (2011). Facilitating change in undergraduate STEM
instructional practices: An analytic review of the literature. Journal of Research in Science Teaching, 48(8),
952–984.
Hidi, S. (2006). Interest: A unique motivational variable.Educational Research Review, 1(2), 69–82.
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 433
Hinkel, E. (2012). Cognitive-code learning. In: Encyclopedia of the sciences of learning (pp. 625–626).
Springer: US.
Houseal, A. K., Abd-El-Khalick, F., & Destefano, L. (2014). Impact of a student–teacher–scientist
partnership on students’ and teachers’ content knowledge, attitudes toward science, and pedagogical
practices. Journal of Research in Science Teaching, 51(1), 84–115.
Huang, M. (2006). Flow, enduring, and situational involvement in the Web environment: A tripartite
second-order examination. Psychology and Marketing, 23(5), 383–411.
Ibrahim, M., Callaway, R. (2012, March). Assessing the correlations among cognitive overload, online
course design and student self-efficacy. In: Society for Information Technology & Teacher Education
International Conference (Vol. 2012, No. 1, pp. 463–470).
Ifenthaler, D., & Eseryel, D. (2013). Facilitating complex learning by mobile augmented reality learning
environments. In: Reshaping learning (pp 415–438). Berlin Heidelberg: Springer.
Johri, A., & Olds, B. M. (2011). Situated engineering learning: Bridging engineering education research
and the learning sciences. Journal of Engineering Education, 100(1), 151–185.
Johnston, R. C., & Viadero, D. (2000). Unmet promise: Raising minority achievement.Edweek.org.
Kappes, A., Oettingen, G., & Pak, H. (2012). Mental contrasting and the self-regulation of responding to
negative feedback. Personality and Social Psychology Bulletin, 38(7), 845–857.
Ketelhut, D.J., (2006) Do students with higher self-efficacy exhibit greater and more diverse scientific
inquiry skills: An exploratory investigation in.
Kirsh, D. (2009). Problem solving and situated cognition. The Cambridge handbook of situated
cognition, 264–306.
Konstantinou, N., Bahrami, B., Rees, G., & Lavie, N. (2012). Visual short-term memory load reduces
retinotopic cortex response to contrast. Journal of Cognitive Neuroscience, 24(11), 2199–2210.
Kimmons, R., Liu, M., Kang, J., & Santana, L. (2011). Attitude, achievement, and gender in a middle school
science-based ludic simulation for learning. Journal of Educational Technology Systems, 40(4), 341–370.
Klassen, S. (2006). Contextual assessment in science education: Background, issues, and policy.
Science Education, 90(5), 820–851.
Klassen, R. M., & Usher, E. L. (2010). Self-efficacy in educational settings: Recent research and
emerging directions. Advances in Motivation and Achievement, 16, 1–33.
Krapp, A. (2007). An educational-psychological conceptualization of interest. International Journal for
Educational and Vocational Guidance, 7(1), 5–21.
Lamb, R. L., & Annetta, L. (2013). The use of online modules and the effect on student outcomes in a
high school chemistry class. Journal of Science Education and Technology,22(5), 603–613.
Lamb, R., Vallett, D., & Annetta, L. (2014a). Development of a short form measure of science and
technology self-efficacy using Rasch analysis. Journal of Science Education and Technology, 1–17.DOI:
10.1007/s10956-014-9491-y
Lamb, R., Annetta, L., Vallett, D., & Sadler, T. (2014b). Cognitive diagnostic like approaches using
neural-network analysis of serious educational videogames. Computes & Education, 70, 92–104.
Lamb, R., Annetta, L., Meldrum, J., & Vallett, D. (2012). Measuring science interest: Rasch validation
of the science interest survey.International Journal of Science and Mathematics Education, 10(3), 643–668.
LeDoux, J. (2012). Rethinking the emotional brain. Neuron, 73(4), 653–676.
Lee, S. Y., Song, X. Y., Skevington, S., & Hao, Y. T. (2005). Application of structural equation models to
quality of life. Structural Equation Modeling, 12(3), 435–453.
