Content uploaded by Michael F Dahlstrom
Author content
All content in this area was uploaded by Michael F Dahlstrom on May 05, 2020
Content may be subject to copyright.
This article was downloaded by: [Iowa State University], [Michael F. Dahlstrom]
On: 04 September 2012, At: 09:24
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Media Psychology
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/hmep20
The Persuasive Influence of Narrative
Causality: Psychological Mechanism,
Strength in Overcoming Resistance, and
Persistence Over Time
Michael F. Dahlstrom a
a Greenlee School of Journalism and Communication, Iowa State
University, Ames, Iowa, USA
Version of record first published: 04 Sep 2012
To cite this article: Michael F. Dahlstrom (2012): The Persuasive Influence of Narrative Causality:
Psychological Mechanism, Strength in Overcoming Resistance, and Persistence Over Time, Media
Psychology, 15:3, 303-326
To link to this article: http://dx.doi.org/10.1080/15213269.2012.702604
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
Media Psychology, 15:303–326, 2012
Copyright ©Taylor & Francis Group, LLC
ISSN: 1521-3269 print/1532-785X online
DOI: 10.1080/15213269.2012.702604
The Persuasive Influence of Narrative
Causality: Psychological Mechanism,
Strength in Overcoming Resistance,
and Persistence Over Time
MICHAEL F. DAHLSTROM
Greenlee School of Journalism and Communication, Iowa State University, Ames, Iowa, USA
The impact of narrative persuasion depends on the location of its
persuasive information relative to the cause-and-effect structure
within the narrative, yet, the bounds of this structural influence
remain unknown. This study examines the a) underlying psycho-
logical mechanism, b) strength in overcoming psychological resis-
tance, and c) persistence over time of narrative causality effects
on information acceptance. Results suggest causality effects occur
during initial stages of comprehension, which serve to shield the
influence from external moderators, such as preexisting world-
views. The effect also remained constant over a two-week delay.
Results serve to psychologically explain the narrative causality
effect and suggest it remains robust over a wide range of condi-
tions, potentially being useful for persuasion of otherwise resistant
audiences.
Narratives hold a particular power over how individuals perceive the world.
By explaining the actions and intentions of characters over time, narratives
mirror everyday thought (Schank & Abelson, 1995) and are often the tactic
of choice employed by communication campaigns to promote prosocial
behaviors, ranging from family planning and AIDS prevention techniques
(Slater & Rouner, 2002; Yoder, Hornik, & Chirwa, 1996) to spousal abuse and
domestic violence (Usdin, Singhal, Shongwe, Goldstein, & Shabalala, 2004).
Even the Centers for Disease Control and Prevention (2011) have begun
working with Hollywood to monitor the truthfulness of medical information
in television dramas.
Address correspondence to Michael F. Dahlstrom, Greenlee School of Journalism and
Communication, Iowa State University, 215 Hamilton Hall, Ames, IA 50010, USA. E-mail:
mfd@iastate.edu
303
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
304 M. F. Dahlstrom
The field of narrative persuasion seeks to better understand this persua-
sive influence of narratives. A recent article claimed that narrative persuasion
in general has overlooked potentially meaningful variance within narratives
by treating the narrative itself as a unit of analysis (Dahlstrom, 2010). By
manipulating the internal cause-and-effect structure of narrative, Dahlstrom
(2010) was able to show that information that caused future events in a
narrative to occur was more persuasive than the same information within
the same narrative when it had no causal impact on the narrative’s events.
Previous research from advertising and law contexts also support such an
influential role of information related to the plot of a narrative (Russell, 2002;
Voss, Wiley, & Sandak, 1999).
While the existence of such a narrative causality effect provides an
example of overlooked variance within narratives, the bounds of influence
regarding this effect remain untested. In response, the purpose of this study
is to move beyond showing the existence of meaningful, within-narrative
variance to exploring its a) underlying psychological mechanism, b) strength
in overcoming psychological resistance, and c) persistence over time. In
addition, these objectives can also provide recommendations as to whether
communication planners should consider leveraging the internal causal struc-
ture when crafting persuasive narratives. To begin exploring these objectives,
it will be useful to contrast the theoretical foundations between narrative
causality and current models of narrative persuasion.
NARRATIVE PERSUASION
Narrative persuasion investigates how narratives, often from fictional enter-
tainment media, can persuade about real world topics. Results from attitude-
based studies suggest that individuals are often more willing to accept nor-
mative evaluations from narratives than factual arguments (Slater & Rouner,
2002; Strange, 2002) and a fictional label does little to reduce the impact
of a narrative (Busselle & Bilandzic, 2008; Green & Brock, 2000). Results
from belief-based studies, which examine the acceptance of specific, factual
assertions made within narratives and their incorporation into mental belief
structures about the world, generally find that individuals do tend to accept
narrative assertions and utilize them to answer questions about the world
(Appel & Richter, 2007; Dahlstrom, 2010; Marsh, Meade, & Roediger, 2003).
Several models have been proposed to explain the persuasive influence
of narratives, including the transportation-imagery model (Green & Brock,
2000), extended elaboration likelihood model (Slater & Rouner, 2002), and
entertainment overcoming resistance model (Moyer-Gusé, 2008), all which
share the belief that transportation, or engagement with the narrative, as well
as identification with characters serve to increase persuasive impact through
reducing the formation of counterarguments, lessening message scrutiny,
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 305
and inhibiting psychological resistance. Based on these processes, narrative
persuasion is often considered a covert mode of persuasion, where infor-
mation is generally accepted first and only scrutinized later with additional
cognitive effort. Supporting this description, narrative persuasion has been
found to lose much of its influence once the reader becomes aware of the
persuasive intent (Moyer-Gusé, 2008; Weinstein, Grubb, & Vautier, 1986).
These current models of narrative persuasion are generally extensions of
the earlier theories of flow (Csikszentmihalyi, 1991), social cognitive theory
(Bandura, 1986, 2002), and the elaboration likelihood model (Cacioppo &
Petty, 1984). Following this tradition, the narrative message is often treated
as the unit of analysis, either by comparing the effectiveness of narrative
to nonnarrative messages (de Wit, Das, & Vet, 2008; Mazor et al., 2007) or
by measuring factors associated with narrative messages as a whole, such
as transportation, emotional involvement, or perceived realism (Busselle
& Bilandzic, 2008; Morgan, Movius, & Cody, 2009; Moyer-Gusé & Nabi,
2010). Such a level of theoretical understanding is essential for exploring the
persuasive influence of narrative, but is limited in the specificity of questions
that may be answered.
NARRATIVE CAUSALITY
In contrast to narrative persuasion, narrative causality arises from the the-
oretical background of discourse psychology that treats factors within the
narrative message as the unit of analysis. Narrative causality describes the
relationship between the internal cause-and-effect structure of a narrative
and the acceptance of assertions placed within that narrative. It was found
that assertions placed at locations that served to change the course of action
within the narrative, termed causal locations, were accepted as more truthful
then the same assertions placed at locations that had no impact on future
narrative events, termed noncausal locations (Dahlstrom, 2010).
