ArticlePDF Available

Transportation and Need for Affect in Narrative Persuasion: A Mediated Moderation Model

Authors:

Abstract and Figures

Two experiments investigated the idea that individual differences in need for affect are critical for narrative persuasion. Need for affect, that is, the disposition to approach emotions, was assumed to facilitate the experience of being transported into the mental world of the narrative. An intense experience of transportation, in turn, should enhance the persuasive impact of narrative information on readers' beliefs. A mediated moderation analysis was used to test these assumptions. In both experiments (N = 314), need for affect (approach) and transportation moderated the persuasive effects of a fictional narrative compared to a belief-irrelevant control story (Experiment 1) and the persuasive effects of a story with high emotional content compared to a story with low emotional content (Experiment 2). The moderator effects of need for affect were shown to be mediated by the moderator effects of transportation. In sum, the magnitude of a person's need for affect determines whether and to what extent the person experiences transportation into the story world and is persuaded by the information presented in the narrative.
Content may be subject to copyright.
Transportation and need for affect
1
Running Head: NARRATIVE PERSUASION AND NEED FOR AFFECT
Transportation and Need for Affect in Narrative Persuasion:
A Mediated Moderation Model
Markus Appel Tobias Richter
Johannes Kepler University Linz University of Cologne
Author Posting. (c) Taylor & Francis Group, LLC, 2010.
This is the author's version of the work. It is posted here by permission of Taylor & Francis Group, LLC, for
personal use, not for redistribution.
The definitive version was published in Media Psychology, Volume 13 Issue 2, April 2010.
doi:10.1080/15213261003799847 (http://dx.doi.org/10.1080/15213261003799847)
Transportation and need for affect
2
Transportation and need for affect
3
Abstract
Two experiments investigated the idea that individual differences in need for affect are critical
for narrative persuasion. Need for affect, i.e. the disposition to approach emotions, was assumed
to facilitate the experience of being transported into the mental world of the narrative. An intense
experience of transportation, in turn, should enhance the persuasive impact of narrative
information on reader's beliefs. A mediated moderation analysis was used to test these
assumptions. In both experiments (N = 314), need for affect (approach) and transportation
moderated the persuasive effects of a fictional narrative compared to a belief-irrelevant control
story (Experiment 1) and the persuasive effects of a story with high emotional content compared
to a story with low emotional content (Experiment 2). The moderator effects of need for affect
were shown to be mediated by the moderator effects of transportation. In sum, the magnitude of a
person's need for affect determines whether and to what extent the person experiences
transportation into the story world and is persuaded by the information presented in the narrative.
Keywords: narrative persuasion, need for affect, transportation, mediated moderation
Transportation and need for affect
4
Transportation and Need for Affect in Narrative Persuasion:
A Mediated Moderation Model
Every day, people around the world spend a substantial amount of time with narratives,
for example, reading novels and short stories, viewing soap operas, or following journalistic first-
hand accounts of individual tragedies. A number of studies have shown that reading or listening
to a narrative can alter beliefs that recipients hold about the world, even if the characters and
events described in the narrative are fictitious (e.g., Fazio & Marsh, 2008; Gerrig & Prentice,
1991; Green & Brock, 2000; Marsh & Fazio, 2006; Prentice, Gerrig & Bailis, 1997; Strange &
Leung, 1999). Despite commonalities between persuasion through narrative and other forms of
persuasive messages, some of the mechanisms underlying narrative persuasion are likely to differ
from those involved in persuasion through non-fictional, argumentative texts. Gerrig (1993) and
Green and Brock (2002) have proposed that the persuasive impact of fictional narratives is based
on an experiential state called “transportation,” which makes the recipients’ beliefs more
susceptible to influences by information provided in the narrative. The idea that transportation is
a general mechanism that underlies persuasion through narratives has been supported by a
number of studies (Escalas, 2004, 2007; Green, 2004; Green & Brock, 2000; Mazzocco, Green,
& Brock, 2007; Vaughn, Hesse, Petkova, & Trudeau, 2009). Whereas these studies were
primarily concerned with situational influences, the focus of the present article is on individual
differences in transportation and narrative persuasion. We will argue that the Need for Affect
(Maio & Esses, 2001) is a personality trait that can explain individual differences in narrative
persuasion. We assume that for individuals with a strong disposition to approach emotions, the
experience of transportation during processing is particularly intense and, as a consequence,
persuasive effects are particularly strong (Green & Brock, 2002). These assumptions amount to a
Transportation and need for affect
5
mediated moderation model of the interplay of need for affect, transportation, and the emotional
content of the narrative. Two experiments tested the predictions of this model.
Persuasion Through Fictional Narratives
Reading or listening to fictional and non-fictional narratives can alter beliefs that
recipients hold about the world. Research has demonstrated persuasive effects in response to the
central theme of a narrative (e.g., a dangerous psychiatric patient; Green & Brock, 2000) as well
as in response to information that was no central part of the plotline (e.g., Prentice et al., 1997;
Wheeler, Green, & Brock, 1999), and such effects seem to be durable (Appel, 2008a; Appel &
Richter, 2007). Thus, despite the fact that the authors of narratives often create fictitious worlds
in which invented events and characters occur, narratives carry information about the real world
that can have a profound influence on recipient beliefs (for a review, see Green, Garst, & Brock,
2004).
Whereas a number of the mechanisms of belief change that have been described in
general models of persuasion such as the Elaboration Likelihood Model (Petty & Cacioppo,
1986; Petty & Wegener, 1999) or the Heuristic-Systematic Model (Chen & Chaiken, 1999) also
apply to persuasion through fictional narratives, narrative persuasion cannot be fully captured by
these models. One reason is that narratives differ from other types of persuasive messages, most
notably from texts that present claims whose validity is backed up by arguments. People read or
listen to argument texts in order to get an informed world-view. The comprehension of such texts
frequently requires active elaboration, and comprehenders often fail to construct an adequate
representation of what the text is about. Narratives, in contrast, do not involve any claims about
the validity of the presented information, at least not if they are pieces of fiction. They usually
describe the actions and experiences of one or more protagonists and a plot line with certain
schematic elements (e.g., setting, event, attempt, reaction, and consequence, Rumelhart, 1975). In
Transportation and need for affect
6
many cases, people read or listen to narratives for pleasure (Nell, 1988). The comprehension of
narrative texts normally proceeds smoothly and effortlessly, and comprehenders have no
difficulty to construct and continuously update a situation model of the events that unfold when
the narrative proceeds (Zwaan & Radvansky, 1998). In the light of these differences, alternative
accounts of narrative persuasion have been developed that will be described in the following
section.