Lent, R. W., & Hackett, G. (1987). Career self-efficacy: Empirical status and future directions. Journal
of Vocational Behavior,30, 347–382.
Lent, R. W., Lopez, F. G., Brown, S. D., & Gore, P. A., Jr (1996). Latent structure of the sources of
mathematics self-efficacy. Journal of Vocational Behavior,49(3), 292–308.
Lent, R. W.,Miller, M. J., Smith, P. E., Watford, B. A., Lim, R. H., Hui, K., & Williams, K. (2013). Social
cognitive predictors of adjustment to engineering majors across gender and race/ethnicity. Journal of
Vocational Behavior.
Lopez, F. G., & Lent, R. W. (1992). Sources of mathematics self-efficacy in high school students. The
Career Development Quarterly, 41(1), 3–12.
Journal of Research in Science Teaching
434 LAMB, AKMAL, AND PETRIE
Lowe, R., & Ziemke, T. (2011). The feeling of action tendencies: On the emotional regulation of goal-
directed behavior. Frontiers in Psychology, 2.
Lubinski, D. (2010). Spatial ability and STEM: A sleeping giant for talent identification and
development. Personality and Individual Differences,49(4), 344–351.
Mamor, G. S. (1975). Development of kinetic images: When does the child first represent movement in
mental images? Cognitive Psychology, 7, 548–559.
Mathewson, J. H. (1999). Visual-spatial thinking: An aspect of science overlooked by educators.
Science Education, 83(1), 33–54.
Matsui, T., Matsui, K., & Ohnishi, R. (1990). Mechanisms underlying math self-efficacy learning of
college students. Journal of Vocational Behavior, 37, 223–238.
McDonald, C. V. (2010). The influence of explicit nature of science and argumentation instruction on
preservice primary teachers’ views of nature of science. Journal of Research in Science Teaching, 47(9),
1137–1164.
McGuire, W. J. (1969). The nature of attitudes and attitude change. In: G. Lindzey & E. Aronson (Eds.),
The handbook of social psychology (Vol. 3: pp. 136–314). Reading, MA: Addison-Wesley.
Michael, W. B., Guilford, J. P., Fruchter, B., & Zimmerman, W. S. (1957). The description of spatial-
visualization abilities. Educational and Psychological Measurement.
Monk, C., Georgieff, M. K., & Osterholm, E. A. (2013). Research review: Maternal prenatal distress and
poor nutrition–mutually influencing risk factors affecting infant neurocognitive development. Journal of
Child Psychology and Psychiatry, 54(2), 115–130.
Murray, B. D., & Kensinger, E. A. (2012). The effects of emotion and encoding strategy on associative
memory.Memory & cognition, 40(7), 1056–1069.
Muthe
´n, L. K., & Muthe
´n, B. O. (2007). Mplus. Statistical analysis with latent variables. Version, 3.
NGSS Lead States, 2013. Next Generation Science Standards: For States, By States. Washington, DC
The National Academies Press.
Noguera, P., (2011) Creating the opportunity to learn: Moving from research to practice to close the
achievement gap. ASCD.
Oppezzo, M., & Schwartz, D. L. (2013). A behavior change perspective on self-regulated learning with
teachable agents. In: International handbook of metacognition and learning technologies (pp. 485–500). New
York: Springer.
Osborne, J. (2010). Arguing to learn in science: The role of collaborative, critical discourse. Science,
328(5977), 463–466.
Paas, F., & Sweller, J. (2012). An evolutionary upgrade of cognitive load theory: Using the human motor
system and collaboration to support the learning of complex cognitive tasks. Educational Psychology
Review, 24(1), 27–45.
Patrick, H., Mantzicopoulos, P., & Samarapungavan, A. (2009). Motivation for learning science in
kindergarten: Is there a gender gap and does integrated inquiry and literacy instruction make a difference.
Journal of Research in Science Teaching, 46(2), 166–191.