Discourse psychology investigates how the mind comprehends text and
forms mental representations of its content. Research suggests three levels
of comprehension when reading a narrative: the surface code, textbase, and
situation model (Graesser, Millis, & Zwaan, 1997; Graesser, Olde, & Klettke,
2002). The surface code represents the depiction of the actual letters of the
text in the mind. The surface code is quickly converted to the textbase,
which represents the propositional meanings underlying the words them-
selves. These propositions are then combined with preexisting knowledge to
generate a complex mental depiction of the narrative world called a situation
model. While strings of text may remain memorable in certain circumstances,
individuals do not react directly from communicated information but, rather,
indirectly through the perceived meaning contained in their individually
created situation models (van den Broek & Gustafson, 1999). Therefore, pre-
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
306 M. F. Dahlstrom
dicting how narratives influence beliefs requires an understanding of how the
mind combines the a) content present in the narrative text with b) preexisting
knowledge to form a complex understanding of the narrative world.
To build a situation model, individuals must first identify connections
within the textbase in a bottom-up fashion (van den Broek & Gustafson,
1999). This construction is largely driven by a search for cause-and-effect
relationships. The investigation of causality has been shaped in large part
by the causal network model (Trabasso & Sperry, 1985), which claims that
causal relations represent the glue that holds narratives together (Magliano,
1999). Studies using the causal network model identify causal relations using
the logical criterion of necessity (Mackie, 1980), which states that element A
is considered causally related to element B if element B could not occur in
the narrative without element A. Narrative statements with a greater number
of causal connections are easier to recall, are recalled faster, and receive
higher ratings of importance to the narrative (Trabasso & Sperry, 1985).
As these cause-and-effect relationships are organized into a cohesive
structure, preexisting knowledge is used to fill any gaps not explicitly stated
in the narrative text by generating inferences (Magliano, 1999). These gaps
could represent severe breaks in continuity, but more commonly refer to the
shared sociocultural understanding an author assumes he or she need not
state directly. For example, an author may describe a character driving to a
destination without explaining how the character operates the car. It is up to
the audience to fill such a gap by generating an inference from preexisting
knowledge.
Inferences are largely driven by a need to explain cause-and-effect
relationships that may not be fully stated in the text (Graesser et al., 2002;
Magliano, 1999), and this desire for causal coherence has been found to be
nearly invariant among individuals, genres, and reading situations (Graesser
& Wiemer-Hastings, 1999). Other research suggests that individuals rely mainly
on information from the narrative text to explain gaps within the cause-and-
effect structure but rely to a greater degree on preexisting knowledge to fill
other gaps not related to causality, such as setting or normative expecta-
tions (Magliano, 1999). This difference was observed in sentence reading
time where causal information was read faster than noncausal informa-
tion because mentally searching through previous narrative information was
suggested to be less time-consuming that searching through vast stores of
preexisting knowledge (Magliano, 1999). Models such as the construction-
integration model (Kintsch, 1998), constructionist model (Graesser, Singer,
& Trabasso, 1994), event-indexing model (Zwaan, Langston, & Graesser,
1995), and resonance model (O’Brien & Myers, 1999) all agree that preex-
isting knowledge is necessary for building situation models and can lead to
differences in how individuals comprehend narrative texts.
In summary, the primary organizing factor the mind uses to comprehend
narrative text is its cause-and-effect structure, the search for which remains
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 307
primarily within the given narrative text. The gaps in comprehension not
fulfilled by this causal structure are more likely to be filled by inferences
generated from preexisting information.
STUDY OBJECTIVES
The purpose of this study is to explore the bounds of influence regarding
narrative causality on information acceptance by investigating its a) under-
lying mechanism, b) strength in overcoming psychological resistance, and
c) persistence over time.
The initial processes of narrative comprehension likely influence the
mechanism underlying the narrative causality effect. To comprehend a narra-
tive message, the mind must first identify cause-and-effect relations within the
propositional meanings of the text and use those relations to build a growing
framework for the mental situation model. To fill gaps within this causal
framework, the mind then generates inferences to permit comprehension of
the narrative. However, by forcing the mind to actively validate the causal
framework through inference generation before comprehension is complete,
the legitimacy of the causal relationships may be bolstered even before
evaluation of the narrative is possible. This procedural legitimacy may lead
to greater perceived truthfulness, or relevance of the information to the real
world, which is typically the desired outcome for most belief-based examples
of narrative persuasion. Such a relationship would therefore be mediated
by the more intense information processing allocated to understanding the
causal structure of a narrative. One indicator that could be used to reflect
this cognitive process is recall. Increased recall has already been associated
with causal information within a narrative (Trabasso & Sperry, 1985) and can,
therefore, serve as a proxy measure for intensity of information processing
during comprehension.
H1: Using recall as an indicator of information processing during compre-
hension, the relationship between causal location and perceived truth-
fulness will be mediated by recall, such that increased recall indicates
increased perceptions of truthfulness.
Because of the contrasting theoretical foundations between narrative
causality and current models of narrative persuasion, their underlying mech-
anisms may be fundamentally different. Narrative causality is based on the-
ories of comprehension that examine factors within the narrative message
as the unit of analysis while most models of narrative persuasion are based
on theories that treat the entire narrative message as the unit of analysis.
Therefore, the factors that moderate and mediate narrative persuasion may
be dependent on the unit of analysis used.
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
308 M. F. Dahlstrom
In the case of psychological resistance, for instance, narrative informa-
tion may not lead to attitude change if preexisting belief structures strongly
contradict the content of the narrative (Slater, Rouner, & Long, 2006). If the
entire narrative was treated as the unit of analysis, one could measure if a
narrative that contains information that contradicts existing belief structures
leads to more attitude change than a narrative that did not. Likewise, because
greater transportation, or engagement with a narrative, is often found to
increase narrative persuasion (Green & Brock, 2000), one could also mea-
sure if the transportation activated by either narrative might moderate such
rejection.
However, when the unit of analysis moves inside the narrative, the
analogous question becomes what aspects within a single narrative result
in greater attitude change or greater transportation. Because Hypothesis 1
proposes that the identification and mental validation of causal information
occurs before preexisting knowledge enters the comprehension process,
psychological resistance would be less likely to influence the acceptance
of causal statements. In fact, most of the global moderators associated with
narrative persuasion, including both preexisting beliefs and transportation,
would likely exert their influence downstream of causal effects, somewhat
insulating the effects of narrative causality from moderators external to the
comprehension process. Therefore, such a relationship would predict that
the moderating effects of preexisting beliefs and transportation would not yet
exert their influence when treating causality as the internal unit of analysis
and would only appear when treating the entire narrative as the unit of anal-
ysis. Likewise, the proposed mediating effect hypothesized in Hypothesis 1
would likely be reduced when treating the entire narrative as the unit of
analysis as it must compete with the additional moderating factors.