Narrative Persuasion and Transportation
According to the Transportation-Imagery Model by Green and Brock (2002), recipients of
narrative stories experience a state of transportation (Gerrig, 1993) while they are reading,
watching, or listening to a narrative. The term transportation is based on the metaphor that
readers undertake a mental journey into the world of a narrative. When an individual is
transported into the narrative world “all mental systems and capacities become focused on the
events occurring in the narrative” (Green & Brock, 2000). The resulting mental state has been
conceptualized as a co-activation of attention, imagery, and emotions (e.g., Green, 2004; Green &
Brock, 2000). Accordingly, transportation is a rather broad concept incorporating several aspects
of an absorbed reception of information. The construct of transportation and the Transportation
Scale proposed by Green and Brock (2000) for its measurement have been widely used in
research on narrative persuasion. Importantly, the amount of transportation has repeatedly been
found to predict the extent to which information contained in a narrative influenced recipients’
real-world beliefs (e.g., Green & Brock, 2000; see also Wyer, Adaval, & Colcombe, 2002).
Three different ways have been proposed how transportation facilitates narrative belief
change: First, transportation may reduce basic cognitive and elaborative activities that underlie
resistance to persuasion (such as epistemic monitoring, Richter, Schroeder, & Wöhrmann, 2009;
Schroeder, Richter, & Hoever, 2008; or counterarguing, Green & Brock, 2000). Second,
Transportation and need for affect
7
transportation goes along with a vivid mental simulation of the events described in a narrative,
with the consequence that these events may be misremembered as if they were actual real-world
experiences (Johnson, Hashtroudi, & Lindsay, 1993). Third, transportation involves strong
emotional experiences that can facilitate narrative persuasion via positive mood (transportation is
usually enjoyable, Green, Brock, & Kaufman, 2004), arousal (Clore & Schnall, 2005), or positive
identification with story characters (Oatley, 1994; Zillmann, 1991; cf., Mar & Oatley, 2008).
It is important to note that transportation is distinct from elaboration, the major cognitive
mechanism underlying persuasion according to general two-process models of persuasion such as
the Elaboration Likelihood Model (Petty & Wegener, 1999) and the Heuristic-Systematic Model
(Chen & Chaiken, 1999). Unlike elaboration, the mechanisms that have been proposed to explain
the impact of transportation are not based on distanced analytical processes such as propositional
reasoning or critical thinking. Quite to the contrary, transportation is conceived as a holistic
experiential state characterized by a close connection of the recipient to the story world and
emotional components. Summarizing these differences, Green and Brock (2000, p. 702) note that
elaboration can be construed as a divergent process because individuals engaged in elaboration
use information differently from that presented in a text (e.g., their prior knowledge, opinions,
and real-world experiences) to evaluate the arguments presented. By contrast, transportation can
be described as a convergent process as the individual is fully focused on the narrative itself.
Transportation and Other Forms of Immersed Reception
The experiential state of transportation is related to other constructs that imply an
attentive and absorbed reception of information. In order to clarify our rationale for basing our
own research on transportation, we will briefly discuss these other constructs. The first possible
conceptual alternative to transportation is the narrative engagement which has recently been
proposed by Busselle and Bilandzic (2008; 2009) as a multifaceted construct with four
Transportation and need for affect
8
dimensions (narrative understanding, attentional focus, emotional engagement, and narrative
presence). Incorporating narrative understanding, narrative engagement is defined even more
broadly than transportation. Narrative understanding refers to the extent to which recipients are
able to grasp the characters’ goals and actions and the events unfolding in a narrative. In contrast,
the other three components tapping experiential states (attentional focus, emotional engagement,
and narrative presence) directly correspond to central aspects of transportation, creating a large
overlap of these two constructs. In fact, Busselle and Bilandzic (2009) report very high
correlations of their narrative engagement scale (including all four components) with the
transportation scale (ranging from .73 to .82). In addition, the bivariate relationships of narrative
engagement and transportation with story-related attitudes (an indicator of narrative persuasion)
are moderate and almost identical for both scales. Thus, there seems to be a large overlap of the
two constructs both conceptually as well as empirically.
Other theories and concepts that have been used to describe an attentive and absorbed
way of reception have traditionally focused on more specific aspects of immersion such as the
illusion of non-mediation (presence, Lombard & Ditton, 1997) or the recipients’ experiential
state with regard to a story’s characters (e.g., suspense, Zillmann, 1991; identification, Mar &
Oatley, 2008; parasocial interaction, Klimmt & Hartmann, 2006). The alternative concept of
involvement is rather broad. However, it bears a puzzling heterogeneity with respect to the extant
definitions and operationalizations (Wirth, 2006). Finally, the concept of flow (Czikszentmihalyi,
1990) and transportation share a highly similar phenomenological experience (Green & Brock,
2002). However, general flow theory does not posit a connection between flow and persuasion.
Unfortunately, there is no standard method to measure the state of flow; in fact, many empirical
studies on the flow experience lacked a distinction between the predictors of flow and the
experiential state itself (Engeser & Rheinberg, 2008).
Transportation and need for affect
9
The present research is based on transportation rather than on any of the other constructs
because a) transportation rests on a coherent and well-established theoretical foundation that
focuses on the processing of narratives (Gerrig, 1993; Green & Brock, 2002), b) it captures a
large part of the phenomenological experience of being immersed in a narrative, c) it has been
used most frequently in previous studies on narrative persuasion which facilitates relating our
research to previous research, and d) it comes with a standardized self-report measure that has
been tested by a number of different research groups and showed good reliability and validity
(Green & Brock, 2000).
Individual Differences in Transportation
Transportation theory does not adopt the text hegemony hypothesis (Bloom, 1994),
according to which textual factors alone, i.e. independent of situational and individual factors,
can draw the reader into an absorbed and attentive way of reception. In contrast, Green and Brock
(2000, p. 703) assume that it is not only the text that determines the amount of transportation and
the persuasive impact of a story: Transportation and persuasion may vary with regard to the
extra-textual situational context and a general proclivity to get immersed into a story world (see
also Green, Garst, & Brock, 2004). In previous studies, the impact of situational factors has often
been weak and inconsistent, e.g., with regard to the impact of fact/fiction labels (Green & Brock,
2000; Green, Garst, Brock, & Chung, 2006; Prentice et al., 1997; Strange & Leung, 1999) or
specific reading goal instructions (Green, 2004; Green & Brock, 2000). This general finding
raises the question whether there are stable individual dispositions that regularly influence the
degree to which an individual experiences transportation and, in turn, is persuaded by a narrative
(Green, 2004, 2006). From a theoretical perspective, identifying individual dispositions that
moderate the impact of narratives on transportation and beliefs is an important objective because
adding these variables to the existing theoretical framework of transportation would strengthen its
Transportation and need for affect
10
explanatory power (cf., Underwood, 1975) and fill a theoretical gap. Technically, individual
dispositions may account for variance previously unexplained by text characteristics or
situational factors.