Penuel, W.R., & Fishman, B. J. (2012). Large-scale science education intervention research we can use.
Journal of Research in Science Teaching, 49(3), 281–304.
Pellegrino, J. W. (2012). Assessment of science learning: Living in interesting times. Journal of
Research in Science Teaching, 49(6), 831–841.
Pezzulo, G., Barsalou, L. W., Cangelosi, A., Fischer, M. H., McRae, K., & Spivey, M. J. (2012).
Computational grounded cognition: A new alliance between grounded cognition and computational
modeling. Frontiers in Psychology, 3, 612–623.
Piaget, J., & Inhelder, B. (1963). Les images mentales. Traite
´de psychologie expe
´rimentale.
Piety, P. J. (2013). Assessing the educational data movement: 0. New York:Teacher College Press,
108–127.
Pinheiro, J. C., & Bates, D. M. (2000). Linear mixed-effects models: Basic concepts and examples. (pp.
3–56). New York: Springer.
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 435
Plass, J. L., Milne, C., Homer, B. D., Schwartz, R. N., Hayward, E. O., Jordan, T., & Barrientos, J.
(2012). Investigating the effectiveness of computer simulations for chemistry learning. Journal of Research
in Science Teaching, 49(3), 394–419.
Pruden, S. M., Levine, S. C., & Huttenlocher, J. (2011). Children’s spatial thinking: Does talk about the
spatial world matter? Developmental Science, 14(6), 1417–1430.
Raykov, T., & Marcoulides, G. A. (2010). Introduction to psychometric theory. New York: Taylor
Francis, 269–301.
Rahm, J. (2014). Reframing research on informal teaching and learning in science: Comments
and commentary at the heart of a new vision for the field. Journal of Research in Science Teaching,
51(3), 395–406.
Revelle, W., Wilt, J., & Rosenthal, A. (2010). Individual differences in cognition: New methods for
examining the personality-cognition link. In: Handbook of individual differences in cognition. (pp. 27–49).
NY:Springer New York.
Richetin, J., Conner, M., & Perugini, M. (2011). Not doing is not the opposite of doing: Implications for
attitudinal models of behavioral prediction. Personality and Social Psychology Bulletin, 37(1), 40–54.
Rivet, A. E., & Kastens, K. A. (2012). Developing a construct-based assessment to examine students’
analogical reasoning around physical models in Earth Science. Journal of Research in Science Teaching,
49(6), 713–743.
Sawtelle, V., Brewe, E., & Kramer, L. H. (2012). Exploring the relationship between self-efficacy and
retention in introductory physics. Journal of Research in Science Teaching,49(9), 1096–1121.
Scherer, K. R. (1988). Criteria for emotion-antecedent appraisal: A review. In: Cognitive perspectives
on emotion and motivation. (pp. 89–126). Netherlands: Springer.
Schoon, I., Ross, A., & Martin, P. (2007). Science related careers: Aspirations and outcomes in two
British cohort studies. Equal Opportunities International, 26(2), 129–143.
Schiefele, U. (1998). Individual interest and learning, what we know and what we don’t know. In: L.
Hoffman, A. Krapp, K. Renninger, & J. Baumert (Eds.), Interest and learning: Proceedings of the Seeon
Conference on Interest and Gender. Kiel, Germany: Kiel Publishing.
Schweingruber, H., Keller, T., & Quinn, H. (Eds.). (2012). In: A framework for K-12 science education:
Practices, crosscutting concepts, and core ideas. National Academies Press. 7–311.
Seel, M. N., & Winn, W. D. (2012). Research on media and learning: Distributed cognition and
semiotics. Theory, research, and models ofinstructional design: International perspective, 1, 32–293.
Shepard, R.N., Metzler, J. (1971) Mental rotation of three-dimensional objects.
Smetana, J., Campione-Barr, N., & Metzger, A. (2006). Adolescent development in interpersonal and
societal contexts. Annual Review of Psychology, 57, 255–284.