The preexisting belief structure chosen for this study will be pro-environ-
mental worldviews. Perceptions of the environment have been said to be
especially susceptible to the effects of media influence (McComas & Shana-
han, 1999) and do not necessarily mirror reality (Ader, 1995), leading to a
potentially large moderation. Likewise, there exists a generally well-tested
scale for the measurement of pro-environmental worldviews (Dunlap, Van
Liere, Mertig, & Jones, 2000). Finally, previous experiments using a similar
participant pool suggests that the scores will cluster on the high end of the
scale, assisting in stimulus and assertion creation.
H2: The moderating effects of pro-environmental worldviews and trans-
portation will be stronger when treating the narrative as a unit of analysis
than when treating causality as the internal unit of analysis.
H3: When treating the narrative as the unit of analysis, the mediating effect
of recall between narrative causality and perceived truthfulness will be
reduced.
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 309
Finally, the persistence of the narrative causality effect over time remains
unknown. Recall is well known to decrease rapidly over time as newer
information takes the place of prior information (Oberauer & Kliegl, 2006)
and while most studies of narrative persuasion focus on immediate effects,
there is a growing body of research examining the delayed effect of narrative
persuasion (Appel & Richter, 2007; Jensen, Bernat, Wilson, & Goonewar-
dene, 2011; Marsh et al., 2003). Focusing on belief-based studies that mirror
the current methodology of measuring the acceptance of assertions, research
suggests that fictional narrative information is often incorporated into pre-
existing knowledge but also marked with a source cue labeling the content
as coming from a potentially misleading source (Marsh et al., 2003). The
delayed effect of narrative persuasion may decrease over time (Marsh et al.,
2003) or potentially increase if the source cues are forgotten and the content
becomes more associated with previous knowledge (Appel & Richter, 2007).
Research into the delayed effects of narrative persuasion has thus far
used narrative as the unit of analysis and, therefore, incorporating the pro-
posed within-narrative factors of causality introduces more complexity with
little precedent on which to form hypotheses. Nonetheless, it is possible to
suggest two contrasting relationships based on the existing research and the
proposed influences from this study.
The first relationship states that because recall is proposed to indicate the
mediated relationship between causality on perceived truthfulness, they will
remain coupled and perceived truthfulness will follow recall as it decreases
over time. While both causal and noncausal information would decrease
under this proposed relationship, the causal information would likely lose
its advantage and begin to converge with that of the noncausal information.
However, situation models can exist in memory for years and are re-
membered over the narrative text itself (Bower & Morrow, 1990). There-
fore, the contrasting relationship suggests that recall may merely indicate
the mechanism of comprehension at the immediate time of exposure, and
may de-couple once the situation model is complete, leaving the different
effects of causality fixed for a longer period of time. Under this proposed
relationship, the perceived truthfulness of the causal information would have
been incorporated into existing knowledge and remain fairly stable as recall
decreased over time. The noncausal information would either also remain
stable at its reduced level of impact or decrease over time if portions of it
were not incorporated into existing knowledge during exposure.
If the former relationship is supported, the influence of narrative causal-
ity would be brief and not offer practical use for communication planners. If
the latter relationship is supported, the influence of narrative causality may
be long lasting and represent a promising tool for communication planners.
RQ1. How will the relationship between causal location and perceived truth-
fulness change over time?
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
310 M. F. Dahlstrom
METHODS
Participants
One hundred one undergraduate students served as participants for this
study and were offered extra credit as compensation. The mean age was
19.91 (SD D1.73) and the group consisted of 68% females. These subjects
were contacted two weeks after completing the study and asked to complete
a follow-up questionnaire for additional extra credit. Ninety-four participants
successfully completed the follow-up questionnaire, representing a 93% re-
turn rate. Participants were asked if they had researched any of the assertions
between experiments. Two participants replied yes and were removed from
the analysis, resulting in a sample size of 92. The mean age of the second
participant pool was 19.95 (SD D1.75) and 66% females.
Experimental Design and Procedure
This study used a two-factor, within-subjects design consisting of three-
levels of assertion placement with regard to a stimulus narrative: a) causal,
b) noncausal, and c) absent, and two levels of time: a) immediate and
b) delayed. Potential moderators consisted of two between-subject fac-
tors: transportation and pro-environmental worldviews. Participants were
informed that they would be involved in a study addressing how information
is presented in the media and would be asked to read a narrative and answer
questions about their opinion of what they read. Participants were initially
contacted by email and provided with a Web link to an online experiment
accessible from any computer with Internet access. They were randomly
assigned by an HTML algorithm to read one of three stimulus narratives.
After reading, they were presented with a set of questions measuring the
between-subject moderators and demographics. Participants were finally
presented with a set of questions designed to measure their levels of cued-
recall and perceived truthfulness regarding assertions from the narrative.
The moderator and demographic questions did not refer to assertions from
the narrative and were, therefore, asked before the recall questions to
serve as mental distraction to lessen the effect of recency on the recall
measure.
Two weeks after completion of the previous questionnaire, the par-
ticipants were requested by e-mail to complete a follow-up questionnaire
regarding their opinions about the material previously viewed. Participants
were presented with the same set of questions measuring cued-recall and
perceived truthfulness. Participants were finally asked if they had investigated
any of the assertions between experiments, debriefed about the truth or
falsity of the assertions, and thanked for their time.
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 311
Preexisting Contradictory Beliefs
All measures of pro-environmental worldviews were obtained by use of the
new ecological paradigm scale (Dunlap et al., 2000). A high score on this
scale represents the existence of a strongly held set of pro-environmental
beliefs and the assertions manipulated in this study would represent a chal-
lenge to this preexisting belief structure. In contrast, a low score on the new
ecological paradigm scale represents the existence of a strongly held set of
anti-environmental beliefs where the assertions manipulated in this study
would have little conflict with preexisting belief structures.
Stimulus
The stimulus consisted of a fictional narrative in the style of a short story that
would be read for entertainment. Creation of the stimulus narrative required
the development of a) a list of assertions and b) a template narrative within
which the assertion locations could be manipulated.
The assertions used in this experiment had to meet certain require-
ments. First, the content of each assertion needed to consist of a familiar
environmental issue about which the participant would most likely have
preexisting beliefs. Second, each assertion needed to challenge its respec-
tive preexisting belief. Finally, each assertion needed to be false to avoid
potential effects with previous knowledge. An alternative method would
have been to measure and then control for previous knowledge, but this
introduces potential priming effects that could be avoided by using false
assertions.
Therefore, a pool of 17 false, anti-environment assertions was created
to fulfill these requirements, such as, ‘‘More energy is saved by leaving
your computer on overnight than by turning it off and starting it back
up in the morning,’’ and ‘‘Urban and agricultural areas currently account
for less than 15% of available land area.’’ Six individuals of differing pro-
environmental worldviews were recruited to test for validity. Three individ-
uals came from environmental occupations and held extremely strong pro-
environmental worldviews. Three individual came from nonenvironmental
occupations and held weak to average pro-environmental worldviews. All
six individuals rated the pool of assertions regarding perceived truthfulness
and the 12 assertions receiving both the lowest rating of truthfulness by
those with strong pro-environmental worldviews and the largest differences
between the two groups were selected. This pretest helps to ensure that the
selected assertions are not just false, but also perceived as false relative to
the preexisting beliefs of interest. Four pro-environmental assertions were
also identified by the individuals and included into the narrative to help
disguise the persuasive intent of the narrative and avoid the appearance of
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
312 M. F. Dahlstrom
a completely anti-environmental narrative, as narrative persuasion has been
found to decrease if the persuasive intent becomes salient (Moyer-Gusé &
Nabi, 2010). All four pro-environmental assertions were not manipulated nor
included in any analyses.