However, previous approaches to the issue of explaining individual differences in
transportation are somewhat unsatisfactory. Dal Cin, Zanna, and Fong (2004) have suggested a
self-report scale to measure individual differences in transportability. Similarly, Brock and
Livingstone (2004) have introduced a need for entertainment scale. A related construct is
openness to absorption that has been defined by Tellegen and Atkinson (1974) as the general
tendency to become absorbed by some activity. The problem with transportability, need for
entertainment, and absorption is that the conceptual definitions of these constructs and the
methods proposed for assessing them are very close to the situation-specific experiential state of
transportation. As a consequence, the construct of transportability merely describes rather than
explains the fact that there are stable individual differences in the degree of transportation. As
yet, it is an open question as to whether there are more broadly and independently defined
individual dispositions that determine the extent to which people experience transportation while
reading narrative texts and, as a consequence, their susceptibility to narrative persuasion.
In the context of general models of persuasion such as the Elaboration Likelihood Model
(ELM, Petty & Cacioppo, 1986; Petty & Wegener, 1999), need for cognition has been established
as the primary motivational disposition that determines the amount of cognitive processing of a
message. Individuals high in the need for cognition are more likely to engage in elaborative
processing of the information presented in a persuasive message, with the consequence that high-
quality arguments can lead to a belief change that is stable and resistant against further persuasive
attempts. However, we hold that in narrative persuasion, transportation rather than elaboration is
the key mechanism underlying persuasion. Given that transportation involves the experience of
Transportation and need for affect
11
emotions as a regular and essential component, we propose that individual differences in need for
affect rather than need for cognition are likely to play a major role in narrative persuasion.
Need for Affect as a Moderator of Narrative Persuasion
Need for affect has been proposed by Maio and Esses (2001) as the affective counterpart
to the need for cognition. They define the need for affect as the "general motivation of people to
approach or avoid situations and activities that are emotion inducing for themselves and others"
(p. 585). Emotion and affect are conceptualized in a broad sense that includes moods, emotions,
preferences, and related evaluations with an affective component. Although some affective states
(e.g., good feelings) are perceived as more positive than others (e.g., sad feelings), Maio and
Esses (2001) insist that "meaningful individual differences in the pursuit of affect on average” (p.
586) exist. Accordingly, need for affect has been described as a trait-like meta-emotion, i.e., a
generalized attitude regarding one’s own primary emotions (Bartsch, Vorderer, Mangold, &
Viehoff, 2008; Meyer & Gaschke, 1988). Maio and Esses (2001) have developed a Need for
Affect Questionnaire, which is a self-report measure that comprises of an approach and an
avoidance subscale.
There are strong theoretical reasons linking the need for affect to narrative transportation.
According to Maio and Esses (2001), individuals high in the need for affect actively seek out
emotional situations, and in such a situation they tend to intensify their emotional experiences.
Whereas effects of the avoidance aspect of need for affect seem to be confined to real,
unmediated situations, the approach aspect seems to be particularly relevant for experiencing
emotions during media reception (Bartsch, Appel, & Storch, 2010). In studies by Maio and Esses
(2001), for example, the approach subscale (but not the avoidance subscale) of the Need for
Affect Questionnaire predicted the selection of emotional vs. non-emotional movies, the
emotionality of the respondents' favorite television show, and the intensity of their emotional
Transportation and need for affect
12
reactions towards the death of Princess Diana (for similar results with the German version of the
Need for Affect scale, see Appel, 2008b). Intense emotional responses to the events described in
narratives and emotional identification with its characters are central components of
transportation (Gerrig, 1993; Green & Brock, 2000). Accordingly, individuals with a strong need
for affect should be inclined to experience high levels of transportation. The most important
precondition seems to be that the narrative must provide a certain amount of emotional content
that allows individuals high on need for affect to respond emotionally.
A Mediated Moderation Model of Narrative Persuasion
Our assumptions concerning the interplay of need for affect and transportation in
persuasion through fictional narratives imply a set of interrelated predictions that we tested in two
consecutive experiments. The goal of Experiment 1 was to establish a mediated moderation
model of narrative persuasion by examining whether the persuasive effects of a fictional narrative
depend on need for affect. Experiment 2 extended and refined this model by including the
emotional content of the narrative as predictor. To this end, we compared two versions of the
same narrative that differed in emotional content.
In Experiment 1, we predicted that reading a fictional narrative would cause a shift in
participants’ real-world beliefs as a result of information included in a fictional narrative
(Hypothesis 1). Analogous hypotheses have been tested and corroborated in several studies on
persuasion through fiction (e.g., Appel & Richter, 2007; Green & Brock, 2000, Experiment 4;
Prentice, Gerrig, & Bailis, 1997; Wheeler, Green, & Brock, 1999). The narrative used in
Experiment 1 (Murder at the Mall, Nuland, 1994) describes the murder of a young girl by a
psychiatric patient. Thus, we expected that individuals who read this narrative would perceive
psychiatric patients as more dangerous than those who read a control story on an unrelated topic.
The second prediction referred to the assumption that transportation is the key mechanism in
Transportation and need for affect
13
persuasion through fiction (Gerrig, 1993; Green & Brock, 2002). According to this assumption,
the degree of transportation experienced by an individual should moderate the strength of the
persuasive effect, with a higher degree of transportation yielding a stronger shift towards
believing that psychiatric patients are dangerous (Hypothesis 2).
Whereas the first two predictions have already received support in previous studies the
additional predictions concerning need for affect add novel aspects to the picture. Generally, we
expected that the persuasive impact of a fictional narrative would depend on the recipient’s level
of need for affect and that the influence of need for affect would be mediated by the degree of
transportation. These assumptions imply three empirical predictions. First, need for affect should
moderate persuasive effects of fictional narratives, with larger persuasive effects for individuals
high on need for affect (Hypothesis 3). Second, we expected that individuals high on need for
affect should also experience higher degrees of transportation (Hypothesis 4) because need for
affect should promote readers' immersion into the fictional world of the narrative and emotional
responses to narrative events. Thus, the role attributed to need for affect in narrative persuasion
parallels the role assumed for need for cognition in persuasion through argument texts (Chen &
Chaiken, 1999; Petty & Wegener, 1999) with the difference that transportation replaces
elaboration as the mediating mechanism. Third, if transportation is indeed the mediating
mechanism that underlies effects of need for affect, the hypothesized moderating effect of need
for affect should be mediated by the hypothesized moderating effect of transportation on
persuasion through fiction (Hypothesis 5).
These five predictions can be summarized in the mediated moderation model depicted in
Figure 1c. This model meets the defining characteristics of mediated moderation and resembles
the second type of mediated moderation model described by Muller, Judd, and Yzerbyt (2005).