Somerville, L. H., Wagner, D. D., Wig, G. S., Moran, J. M., Whalen, P. J., & Kelley, W. M. (2013).
Interactions between transient and sustained neural signals support the generation and regulation of anxious
emotion. Cerebral Cortex, 23(1), 49–60.
Spence, I., & Feng, J. (2010). Video games and spatial cognition. Review of General Psychology,
14(2), 92.
Stieff, M. (2007). Mental rotationand diagrammatic reasoning in science. Learning and Instruction, 17(2),
219–234.
Stricker, G. (2012). The test of time: Strupp and Hadley’s tripartite model of mental health. Clinical
Child Psychology and Psychiatry, 17(4), 631–637.
Swarat, S., Ortony, A., & Revelle, W. (2012). Activity matters: Understanding student interest in school
science. Journal of Research in Science Teaching, 49(4), 515–537.
Tan, E., Calabrese, B., Kang, A., & O’Neill, H. (2013). Desiring a career in STEM-related fields: How
middle school girls articulate and negotiate identities-in-practice in science. Journal of Research in Science
Teaching, 50(10), 1143–1179.
Tal, T., & Dierking, L. D. (2014). Learning science in everyday life. Journal of Research in Science
Teaching, 51(3), 251–259.
Tal, T., & Morag, O. (2013). A longitudinal study of environmental and outdoor education: A cultural
change. Journal of Research in Science Teaching, 50(9), 1019–1046.
Journal of Research in Science Teaching
436 LAMB, AKMAL, AND PETRIE
van Dinther, M., Dochy, F., & Segers, M. (2011). Factors affecting students’ self-efficacy in higher
education. Educational Research Review, 6(2), 95–108.
van Eijck, M., & Roth, W. M. (2011). Cultural diversity in science education through novelization:
Against the epicization of science and cultural centralization. Journal of Research in Science Teaching,
48(7), 824–847.
van Gog, T., & Rummel, N. (2010). Example-based learning: Integrating cognitive and social-cognitive
research perspectives. Educational Psychology Review, 22(2), 155–174.
Vedder-Weiss, D., & Fortus, D. (2013). School, teacher, peers, and parents’ goals emphases and
adolescents’ motivation to learn science in and out of school. Journal of Research in Science Teaching, 50(8),
952–988.
Vedder-Weiss, D., & Fortus, D. (2011). Adolescents’ declining motivation to learn science: Inevitable or
not? Journal of Research in Science Teaching, 48(2), 199–216.
Wang, X. (2013). Why students choose stem majors motivation, high school learning, and
postsecondary context of support. American Educational Research Journal, 50(5), 1081–1121.
Wright, B. D., & Stone, M. H. (2004). Making measures. Chicago, IL, Phaneron Press.
Wu, H. K., & Shah, P. (2004). Exploring visuospatial thinking in chemistry learning. Science Education,
88(3), 465–492.
Yassa, M. A., Mattfeld, A. T., Stark, S. M., & Stark, C. E. (2011). Age-related memory deficits linked to
circuit-specific disruptions in the hippocampus. Proceedings of the National Academy of Sciences, 108(21),
8873–8878.
Zeineddin, A., & Abd-El-Khalick, F. (2010). Scientific reasoning and epistemological commitments:
Coordination of theory and evidence among college science students. Journal of Research in Science
Teaching,47(9), 1064–1093.
Zeldin, A. L., & Pajares, F. (2000). Against the odds: Self-efficacy beliefs of women in mathematical,
scientific, and technological careers. American Educational Research Journal, 37(1), 215–246.
Zimmerman, B. (1989). A social cognitive view of self-regulated academic learning, Journal of
Educational Psychology, 81(3), 329–339.
Zimmerman, B. J., & Schunk, D. H. (2012). Handbook of self-regulation of learning and performance
(Educational psychology handbook). New York: Routledge.
Supporting Information
Additional supporting information may be found in the online version of this article at the
publisher’s web-site.
Journal of Research in Science Teaching
DEVELOPMENT OF A COGNITION-PRIMING MODEL 437