The template narrative into which these assertions could be manipulated
also had to meet certain requirements. First, the structure of the template
narrative needed to contain multiple slots, each of which could be filled with
various assertions without disrupting the logic of the narrative. Second, the
topic of the template narrative also needed to conceal its persuasive attempt
for the same reason as with the assertions; the narrative needed to state
anti-environmental assertions without the theme of the narrative relating to
those assertions. Finally, to align with common stimuli in existing narrative
persuasion research, the template narrative needed to be structured as a
fictional short story written primarily for entertainment purposes.
To meet these requirements, the template narrative told the story of
Lucas, a science center employee in competition with a nearby museum to
attract more visitors to its Earth Day exhibits. Lucas was in charge of creating
the exhibits and he repeatedly focused on the artistic presentation of each
exhibit at the expense of scientific accuracy. His competitor attempted to
reveal these inaccuracies, thereby increasing traffic to his competing mu-
seum, by handing out flyers of environmental facts that always contradicted
Lucas’s exhibit. After continuously being reprimanded by the director for
his incorrect environmental facts, Lucas finally accepted the importance of
accuracy and vowed to improve his performance for future exhibits. The
neutral setting of a museum was selected over a more controversial en-
vironmental setting to better hide the persuasive intent, and the conflict
between Lucas and his competitor allowed for environmental assertions to
be inserted according to the logic of the narrative without moralizing about
environmental themes.
Experimental Manipulation
Causal placement was manipulated by inserting assertions at one of three
locations relative to the narrative: causal, noncausal, or absent. Causal loca-
tions represent events that lie on the main cause-and-effect chain within the
narrative and, therefore, cause future events to occur. Noncausal locations
represent statements that lie off the main cause-and-effect chain within the
narrative and have no impact on future events. Absent locations represent a
nonnarrative control and were not included in the text of the narrative.
The template narrative was written to contain four causal and four non-
causal slots that could be logically filled with various assertions. To account
for any differences of memorability or perceived truthfulness between the
assertions, three versions of the narrative were created to allow each assertion
to be measured in each location.
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 313
To illustrate, at one point in the story, Lucas creates an exhibit showing
the dangers of water pollution. The exhibit displays a large fish kill, long-term
damage from an oil spill and massive amounts of chemical runoff dumping
into rivers. Visitors then enter the museum carrying flyers of environmen-
tal facts distributed by Lucas’s competitor. Lucas overhears visitors reading
assertions from their flyers not related to water pollution. These assertions
represent noncausal slots because their declaration in the narrative does not
cause Lucas or any other character to alter their actions. However, one of
the causal slots occurs when Lucas overhears a visitor read an assertion
that is related to water pollution but does not align with his exhibit. The
visitor complains about this discrepancy, which motivates Lucas to change
his course of action and address the shortcoming in the following exhibit.
This particular excerpt from the narrative is provided in the Appendix.
Associated with this particular causal slot were three assertions that the
visitor could read to contradict the exhibit and serve as a causal assertions:
a) Fish populations are now increasing in the wild; b) the effects of an oil spill
typically last less than a one year; and c) over 90% of runoff contaminants are
eliminated by microbes before entering water systems. In the first version of
the narrative, the first assertion was placed into this causal slot, the second
assertion was placed into a noncausal slot later in the narrative when the
water pollution exhibit was no longer the focus and, thus, had no impact on
the narrative’s events. The third assertion was left out of the narrative as a
control.
The second version of the narrative rotated the order so that the second
assertion was placed into this causal slot, the third assertion was placed into
the noncausal slot and the first assertion was left out of the narrative. The
third version completed this rotation so that each assertion was measured
at each possible location. The final narrative contained four causal and four
noncausal slots and these slots alternated to control for order effects. The
final versions of the narrative consisted of 2,974, 2,986, and 2,965 words.
The three versions of the narrative were created to control for variabil-
ity in the assertions and do not represent meaningful manipulations in of
themselves. Of interest is comparing the within-subject influence of causal
category regardless of the assertions that were placed in those locations.
Dependent Measures
RECALL
Participants were given the starting word or phrase of each assertion as a cue
and prompted to recall the full assertion in an open-ended text field. Recall
was measured after demographic and environmental worldview questions
to limit the effect of recency. Responses were coded as either correct or
incorrect based on similarity to the actual assertion. Two coders were used
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
314 M. F. Dahlstrom
to ensure unbiased coding. One coder rated all responses for each of the
12 assertions and a second coder rated 20% of the responses for each
assertion to test for reliability. All measures of Cohen’s Kappa between the
two coders were found to be above 0.84. After ensuring reliability, the first
coder’s scores were used to calculate all recall variables.
Recall for each of the three possible causal categories was calculated as
the average of the four specific assertions an individual saw that represented
those categories in their narrative: recall of causal assertions (MD0.52,
SD D0.33), recall of noncausal assertions (MD0.21, SD D0.26), and recall
of assertions not present (M<0.01, SD D0.03). Because the coding values
were either 0 or 1, a participant’s average recall measure also represents the
percent of correctly recalled assertions (MD0.25, SD D0.17). Average recall
measures were not significantly different between versions of the stimulus
narrative, F(2, 95) D0.21, pD.81, 2<0.01.
DELAYED RECALL
Delayed recall was calculated in the same manner at the second time point:
delayed recall of causal assertions (MD0.28, SD D0.29), delayed recall of
noncausal assertions (MD0.11, SD D0.18), and delayed recall of assertions
not present (MD0.02, SD D0.07).
It is important to note that the control situation asked participants to
recall information they had not seen. Responses were therefore almost uni-
formly zero for this control group because anything other than the correct
assertion was coded as incorrect. While this created a needed baseline for use
in the mediation analysis, it also resulted in significantly less variance. In or-
der to correct for this violation of the assumption of homogeneity of variance,
all measures of recall were transformed using an arcsin transformation to
stabilize the variances (Laubscher, 1961). While these transformed measures
were used in all analyses regarding recall, the untransformed means are
reported because of the greater ease in interpretation.
PERCEIVED TRUTHFULNESS
Participants were asked to rate the likelihood of each assertion being true
in the real world on a 7-point scale with possible responses ranging from
absolutely false, probably false, possibly false, don’t know, possibly true,
probably true, to absolutely true. Responses were coded so larger numbers
represented greater levels of perceived truthfulness (MD4.32, SD D1.08).
The complete assertion was given prior to each measure to ensure perceived
truthfulness was not dependent on the accuracy or completeness of the
previous recall measure.