However, rather than testing whether the interaction effect of need for affect and the story
Transportation and need for affect
14
manipulation (narrative with belief relevant information vs. control story on an unrelated topic) is
mediated by the main effect of another variable, the model proposed puts the focus on another
interaction effect as the mediator, i.e. the interaction of transportation with the story
manipulation.
1
Experiment 2 extended the perspective of Experiment 1 by applying an analogous
mediated moderation model to examine the role of emotional content more closely. Emotional
content such as portrayals of critical life events or descriptions of strong feelings in a narrative is
a precondition for emotional responses, which are thought to occur when readers are transported
into the fictional world of the narrative (Green & Brock, 2002). Accordingly, we expected that
transportation is a stronger predictor of persuasive effects when the story includes high emotional
content such as portrayals of critical life events or descriptions of strong feelings in a narrative as
compared to a story that includes low emotional content. The persuasive effect of a narrative with
strong emotional content is expected to increase with the degree of transportation that participants
experienced during reading. This effect should be alleviated or even absent when participants
read a narrative with weak emotional content (Hypothesis 6). A parallel interaction effect was
expected for need for affect because participants high on need for affect should be more likely to
respond to emotional content in a narrative (Hypothesis 7). As in Experiment 1, we predicted that
need for affect would have a positive effect on transportation (Hypothesis 8), and that the
moderator effect of need for affect would be mediated by the moderator effect of transportation
(Hypothesis 9, Figure 1c).
Experiment 1
Experiment 1 was designed to provide a first test of the mediated moderation model of
narrative persuasion outlined in the previous section (Figure 1). Persuasive effects were
determined by comparing participant responses to a fictional narrative that contained information
Transportation and need for affect
15
about violent acts of a psychiatric patient with those responding to a narrative that did not contain
this information. After reading the narrative, participants' beliefs about the dangerousness of
psychiatric patients were assessed. As potential moderator variables, the transportation that
recipients experienced during reading the narrative and their need for affect were included.
Method
Participants. Participants were 181 adult volunteers (96 women) between the ages of 18
and 39 (M = 24.4 years; SD = 4.1). They were personally recruited by student research assistants,
inside and outside the campus of the University of Linz (Austria). Seventy-five percent of the
respondents were students.
Text material. The experimental text material was based on the short story Murder at the
Mall (Nuland, 1994), which had been used in the persuasion studies of Green and Brock (2000,
Experiments 1-3). This story is about two women and their children who enjoy a day together at
the mall. A lengthy sequence describes how the daughter of one mother is brutally stabbed to
death by a man who is revealed at the end of the story to be a psychiatric patient with a history of
aggressive outbursts. In sum, the story implies that psychiatric patients are a potential danger and
that there is a necessity to be protected from them. The German translation of the story was
shortened and rewritten from a third-person-only perspective in order to make the style more
typical for a fictional narrative. In the control condition, we used a story called Long Live
Marrakech! (Meyer, 2005), which describes a dinner at a fancy restaurant, a topic unrelated to
homicide or mental illness. The experimental story and the control story were comparable in
writing style and length (experimental story: 76 lines, 855 words; control story: 76 lines, 801
words).
Dependent variable. As the dependent variable, we used a measure similar to the two-
item psychiatric patient index developed by Green and Brock (2000). Our psychiatric patient
Transportation and need for affect
16
index consists of four items measuring the belief in the dangerousness of psychiatric patients
(e.g., “Psychiatric patients who live in an institution should be allowed to go out in the
community during the day”) on seven-point scales (1 do not agree, 7 completely agree). The
internal consistency (Cronbach’s α) of this scale was .65 which is comparable to the internal
consistency (Cronbach’s α) of .69 reported by Green and Brock (2000). Higher scores indicated
more negative beliefs about psychiatric patients.
Moderator Variables. As potential moderators of persuasive effects of the experimental
story, we assessed individual differences in need for affect and the degree of transportation that
participants experienced while reading the stimulus texts.
Need for affect was assessed with the Need for Affect Questionnaire (2001; German
version: Appel, 2008b) which included approach- and avoidance-items. Only the approach
subscale of this measure was included in our analysis because in previous research, this subscale
was found to be related to emotional experiences during media reception (Appel, 2008b; Maio &
Esses, 2001). The scale is based on 13 items (with 7-point response scales ranging from -3 to +3)
that capture the individual disposition to approach emotions (e.g., “It is important to me to be in
touch with my feelings,” “I approach situations in which I expect to experience strong
emotions”). In the present sample of participants, the internal consistency (Cronbach's α) was
.84.
Transportation was assessed with a pre-tested German adaptation of the transportation
scale developed by Green and Brock (2000). This scale is intended to measure the state of
transportation experienced during reading a fictional narrative via retrospective self-reports. It is
based on 14 items (with 7-point response scales, ranging from 1 to 7) that refer to affective and
imaginative aspects of transportation (e.g., “The narrative affected me emotionally,” "While I
Transportation and need for affect
17
was reading the narrative, I could easily picture the events in it taking place"). The internal
consistency (Cronbach's α) of the transportation scale was .82.
Procedure and design. Experimenters provided the participants with a booklet that
contained either the experimental story or the control story, the dependent measure, a filler
questionnaire, the transportation scale, the need for affect scale, two questionnaires on media
exposure that are unrelated to the present research, and socio-demographic questions. After
completing the questions in the booklet, participants were thanked and debriefed. The design was
a one factorial between-subjects design with random assignment of participants to either the
experimental story with belief-relevant information or the control story.
Results and Discussion
The relationships specified in the mediated moderation model in Figure 1 were tested by
two alternative procedures. The first one was based on an adaptation of the hierarchical
regression procedure proposed by Muller et al. (2005), which is widely accepted as the standard
method for analyzing mediated moderation models. As a particular benefit, the sequence of
nested regression models proposed by Muller et al. (2005) allows a detailed evaluation of the
moderation part of mediated moderation models because the versatile tools of moderated
regression analysis for interpreting interactions can be used (Aiken & West, 1991; Cohen, Cohen,
West, & Aiken 2003). The second procedure used for analyzing the data was based on structural
equation modeling (SEM). Compared to the hierarchical regression procedure, using SEM has
the advantages that mediation hypotheses can be tested simultaneously in one single structural
model and on the level of latent variables (e.g., Bollen, 1989).
Transportation and need for affect
18
All significance tests reported in this article were based on a Type-I error probability of
.05. Descriptive statistics and intercorrelations of all variables in Experiment 1 are given in Table
1 (upper part).
Nested Regression Models
In all regression models reported in this section, the continuous predictors were z-
standardized before computing the interaction terms and entering them into the regression
equations in order to avoid non-essential multicollinearity (e.g., Cohen et al., 2003). The story
manipulation was coded with contrast coding (experimental story with belief-relevant
information: 1, control story: -1). The hypothesis concerning the mediation of the moderator
effect of need for affect through the moderator effect of transportation was addressed by
estimating a series of nested regression models (Baron & Kenny, 1986). According to the logic of
this approach, a mediation hypothesis is corroborated only if the following conditions are met:
(1) There is an effect of the distal predictor (here, the interaction of need for affect and the story
manipulation) on the outcome variable, (2) there is an effect of the potential mediator (here, the
interaction of transportation and the story manipulation) on the outcome variable, (3) there is an
effect of the distal predictor on the potential mediator, and (4) the effect of the distal predictor on
the outcome variable disappears or is considerably weaker after including the mediator in the
model. One problem of the stepwise approach is that it is susceptible to artificial results in cases
where power is insufficiently low (McKinnon, Lockwood, Hoffman, West, & Sheets, 2002). For
this reason, we also estimated the hypothesized indirect effect and tested it via the Sobel test
(Sobel, 1982). The regression coefficients estimated in the series of nested regression models and
the corresponding significance tests are provided in the left columns of Table 2.
Overall persuasive effect of the fictional narrative. Hypothesis 1 predicted that
participants who read the experimental story about a murder by a psychiatric patient would
Transportation and need for affect
19
express stronger beliefs in the dangerousness of psychiatric patients than participants who read
the control story about an unrelated topic. In line with this hypothesis, the psychiatric patient
index was higher after reading the experimental story (M = 5.03, SE
M
= 0.12) compared to
reading the control story (M = 4.31, SE
M
= 0.10; Table 2, left columns, Model 1). Thus,
Experiment 1 replicated the persuasive effect of fictional narratives that has been found in
previous studies (e.g., Green & Brock, 2000, Experiment 4).
Moderator effect of transportation. According to Hypothesis 2, the fictional narrative with
belief-relevant information should cause stronger persuasive effects in participants who reported
a higher degree of transportation into the fictional world of the narrative. In line with this
prediction, there was an interaction effect of the story manipulation with transportation (Table 2,
left columns, Model 2a). Following the recommendations by Cohen et al. (2003), we conducted
simple slope analyses to interpret this interaction by estimating the effect of transportation
separately in the group that read the experimental story and the group that read the control story
(Figure 3a, cf. Cohen et al., 2003). As expected, transportation had a positive effect on beliefs
about the dangerousness of psychiatric patients in participants who had read the experimental
story (B = 0.61, SE
B
= 0.13, p < .001, R
2
= .10). In contrast, there was no such effect in
participants who had read the control story with no belief-relevant information (B = -0.05, SE
B
=
0.11, p = .66, R
2
= .00). In addition to analyzing the simple slopes in the experimental group and
the control group, we estimated the persuasive effect of the experimental story compared to the
control story for participants who reported a high degree of transportation (one standard deviation
above the sample mean) and participants who reported a low degree of transportation (one
standard deviation below the sample mean). In these comparisons, a persuasive effect of the
experimental story occurred only in participants who reported a high degree of transportation (B
Transportation and need for affect
20
= 0.59, SE
B
= 0.12, p < .001, R
2
= .13) but not in participants who reported a low degree of
transportation (B = -0.10, SE
B
= 0.13, p = .45, R
2
= .00). This pattern of effects suggests that in
line with the Transportation-Imagery Model (Green & Brock, 2002), the persuasive effect of the
experimental story depended on the degree of transportation that participants experienced during
reading.
Moderator effect of need for affect. Hypothesis 3 predicted a moderator effect of need for
affect that should run parallel to the moderator effect found for transportation. Congruent with
this hypothesis, there was an interaction effect of this predictor with the story manipulation
(Table 2, left columns, Model 2b). In our simple slope analyses, we first analyzed the effect of
need for affect separately for the group that read the experimental story and the group that read
the control story. As predicted, need for affect had a positive effect on beliefs about the
dangerousness of psychiatric patients only in participants who had read the experimental story (B
= 0.23, SE
B
= 0.11, p < .05, R
2
= .02) whereas there was no significant effect in participants who
had read the control story (B = -0.19, SE
B
= 0.11, p = .08, R
2
= .01, Figure 3b). Again, we
estimated the magnitude of the persuasive effect at a low and a high level of need for affect (one
standard deviation above or below the mean). In these comparisons, a persuasive effect was
obtained at a high level of need for affect (B = 0.60, SE
B
= 0.12, p < .001, R
2
= .13) but not in
participants who reported a low degree of need for affect (B = 0.16, SE
B
= 0.12, p = .17, R
2
=
.01). Thus, the higher an individual's need for affect, the larger was the persuasive impact of the
experimental story on participants' beliefs concerning the dangerousness of psychiatric patients.
Mediated moderation. The core assumption of the mediated moderation model examined
in Experiment 1 (Figure 1c) was that the moderator effect of need for affect would be mediated
by the moderator effect of transportation. As one precondition of mediation, we tested whether
Transportation and need for affect
21
need for affect promoted the degree of transportation that participants experienced during reading
the experimental story or the control story (Hypothesis 4). In line with this hypothesis, need for
affect had a positive effect on transportation (B = 0.23, SE
B
= 0.07, p < .01, R
2
= .05). The
mediation hypothesis proper (Hypothesis 5) was tested by a regression model that included both
the moderator effects of transportation and need for affect (Table 2, left columns, Model 3).
According to Baron and Kenny (1986), the mediated effect should be considerably reduced and
no longer be significant when the mediator is entered into the model. The parameter estimates of
the mediated moderation model conformed to this requirement: Compared to the original model,
the moderator effect of need for affect was much lower and no longer significant (B = 0.14, SE
B
=
0.09, p = .10, R
2
= .01) after entering the moderator effect of transportation into the model. In
contrast, the moderator effect of transportation remained significant (B = 0.31, SE
B
= 0.09, p <
.001, R
2
= .05). In addition to the stepwise procedure, we estimated and tested the indirect effect
of the moderator effect of need for affect via the moderator effect of transportation on the
dependent variable. The indirect effect was estimated as 0.07 and was significant in a Sobel test
(z = 2.47, p < .05). Thus, both common procedures for detecting mediation consistently indicate
that the moderator effect of need for affect was mediated by the moderator effect of
transportation, as hypothesized in the mediated moderation model depicted in Figure 1c.
Structural Equation Model
SEM provides an alternative means to test the assumptions of the hypothesized mediated
moderation model. In the framework of SEM, the mediation part of our theoretical model
corresponds to a structural model that contains one path from need for affect to transportation and
a second path from transportation to beliefs about the dangerousness of psychiatric patients. The
moderation part can be implemented by using multi-sample analysis (Bollen, 1993), with two
Transportation and need for affect
22
samples based on the experimental group and the control group respectively. For the moderation
part, our theoretical model implies that the first path from need for affect to transportation should
be invariant across the two groups (the equivalent of no interaction) whereas the second path
from transportation to beliefs should vary (the equivalent of an interaction of transportation and
the story manipulation). In particular, a positive effect of transportation (and, hence, a positive
indirect effect of need for affect) on beliefs was expected for the group that received the
experimental story with belief-relevant information whereas no effect (and, hence, no indirect
effect of need for affect) on beliefs was expected for the group that received the control text.