Perceived truthfulness for each of the three possible causal locations was
calculated as the average of the four specific assertions each individual saw
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 315
that represented those locations in their narrative version: perceived truth-
fulness of causal assertions (MD4.85, SD D1.58), perceived truthfulness of
noncausal assertions (MD4.48, SD D1.35), and perceived truthfulness of
assertions not present (MD3.66, SD D0.84). Measures of average perceived
truthfulness were not significantly different between versions of the stimulus
narrative, F(2, 88) D1.49, pD.23, 2D0.03.
DELAYED PERCEIVED TRUTHFULNESS
Delayed perceived truthfulness was calculated in the same manner from
the measures at the second time point: delayed perceived truthfulness of
causal assertions (MD4.82, SD D1.43), delayed perceived truthfulness
of noncausal assertions (MD4.36, SD D1.21), and delayed perceived
truthfulness of assertions not present (MD3.86, SD D1.10).
Care was taken to ensure that there was no confusion as to the fictional
status of the narrative. The narrative was also always referred to as a fictional
short story and its fictional status was highlighted once again before exposure
to the questions of perceived truthfulness.
Moderators
Strength of pro-environmental worldview was measured using the 15-item
revised New Ecological Paradigm scale (Dunlap et al., 2000; MD4.82, SD D
0.65, ˛D0.76). Transportation was measured using the 12-item scale (Green
& Brock, 2000; MD4.53, SD D0.75, ˛D.79).
Manipulation Check
Narrative persuasion is reduced when the persuasive intent become salient.
Therefore, perceived bias was collected as a manipulation check to measure
if the participants perceived the narrative to be pushing an agenda. Partici-
pants were asked their agreement with the following statement after all other
questions: ‘‘The specific facts used in the story seemed well-rounded and free
of bias.’’ Responses were coded so larger numbers represented greater levels
of agreement (MD4.75, SD D1.47). A regression test suggests the relation-
ship between perceived bias and strength of pro-environmental worldviews
was not significant, t(79) D0.88, pD.38, suggesting the persuasive intent
of the narrative was successfully hidden.
RESULTS
Hypothesis 1 claimed that the relationship between narrative causality and
perceived truthfulness would be mediated by information processing, which
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
316 M. F. Dahlstrom
could be indicated by recall. Following the Judd method1for within-subject
mediation and moderation tests ( Judd, Kenny, & McClelland, 2001) a set of
two orthogonal contrasts were defined for recall and perceived truthfulness:
causal versus noncausal (1, 1, 0) and narrative versus nonnarrative control
(1, 1, 2). The former contrast measures differences between causal and
noncausal locations and represents treating causality as the unit of analysis
within the larger narrative message. The latter contrast measures differences
between information presented in the narrative (by combining the causal and
noncausal factors) and information absent from the narrative and represents
treating the narrative itself as the unit of analysis.
Each perceived truthfulness contrast was regressed upon the sum of
recall and the full set of recall contrasts. Using this method, mediation is
indicated if two conditions are met: a) there exist significant main effect
differences in both outcome variable and potential mediator and b) the
recall contrast corresponding to the selected perceived truthfulness contrast
is significant.
To test the first requirement, both recall and perceived truthfulness were
tested with a repeated-measure analysis of variance (ANOVA) analysis. For
perceived truthfulness, the omnibus test between the three within-subject
levels of causal location was significant, F(2, 180) D47.39, p<.01, 2D0.35.
Regarding individual contrasts, the mean of the perceived truthfulness of
assertions placed at causal locations (MD4.85, SD D1.58) was significantly
higher than the assertions placed at noncausal locations (MD4.48, SD D
1.35), F(1, 90) D12.43, pD0.01, 2D0.12, and both causal and noncausal
assertions were significantly perceived as more truthful than the control
assertions not present in the narrative (MD3.66, SD D0.84), F(1, 90) D
75.32, p<0.01, 2D0.46, F(1, 90) D41.00, p<0.01, 2D0.31. For recall, the
omnibus test between the three within-subject levels of causal location was
also significant, F(2, 194) D153.91, p<.01, 2D0.61. Regarding individual
contrasts, the mean of the recall of assertions placed in causal locations
(MD0.52, SD D0.33) was significantly higher than the assertions placed in
noncausal locations (MD0.21, SD D0.26), F(1, 97) D85.32, p<.01, 2D
0.47, and both causal and noncausal assertions were recalled significantly
more than the control assertions (M<0.01, SD D0.03), F(1, 97) D276.02,
p<.01, 2D0.74, F(1, 97) D77.61, p<.01, 2D0.44. Therefore, the first
requirement for mediation is met, such that there are significant main effect
differences in the same direction for both variables, as shown in Figure 1.
Because the predicted mediation effect regards the difference between
causal and noncausal locations, testing the first orthogonal mediation con-
trast for significance will provide evidence of the second requirement for
mediation. The corresponding recall contrast was predictive of the selected
perceived truthfulness contrast, t(87) D2.91, p>.01, suggesting recall does
serve as a significant mediator between causal location within the narrative
and perceived truthfulness. Therefore, Hypothesis 1 was supported.
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 317
FIGURE 1 Cued recall and perceived truthfulness of narrative assertions at different causal
locations.
It was also hypothesized in Hypothesis 2 that the moderating effects of
both preexisting belief structures and transportation would be present when
treating the narrative as a unit of analysis but not when treating causality as
the unit of analysis. To test for moderation between repeated measures and a
continuous variable, the previously described orthogonal mediation contrasts
were regressed on the set of continuous moderators. A test of whether the
slope of the moderator differs from zero provides the test of moderation
(Judd et al., 2001).
When treating causality as the unit of analysis by using the first orthog-
onal mediation contrast, neither transportation nor strength of pro-environ-
mental worldview moderated the relationships between treatment and recall,
t(75) D1.32, pD.19, t(75) D 0.34, pD.73, or perceived truthfulness,
t(69) D1.15, pD.26, t(69) D 1.94, pD.06, respectively.
When treating the narrative as the unit of analysis by using the second
orthogonal mediation contrast, both transportation, t(75) D2.35, pD.02, and
strength of pro-environmental worldview, t(75) D2.47, pD.02, significantly
moderated the relationship between treatment and recall, such that both
increased transportation and increased strength of pro-environmental world-
view led to greater recall. Strength of pro-environmental worldview also
moderated the relationship between treatment and perceived truthfulness,
t(69) D 3.20, p<.01, such that increased strength of pro-environmental
worldview led to smaller levels of perceived truthfulness. Transportation was
not a significant moderator for this path of the model, t(69) D1.58, pD.12.
Because moderation was stronger when treating the entire narrative as the
unit of analysis, these data lend support to the second hypothesis that these
moderation effects occur downstream from causality effects.
Because of this difference in moderation effects, it was hypothesized
in Hypothesis 3 that the mediation of the relationship between narrative
causality and perceived truthfulness by recall would be reduced when treat-
ing the entire narrative as the unit of analysis. Because the first requirement
for mediation has already been demonstrated, testing the second perceived
truthfulness orthogonal mediation contrast will provide evidence of the sec-
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
318 M. F. Dahlstrom
ond requirement for mediation. The corresponding recall contrast was not
predictive of the selected perceived truthfulness contrast, t(87) D1.34, pD
.18, showing no evidence of mediation when the narrative is treated as the
unit of analysis. Therefore, Hypothesis 3 was also supported. The results for
the previous hypotheses are modeled in Figure 2.