We tested these predictions with a multi-sample model in which need for affect,
transportation, and beliefs about the dangerousness of psychiatric patients were conceptualized as
latent variables. Each of these latent variables was measured by two item parcels, each of which
contained half of the items of the original scale (items were randomly assigned to the item
parcels). The factor loadings of the item parcels associated with one latent variable were
constrained to be equal. The structural model contained one path from need for affect to
transportation and a second path from transportation to beliefs about the dangerousness of
psychiatric patients but no direct effect from need for affect to beliefs. Thus, full mediation was
assumed for the mediation part of the model. In order to test the moderation part, we performed a
multi-sample analysis with the group that received the experimental story and the group that
received the control story constituting one sample each. Two different multi-sample models, one
nested within the other one, were estimated and tested against each other. In the hypothesized
model, the parameters in the measurement models (factor loadings and error variances) and the
coefficient of the path from need for affect to transportation were constrained to be invariant but
the path from transportation to beliefs was allowed to vary between the two groups. In the
alternative model, the path from transportation to beliefs was constrained to be invariant across
Transportation and need for affect
23
the two groups as well, excluding a moderating role of the text that participants received.
Parameters were estimated with the Maximum Likelihood procedure implemented in Lisrel 8
(Jöreskog & Sörbom, 1996).
The parameter estimates of the hypothesized model are provided in Figure 3a. As
expected and in line with the results of the hierarchical regression models, need for affect had a
positive effect on transportation. Transportation, in turn, had a positive effect on beliefs about the
dangerousness of psychiatric patients only in participants who had received the experimental
story with belief relevant information. It did not have a significant effect on beliefs in participants
who received the control story with no belief-relevant information. Accordingly, the indirect
effect of need for affect on beliefs mediated by transportation was positive and significant only in
the experimental group (Sobel-test: z = 3.02, p < .01) whereas it did not reach significance in the
control group (z = 0.47, p = .64). Overall, the hypothesized model had an acceptable model fit, χ
2
(28, N = 181) = 33.75, p = 0.21, RMSEA = 0.05, Standardized RMR = 0.08, NNFI = 0.98, CFI =
0.98. Compared to the hypothesized model, the fit of the alternative model was considerably
worse, χ
2
(29, N = 181) = 49.3, p < 0.05, RMSEA = 0.09, Standardized RMR = 0.13, NNFI =
0.94, CFI = 0.95. A chi-square difference test revealed that the difference in model fit was
significant, ∆χ
2
(1, N = 181) = 15.5, p < .001. Thus, both the mediation part and the moderation
part of the theoretical model were essential for the model to adequately represent the data.
In sum, the SEM analysis fully corroborates the conclusions that can be drawn from the
results of the hierarchical sequence of regression models. The results of both approaches
substantiate our assumptions concerning the roles of transportation and need for affect in
narrative persuasion. However, the mediated moderation model summarized in Figure 1 entails
the more precise assumption that it is the emotional content of a narrative that drives the
Transportation and need for affect
24
hypothesized effects. We conducted a second experiment that addressed this question and, at the
same time, allowed a constructive replication of the effects found in Experiment 1.
Experiment 2
The primary goal of Experiment 2 was to examine the interplay of emotional content
and need for affect in narrative persuasion in more detail. The topic of the stories used as text
materials was organ donation. In many societies, organ donation is a highly important topic
because despite public campaigns to raise awareness of the problem, the number of organ donors
falls well short of the need of organs (e.g., National Kidney Foundation, 2008). Organ donation is
also a topic that is apt to cause emotional responses when it is communicated via narratives
(Kopfman, Smith, Ah Yun, & Hodges, 1998). In contrast to the previous experiment, Experiment
2 employed two stories that differed in their emotional appeal but provided the same factual
information and message. In one version of the story the protagonist was killed in an accident
(high emotional content) whereas in a second version the protagonist only thought about the
possibility of being killed in an accident (low emotional content).
Narrative descriptions of significant life events have the potential to elicit emotional
responses. According to the theoretical view advocated here, these emotional responses occur to
the extent that recipients are transported into the world of the narrative (Green & Brock, 2002).
Thus, we expected the persuasive effects of the narrative with high emotional content to increase
with the degree of transportation that participants experienced reading the narrative. For the
narrative with low emotional content, in contrast, no such effect of transportation was expected.
A parallel interaction effect was predicted for need for affect as participants high on need for
affect should be more likely to experience emotions in response to descriptions of significant life
events. As in Experiment 1, we expected the moderation effect of need for affect to be mediated
Transportation and need for affect
25
by the moderation effect of transportation as outlined in the mediated moderation model in Figure
1c.
Method
Participants. A self-selected sample from a larger pool of volunteers participated in the
study which was conducted online. The data revealed that five participants did not respond
thoroughly; the time to complete the study indicated that they were merely clicking through the
web-based material. Both experimental conditions had an equal number of drop-outs with 7 and 6
respondents for each of two experimental narratives respectively (see Procedure and Design).
Clickers and drop-out responders were not included in our analyses. The remaining sample
consisted of 133 adults (93 women) with an average age of 30.5 years (SD = 12.2). About one
third of the participants reported 10 years or less of school education, a second third had 12 years
of school education, and the remaining participants had obtained a bachelor’s or master’s degree.
Text material. The experimental text material was inspired by a story used in a previous
study on the processing of narrative information (Kopfman et al., 1998). Two stories were
constructed that highlighted the medical treatment opportunities and the necessity of becoming an
organ donor but differed in the amount of emotional content. Both stories described two young
girls who were active members in organizations that promote organ donation. Factual
information about organ donation (e.g., improved treatment with the help of donated organs, lack
of donors) was worked into the narrative description of the two girls. In the story version with
low emotion content, the beginning sequence introduced a student called René, who, on his way
to class, was reminded of the dangers of car traffic. He subsequently decided to become an organ
donor. In the story version with high emotion content, René, on his way to class, was hit by a car
and died. Before the accident, he had decided to become an organ donor. Both stories were of the
same length (38 lines, 421 words).
Transportation and need for affect
26
Dependent variable. We used a 13-item self-report measure based on a previous
instrument by Parisi and Katz (1986) to assess participant's beliefs about the benefits of organ
donation (e.g., “By agreeing to donate organs at death, one sets a good example for others to
follow”). The items were rated on a seven-point scale (1 = do not agree, 7 = completely agree).