The question regarding the change of narrative causality effects over
time was asked in Research Question 1 and was predicted by two contrasting
hypotheses: a) recall and perceived truthfulness would remain coupled and
decrease over time or b) recall and perceived truthfulness would decouple
and perceived truthfulness would remain stable over time as recall decreased.
The change in recall and perceived truthfulness for both causal and non-
causal locations were tested by repeated-measure ANOVA analysis with post
hoc Bonferroni adjustments.
The change in recall over time significantly decreased between both
immediate causal (MD0.53, SD D0.32) and delayed causal (MD0.28,
SD D0.29), F(1, 91) D61.56, p<.01, 2D0.40, and immediate noncausal
(MD0.23, SD D0.26) and delayed noncausal (MD0.11, SD D0.18), F(1,
91) D26.25, p<.01, 2D0.22, suggesting recall decreased over time as
expected.
The change in perceived truthfulness over time did not significantly
change between immediate causal (MD4.92, SD D1.54) and delayed causal
FIGURE 2 Contrasting mediating and moderating relationships between treating narrative as
the unit of analysis and treating causality as a within-narrative unit of analysis.
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 319
FIGURE 3 Change in cued recall and perceived truthfulness of narrative assertions at
different causal locations over a delay of two weeks.
(MD4.80, SD D1.43), F(1, 82) D0.84, pD.36, 2D0.01, or immediate
noncausal (MD4.47, SD D1.37) and delayed noncausal (MD4.36, SD D
1.22), F(1, 86) D0.93, pD.34, 2D0.01. In response to the research
question, recall and perceived truthfulness decoupled during the delay with
recall decreasing and perceived truthfulness unchanging over time as shown
in Figure 3.
DISCUSSION
The purpose of this study was to offer both psychological understanding and
practical guidance regarding the influence of narrative causality on informa-
tion acceptance by examining its: a) underlying psychological mechanism,
b) strength in overcoming psychological resistance, and c) persistence over
time.
Results support a proposed mechanism that the more intense cognitive
processing associated with causal statements, indicated in this experiment by
increased recall, inherent in the primary stages of narrative comprehension
indirectly influence greater acceptance of casual information after exposure
to the narrative. This mechanism highlights the important difference between
comprehension and evaluation. The act of comprehension requires accep-
tance for understanding while correction or certification requires subsequent
consideration of what has been understood (Gilbert, 1991). Because causal-
ity represents the primary organizing structure of narrative comprehension,
causal information earns acceptance through the requirements of compre-
hension before progressing to the later stages of persuasion.
Further support of this mechanism comes from the apparent immunity
of narrative causality from the moderating influences of preexisting beliefs
and transportation, both of which should enter the comprehension process
after the influence of causality. This lack of moderation suggests the narrative
causality effect may be fairly strong in overcoming psychological resistance.
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
320 M. F. Dahlstrom
The overall acceptance of information did decrease as strength of contradic-
tory beliefs increased, but the positive influence of causal over noncausal
location did not significantly change over this range.
This proposed mechanism was contrasted with an alternative model
that averaged the two levels of causal location within the narrative and
compared it to the control information that was absent from the narrative.
This comparison served to illustrate the potential differences when treating
the narrative as the unit of analysis rather than treating units within the
narrative as the unit of analysis. This alternative model was much simplified
when compared to current models of narrative persuasion, but served its
purpose of representing a model that ignores variance within the narrative
itself. When the internal causal variance was ignored, recall lost its medi-
ating effect and narrative influence became moderated by both preexisting
beliefs and transportation, again supporting the unique properties of the
original mechanism. It is important to note that the recall data for both
mechanisms were collected after the comprehension process was largely
complete. Therefore, an underlying assumption in this interpretation is that
the more intense cognitive processing given to causal information early in the
comprehension process remains static enough to be detected through recall
measures after the later stages of comprehension are complete. Likewise,
showing significant mediation in one model and nonsignificant mediation in
the other does not ensure that the two are different. The Judd method for
within-subject mediation testing does not permit comparing the two models
for a significant difference (Judd et al., 2001), so the proposed difference
should be treated as suggestive, yet preliminary.
Regarding the influence of narrative causality over time, results show
that even as recall diminished as expected, the perceived level of truth-
fulness associated with causal and noncausal locations remained constant
over a two-week period. This supports the second proposed relationships
that claims that recall indicates the mechanism of comprehension at the
immediate time of exposure but once the situation model is complete and
content has become incorporated into existing knowledge, the differing
impact of causal and noncausal information remains fixed for a longer period
of time. This also suggests that leveraging the influence of causal structure
may represent a promising method of promoting and maintaining specific
beliefs through a mediated message. It must be cautioned, however, that
this study focused on the acceptance of specific assertions from within a
narrative and these results may be greater in magnitude compared to studies
that examine the acceptance of larger attitudes or themes from a narrative.
Whereas single assertions are somewhat context free, the larger attitudes
often measured in narrative persuasion, such as support for gay marriage
or attitudes about abortion, are constructed of an interaction of multiple
accepted assertions and personal experience, which have a greater likelihood
of being activated or updated by a wider range of messages. Therefore, both
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 321
the impact of causal placement in a single narrative and its persistence over
time would likely be less influential for general attitudes than the specific
assertions measured in this study.
While this study only used variations of a single stimulus narrative,
it represents the first replication of the narrative causality effect using a
different stimulus, supporting the generalizability of the effect in general.
Likewise, because the primary questions concerned the within-narrative vari-
able of causality, the stimulus itself represented four exemplars of each
construct of interest, again supporting the generalizability of the current
findings. However, the previous article introducing the narrative causality
effect (Dahlstrom, 2010) briefly explored the role of recall and found only
a marginally significant mediating influence whereas this study found the
relationship to be significant. The difference may be due to almost a two-fold
difference in sample size, but likely also to unmeasured factors associated
with the individual narratives. Future studies should continue to explore
narrative causality using different stimuli in hopes of identifying additional
overlooked variance that can strengthen the model. Likewise, even though
causality is said to be the primary impetus for narrative comprehension,
any textual factor that increases information processing during the early
stages of comprehension may influence its own acceptance. Other factors
under investigation in narrative persuasion that increase memorability, such
as vividness or emotional involvement, may gain influence through the same
mechanism.
Causality in this study was conceptualized as dichotomous: either causal
or noncausal. However, nonexperimental narratives are often more complex
and a strict categorical dichotomy is not ideal. Future studies should explore
developing a continuous measure of causality for a more nuanced under-
standing of narrative causality. Early studies in discourse psychology did
conceptualize causality as continuous, assigning each unit in a narrative a
number of causal connections to other narrative units (Trabasso & Sperry,
1985). However, these narratives were on the order of a few hundred words
and such a methodology quickly becomes prohibitive for larger narratives.
The unit within the narrative would also need to be appropriately scaled
for more complex narratives. Event structuring offers some guidance (Britt,
Perfetti, Sandak, & Rouet, 1999), but the appropriate methodology remains
unclear.
Likewise, causality was operationalized as a factor within the text itself,
but it is also likely that individual differences may play a role in how cause-
and-effect relationships are identified. One individual may see the cause of
a failure within a narrative as due to weakness in some necessary character
trait while another individual may see the cause as due to external factors
within the narrative world. This discrepancy is likely to be larger when a
particular narrative provides fewer specifics and requires more inference
generation. While the stimulus was created to explicitly highlight the cause-
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
322 M. F. Dahlstrom
and-effect relationships, there was no manipulation check to measure if
the participants did perceive the causal slots as representing a cause-and-
effect relationship in their mental situation model. At this psychological level,
the mechanism underlying narrative causality effects would likely hold for
whatever content the individual perceives as causal and future studies that
account for potential differences in causal perceptions due to prior attitudes,
beliefs, and experiences may find an increased effect by accounting for the
variance overlooked in this study.
A secondary purpose of this study was to offer practical guidance for
communication practitioners as to whether narrative causality should be
considered in communication campaigns. Because the influence of narra-
tive causality did not differ across a range of preexisting belief strength,
leveraging narrative causality may offer one technique suited to influenc-
ing otherwise resistant populations. Likewise, the duration of the narrative
causality effects appear long lasting, suggesting potential value for long-
term communication campaigns. In summary, creating a narrative in which
the intended persuasive content causes future events in the narrative rather
than serving as background information or as an aside may offer useful
benefits. However, much like narrative persuasion in general, the influence
of narrative causality is covert and its use may raise ethical considerations
that should be explored, both in practice and in the literature.
NOTE
1. Bootstrapping methods for mediation analysis are generally superior to the criterion meth-
ods that do not allow quantitative comparison of effects between models, such as the Judd
method used in this study (Judd et al., 2001). However, bootstrapping methods are not
substantially developed to answer the within-subject questions posed by this study.
REFERENCES
Ader, C. R. (1995). A longitudinal-study of agenda-setting for the issue of environ-
mental-pollution. Journalism & Mass Communication Quarterly, 72, 300–311.
Retrieved from http://jmq.sagepub.com/
Appel, M., & Richter, T. (2007). Persuasive effects of fictional narratives increase
over time. Media Psychology, 10, 113–134. doi:10.108/15213260701301194
Bandura, A. (1986). Social foundation of thought and action: A social cognitive
theory. Englewood Cliffs, NJ: Prentice-Hall.
Bandura, A. (2002). Social cognitive theory of mass communication. In J. Bryant &
D. Zillmann (Eds.), Media effects: Advances in theory and research (2nd ed.,
pp. 121–154). Mahwah, NJ: Erlbaum.
Bower, G. H., & Morrow, D. G. (1990). Mental models in narrative comprehension.
Science, 247(4938), 44–48. doi:10.1126/science.2403694
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 323
Britt, M. A., Perfetti, C. A., Sandak, R., & Rouet, J. (1999). Content integration and
source separation in learning from multiple texts. In S. Goldman, A. Graesser, &
P. van den Broek (Eds.), Narrative comprehension, causality, and coherence:
Essays in honor of Tom Trabasso (pp. 209–234). Mahwah, NJ: Erlbaum.
Busselle, R. W., & Bilandzic, H. (2008). Fictionality and perceived realism in experi-
encing stories: A model of narrative comprehension and engagement. Commu-
nication Theory, 18, 255–280. doi:10.1111/j.1468-2885.2008.00322.x
Cacioppo, J. T., & Petty, R. E. (1984). The elaboration likelihood model of persuasion.
Advances in Consumer Research, 11, 673–675. Retrieved from http://www.acr
website.org/volumes/
Centers for Disease Control and Prevention. (2011). Entertainment education. Re-
trieved from http://www.cdc.gov/healthcommunication/ToolsTemplates/Enter
tainmentEd/
Csikszentmihalyi, M. (1991). Flow: The psychology of optimal experience. New York,
NY: Harper Perennial.
Dahlstrom, M. F. (2010). The role of causality in information acceptance in narratives:
An example from science communication. Communication Research, 37, 857–
875. doi:10.1177/0093650210362683
de Wit, J. B. F., Das, E., & Vet, R. (2008). What works best: Objective statistics
or a personal testimonial? An assessment of the persuasive effects of different
types of message evidence on risk perception. Health Psychology, 27, 110–115.
doi:10.1037/0278-6133.27.1.110
Dunlap, R. E., Van Liere, K. D., Mertig, A. G., & Jones, R. E. (2000). Measuring
endorsement of the new ecological paradigm: A revised NEP scale. Journal of
Social Issues, 56, 425–442. doi:10.1111/0022-4537.00176
Gilbert, D. T. (1991). How mental systems believe. American Psychologist, 46, 107–
119. doi:10.1037/0003-066X.46.2.107
Graesser, A. C., Millis, K. K., & Zwaan, R. A. (1997). Discourse comprehension.
Annual Review of Psychology, 48, 163–189. doi:10.1146/annurev.psych.48.1.163
Graesser, A. C., Olde, B., & Klettke, B. (2002). How does the mind construct and
represent stories? In M. Green, J. Strange, & T. Brock (Eds.), Narrative impact:
Social and cognitive foundations (pp. 229–262). Mahwah, NJ: Erlbaum.
Graesser, A. C., Singer, M., & Trabasso, T. (1994). Constructing inferences during
narrative text comprehension. Psychological Review, 101, 371–395. doi:10.1037/
0033-295X.101.3.371
Graesser, A. C., & Wiemer-Hastings, K. (1999). Situation models and concepts in
story comprehension. In S. Goldman, A. Graesser, & P. van den Broek (Eds.),
Narrative comprehension, causality and coherence: Essays in honor of Tom
Trabasso (pp. 77–92). Mahwah, NJ: Erlbaum.
Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness
of public narratives. Journal of Personality and Social Psychology, 79, 701–721.
doi:10.1037/0022-3514.79.5.701
Jensen, J. D., Bernat, J. K., Wilson, K. M., & Goonewardene, J. (2011). The delay hy-
pothesis: The manifestation of media effects over time. Human Communication
Research, 37, 509–528. doi:10.1111/j.1468-2958.2011.01415.x
Judd, C. M., Kenny, D. A., & McClelland, G. H. (2001). Estimating and testing
mediation and moderation in within-subject designs. Psychological Methods,
6, 115–134. doi:10.1037/1082-989X.6.2.115
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
324 M. F. Dahlstrom
Kintsch, W. (1998). Comprehension: A paradigm for cognition. Cambridge, MA:
Cambridge University Press.
Laubscher, N. F. (1961). On stabilizing the binomial and negative binomial variances.
Journal of American Statistical Association, 56(293), 143–150. doi:10.1080/016
21459.1961.10482100
Mackie, J. L. (1980). The cement of the universe. Oxford, UK: Clarendon.