Exploratory factor analyses yielded a one-factor solution. The internal consistency (Cronbach’s
α) of this scale was .86.
Moderator Variables. As in Experiment 1, individual differences in need for affect were
assessed with the Need for Affect Questionnaire (Maio & Esses, 2001; for the German
adaptation, Appel, 2008b). The approach subscale was included in our analyses; in the present
sample, the internal consistency (Cronbach’s α) of this scale was .87. Transportation was
assessed with the German adaptation of the transportation scale (Green & Brock, 2000). The
scale included eleven text-invariant items and three items that were adjusted to the names of the
protagonists of the organ donation story. Cronbach’s α of this scale was .86.
Procedure and design. Participants were invited to take part in the study through e-mail.
All materials were presented on-line and were accessed by the participants via the web browser
of their home computers. The software used for presenting the experiment and collecting data,
EFS-survey, monitored potential repeat responders through IP protocols (cf., Gosling, Vazire,
Srivastava, & John, 2004) and collected the time spent with different parts of the study.
The material presented contained the low emotion story or the high emotion story, the
dependent measure, the transportation scale, the need for affect scale, and socio-demographic
questions. In addition, it was randomly determined whether these scales were presented after the
transportation items or before the experimental story. After completing the study, participants
Transportation and need for affect
27
were thanked and debriefed. The design was a one-factorial between-subjects design with random
assignment of participants to either the story with low or with high emotional content.
Validation of the emotion content manipulation. We validated the experimental
manipulation of emotional content in a separate online study with 106 participants (this sample
was completely independent from the sample that participated in Experiment 2). These
participants read either the high emotion story or the low emotion story. Subsequently, they rated
the emotionality of the text from the viewpoint of a neutral observer on a six-item scale (e.g.,
“Plot and narrative style of this text are rather neutral, making it unlikely that the text is capable
of evoking emotions in the reader,” reverse-scored item; Cronbach's α = .90). In addition, six
items measured the intensity of participant's own emotions while reading the text (e.g., “The text
touched me emotionally;” Cronbach's α = .83). The items of both scales were rated on seven-
point scales (1 = do not agree, 7 = completely agree). Finally, we measured need for affect with
the approach subscale of the need for affect instrument (Appel, 2008b; Maio & Esses, 2001). As
intended, the high emotion story was rated as more emotional from the viewpoint of a neutral
observer (M = 4.23, SE
M
= 0.16) compared to the low emotion story (M = 3.51, SE
M
= 0.19),
t(105) = 2.9, p < .01, d = 0.58. There was an overall tendency of participants to report more
intense emotions after reading the high emotion story (M = 4.10, SE
M
= 0.17) compared to the
low emotion story (M = 3.88, SE
M
= 0.19), but this difference was not significant, t(105) = 0.9, p
= .40. However, a moderated regression analysis with need for affect as moderator revealed a
significant interaction of emotional content with need for affect, F(1,103) = 4.8, p < .05, η
2
= .05.
In subsequent simple slopes analyses, need for affect had a positive effect on the intensity of
emotional experience only after reading the high emotional text (B = 0.63, SE
B
= 0.15, p < .001,
R
2
= .14) whereas the simple slope of need for affect was not significant after reading the low
Transportation and need for affect
28
emotional text (B = 0.05, SE
B
= 0.22, p = .82). In participants high in need for affect (one
standard deviation above the mean), the high emotion story caused more intense emotions
compared to the low emotion story (B = 0.43, SE
B
= 0.17, p < .05, R
2
= .05) whereas no
difference was found for participants low in need for affect (one standard deviation below the
mean, B = -0.16, SE
B
= 0.20, p = .43). In sum, the high emotion story was judged as more
emotional than the low emotion story by the participants of the validation study. In addition, it
also caused more intense emotional responses in those participants high on need for affect. These
results corroborate the validity of the emotional content manipulation.
Results and Discussion
In order to test the relationships between emotional content, need for affect, and
transportation specified in the mediated moderation model in Figure 1, we used the same two
procedures as in Experiment 1, a sequence of hierarchical regression analyses and an analysis
based on SEM.
Nested Regression Models
All continuous predictors were z-standardized and the story manipulation was coded with
contrast coding (high emotion content: 1, low emotion content: -1). In contrast to the story
manipulation in Experiment 1, we did not expect a main effect of emotional content. However,
we predicted an interaction effect of emotional content and transportation on story-consistent
beliefs and a parallel interaction effect for need for affect. The interaction of need for affect and
emotional content was expected to be mediated by the interaction of transportation and emotional
content. Descriptive statistics and intercorrelations of all variables in Experiment 2 are given in
Table 1 (lower part). Regression coefficients and the corresponding significance tests are
provided in Table 2 (right columns).
Transportation and need for affect
29
Overall effect of emotional content. Overall, the story versions with low and high
emotional content yielded similar attitudes towards organ donation (low emotion: M = 3.79, SE
M
= 0.07; high emotion: M = 3.72, SE
M
= 0.09; Table 2, right columns, Model 1) that did not differ
significantly from each other.
Moderator effect of transportation. Hypothesis 6 predicted that the emotional content of a
narrative should cause stronger persuasive effects in participants who reported a high degree of
transportation. As expected, an interaction effect of the story manipulation with transportation
emerged; Table 2, right columns, Model 2a). We estimated the simple slopes of transportation in
each of the two groups to interpret the interaction (Figure 4a, cf. Cohen et al., 2003).
Transportation had a positive effect on beliefs about the benefits of organ donation in participants
who had read the story with the critical life event (high emotion, B = 0.32, SE
B
= 0.07, p < .001,
R
2
= .14). No such effect was found in participants who had read the story with low emotional
content (B = 0.01, SE
B
= 0.08, p = .86, R
2
= .00). Furthermore, we analyzed the persuasive
effect of the experimental story compared to the control story for participants who reported a high
degree of transportation and participants who reported a low degree of transportation (i.e., one
standard deviation below and above the sample mean). For participants who reported a low
degree of transportation, the story version with high emotional content was less persuasive than
the story version with low emotional content (B = -0.19, SE
B
= 0.07, p < .01, R
2
= .05); a
contrary tendency was found for those who reported a high degree of transportation (B = 0.12,
SE
B
= 0.08, p = .14, R
2
= .01). These results are in line with our prediction that the persuasive
effect of emotional content increases with participants’ transportation into the story world.