Magliano, J. P. (1999). Revealing inference processes during text comprehension.
In S. Goldman, A. Graesser, & P. van den Broek (Eds.), Narrative comprehen-
sion, causality, and coherence: Essays in honor of Tom Trabasso (pp. 55–76).
Mahwah, NJ: Erlbaum.
Marsh, E. J., Meade, M. L., & Roediger, H. L. (2003). Learning facts from fiction. Jour-
nal of Memory and Language, 49, 519–536. doi:10.1016/S0749-596X(03)00092-5
Mazor, K. M., Baril, J., Dugan, E., Spencer, F., Burgwinkle, P., & Gurwitz, J. H.
(2007). Patient education about anticoagulant medication: Is narrative evidence
or statistical evidence more effective? Patient Education and Counseling, 69,
145–157. doi:10.1016/j.pec.2007.08.010
McComas, K., & Shanahan, J. (1999). Telling stories about global climate change—
Measuring the impact of narratives on issue cycles. Communication Research,
26, 30–57. doi:10.1177/009365099026001003
Morgan, S. E., Movius, L., & Cody, M. J. (2009). The power of narratives: The effect of
entertainment television organ donation storylines on the attitudes, knowledge,
and behaviors of donors and non-donors. Journal of Communication, 59, 135–
151. doi:10.1111/j.1460-2466.2008.01408.x
Moyer-Gusé, E. (2008). Toward a theory of entertainment persuasion: Explaining
the persuasive effects of entertainment-education messages. Communication
Theory, 18, 407–425. doi:10.1111/j.1468-2885.2008.00328.x
Moyer-Gusé, E., & Nabi, R. L. (2010). Explaining the effects of narrative in an en-
tertainment television program: Overcoming resistance to persuasion. Human
Communication Research, 36, 26–52. doi:10.1111/j.1468-2958.2009.01367.x
O’Brien, E. J., & Myers, J. L. (1999). Text comprehension: A view from the bottom
up. In S. Goldman, A. Graesser, & P. van den Broek (Eds.), Narrative compre-
hension, causality, and coherence: Essays in honor of Tom Trabasso (pp. 35–
54). Mahwah, NJ: Erlbaum.
Oberauer, K., & Kliegl, R. (2006). A formal model of capacity limits in working mem-
ory. Journal of Memory and Language, 55, 601–626. doi:10.1016/j.jml.2006.
08.009
Russell, C. A. (2002). Investigating the effectiveness of product placements in tele-
vision shows: The role of modality and plot connection congruence on brand
memory and attitude. Journal of Consumer Research, 29, 306–318. doi:10.1086/
344432
Schank, R. C., & Abelson, R. (1995). Knowledge and memory: The real story. In R.
Schank & R. Abelson (Eds.), Knowledge and memory: The real story (pp. 1–86).
Hillsdale, NJ: Erlbaum.
Slater, M. D., & Rouner, D. (2002). Entertainment-education and elaboration likeli-
hood: Understanding the processing of narrative persuasion. Communication
Theory, 12, 173–191. doi:10.1093/ct/12.2.173
Slater, M. D., Rouner, D., & Long, M. (2006). Television dramas and support for con-
troversial public policies: Effects and mechanisms. Journal of Communication,
56, 235–252. doi:10.1111/j.1460-2466.2006.00017.x
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
Exploring Narrative Causality 325
Strange, J. J. (2002). How fictional tales wag real-world beliefs. In M. Green, J.
Strange, & T. Brock (Eds.), Narrative impact: Social and cognitive foundations
(pp. 263–286). Mahwah, NJ: Erlbaum.
Trabasso, T., & Sperry, L. L. (1985). Causal relatedness and importance of story
events. Journal of Memory and Language, 24, 595–611. doi:10.1016/0749-596X
(85)90048-8
Usdin, S., Singhal, A., Shongwe, T., Goldstein, S., & Shabalala, A. (2004). No short
cuts in entertainment-education: Designing Soul City step-by-step. In A. Singhal,
M. Cody, M. Everett, & M. Sabido (Eds.), Entertainment-education and social
change: History, research, and practice (pp. 153–176). Mahwah, NJ: Erlbaum.
van den Broek, P., & Gustafson, M. (1999). Comprehension and memory for texts:
Three generations of reading research. In S. Goldman, A. Graesser, & P. van
den Broek (Eds.), Narrative comprehension, causality, and coherence: Essays
in honor of Tom Trabasso (pp. 15–34). Mahwah, NJ: Erlbaum.
Voss, J. F., Wiley, J., & Sandak, R. (1999). On the use of narrative as argument. In
S. Goldman, A. Graesser, & P. van den Broek (Eds.), Narrative comprehension,
causality, and coherence: Essays in honor of Tom Trabasso (pp. 235–252).
Mahwah, NJ: Erlbaum.
Weinstein, N. D., Grubb, P. D., & Vautier, J. S. (1986). Increasing automobile seat
belt use: An intervention emphasizing risk susceptibility. Journal of Applied
Psychology, 71, 285–290. doi:10.1037/0021-9010.71.2.285
Yoder, P. S., Hornik, R., & Chirwa, B. C. (1996). Evaluating the program effects of a
radio drama about AIDS in Zambia. Studies in Family Planning, 27, 188–203.
doi:10.2307/2137953
Zwaan, R. A., Langston, M. C., & Graesser, A. C. (1995). The construction of situation
models in narrative comprehension: An event-indexing model. Psychological
Science, 6, 292–297. doi:10.1111/j.1467-9280.1995.tb00513.x
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
326 M. F. Dahlstrom
APPENDIX
Excerpt from the stimulus narrative detailing causal and noncausal assertion
placement
Note: The causal assertion in this excerpt is bolded and the noncausal
assertions are italicized. These identification marks were not present in the
experimental stimulus.
Lucas likes to watch as people respond to his work. This is why he now
stands next to his newly erected display, watching. But to his surprise, Lucas
sees many of the visitors shaking their heads in disapproval. He also notices
that the same visitors are carrying a green pamphlet. He decides to investigate
and approaches an old man looking at his display.
‘‘Excuse me sir, what is that pamphlet you have there?’’
‘‘It lists various facts about Earth Day. For instance: ‘‘Increasing levels of
smog have been linked to rising rates of asthma,’’ and ‘‘There are currently
more trees being planted worldwide than are being cut down.’’
‘‘Oh, that’s nice.’’
‘‘Yeah, I just wish whoever made this display would have read it,
because they obviously don’t know what they are talking about.’’
‘‘What?’’
‘‘Listen to this from the pamphlet: ‘‘Over ninety percent of runoff
contaminants are eliminated by microbes before entering water sys-
tems.’’ This display gets it all wrong! It just goes to show the lack of quality
in this exhibit. You know, back in my day :::’’
Lucas stops listening and looks at his display. He focuses on the part
of his work that the pamphlet now shows to be false. It glares at him,
overshadowing the rest of the display. It’s not fair that one mistake should
tarnish the entirety of his work :::
Downloaded by [Iowa State University], [Michael F. Dahlstrom] at 09:24 04 September 2012