Moderator effect of need for affect. Hypothesis 7 predicted a moderator effect for need for
affect that should run parallel to the moderator effect we found for transportation. As expected,
Transportation and need for affect
30
we found an interaction effect of this predictor with the story manipulation (Table 2, right
columns, Model 2b). Subsequent simple slope analyses showed a positive effect of need for
affect on beliefs about the benefits of organ donation in participants who had read the story
version with high emotional content (B = 0.29, SE
B
= 0.08, p < .001, R
2
= .09; Figure 4b), but
no significant effect in participants who had read the story version with low emotional content (B
= 0.02, SE
B
= 0.07, p = .80, R
2
= .00). In addition, we looked at the persuasive effect at a low
and a high level of need for affect (one standard deviation above or below the mean). For
participants low in need for affect, high emotional content had a negative effect on beliefs about
the benefits of organ donation, B = -0.19, SE
B
= 0.08, p < .05, R
2
= .04. For participants high on
need for affect, there was no significant difference between the story versions with and without
emotional content (B = 0.08, SE
B
= 0.07, p = .27, R
2
= .01). Thus, the pattern of the interaction
parallels the one that was found for transportation and emotional content.
Mediated moderation. According to our mediated moderation model, the moderator effect
of need for affect was expected to be mediated by the moderator effect of transportation. In a first
step, we tested whether need for affect had a positive effect on the level of transportation that
participants experienced during reading (Hypothesis 8). The predicted effect was indeed found (B
= 0.24, SE
B
= 0.08, p < .01, R
2
= .06). The mediation effect proper (Hypothesis 9) was tested by
a regression model that included both moderator effects (Table 2, right columns, Model 3). As
expected, after entering the moderator effect of transportation into the model, the moderator
effect of need for affect decreased and was no longer significant (B = 0.12, SE
B
= 0.06, p = .06,
R
2
= .02). However, the moderator effect of transportation remained stable (B = 0.15, SE
B
=
0.06, p < .05, R
2
= .04). Thus, all the requirements for mediation outlined by Baron and Kenny
(1986) were met. Furthermore, we estimated and tested the indirect effect of the moderator effect
Transportation and need for affect
31
of need for affect via the moderator effect of transportation on the dependent variable. The
indirect effect was estimated as 0.04 and turned out to be significant in a Sobel test (z = 2.02, p <
.05). Taken together, both common procedures for detecting mediation effects yielded support for
the mediated moderation model depicted in Figure 1c.
2
Structural Equation Model
The SEM analysis of the data from Experiment 2 mirrored that of the data from the
previous experiment. Accordingly, the mediation part of our theoretical model was represented
by a structural model with one path from need for affect to transportation and a second path from
transportation to beliefs about the benefits of organ donation. The moderation part was
implemented by a multi-sample analysis (Bollen, 1993) with the two experimental groups
constituting the samples. Again, our theoretical model implies that the first path from need for
affect to transportation should be invariant across the two groups whereas the second path from
transportation to beliefs should be positive in the group that received the story version with high
emotional content whereas no effect was assumed for the group that received the story version
with low emotional content. Need for affect, transportation, and beliefs about organ donation
were included as latent variables, each of which was measured by two item parcels with factor
loadings constrained to be equal. As for the data of Experiment 1, we estimated a hypothesized
model that incorporated our theoretical assumptions and compared its fit to a more parsimonious
alternative model in which the path from transportation to beliefs was constrained to be invariant
across the two experimental groups.
The parameter estimates of the hypothesized model are provided in Figure 3b. In line with
our theoretical expectations, need for affect had a positive effect on transportation.
Transportation, in turn, had a positive effect on beliefs about the benefits of organ donation only
in participants who had received the story version with high emotional content. It did not have a
Transportation and need for affect
32
significant effect on beliefs in participants who received the story version with low emotional
content. Consequently, the indirect effect of need for affect on beliefs mediated by transportation
was positive and significant only in the group that received the story version with high emotional
content (Sobel-test: z = 2.52, p < .05). However, it was not significant in the group that received
the story version with low emotional content (z = 0.15, p = .88). The model fit of the
hypothesized model was acceptable, χ
2
(28, N = 133) = 35.97, p = 0.14, RMSEA = 0.07,
Standardized RMR = 0.05, NNFI = 0.97, CFI = 0.97. The fit of the alternative model was worse
than that of the hypothesized model, χ
2
(29, N = 133) = 43.66, p < 0.05, RMSEA = 0.09,
Standardized RMR = 0.11, NNFI = 0.95, CFI = 0.95. The difference in model fit was significant,
∆χ
2
(1, N = 133) = 7.69, p < .01. Thus, as in Experiment 1, the SEM analysis and the analysis
based on the hierarchical sequence of regression models consistently corroborated both the
mediation and the moderation part of our theoretical model.
General Discussion
The aim of the present research was to test a mediated moderation model of narrative
persuasion that includes the recipients’ need for affect (Maio & Esses, 2001) as distal predictor
and their experience of being transported into the world of the narrative (Gerrig, 1993; Green &
Brock, 2000) as proximal predictor. In Experiment 1, need for affect as well as the self-reported
degree of transportation during reading moderated the persuasive impact that a story with belief-
relevant information had compared to a story without that information. Furthermore, need for
affect had a positive effect on the amount of transportation that participants experience while
reading the narrative. In a subsequent mediation analysis, we found the moderator effect of need
for affect to be mediated by the moderator effect of transportation (mediated moderation, cf.,
Muller et al., 2005). Experiment 2 refined and expanded these results by contrasting a narrative
Transportation and need for affect
33
with high emotional content with a narrative with low emotional content but identical belief-
relevant information. Again, we found parallel moderator effects of need for affect and
transportation. Both need for affect and transportation were associated with persuasive effects
when participants read the narrative with high emotional content. However, these effects were not
present when participants read the narrative with low emotional content. Conversely, the
narrative with high emotional content yielded stronger persuasive effects than the narrative with
low emotional content only for those participants high on need for affect and transportation. As in
Experiment 1, the moderator effect of need for affect was mediated by the moderator effect of
transportation, supporting the hypothesized mediated moderation model of narrative persuasion.
Theoretically, these results establish need for affect (Maio & Esses, 2001) as a personal
disposition that influences narrative persuasion in a consistent and significant way: The more
people are inclined to approach emotional situations, the more their beliefs are shifted towards
information that is woven into the plot of the narrative. The results also reveal that the
experiential state of transportation as described by Gerrig (1993) and Green and Brock (2000) is
the mediator by which need for affect exerts its role as a moderator of narrative persuasion.
Apparently, recipients who are disposed to approach emotional situations are also those who
more readily indulge in the experience of being transported into the fictional world of the
narrative. One important component of transportation consists of emotional responses to the
events described in the narrative. On the one hand, it is likely that need for affect facilitates these
emotional responses. On the other hand, there are additional ways by which emotional responses
can influence beliefs, for example, by serving as evaluative cues, by causing mood-induced
changes in the way people process the information presented in a narrative, or by creating
physical arousal that makes the described information more memorable (cf., Clore & Schnall,
Transportation and need for affect
34
2005). Future research may directly test these paths which depend on the interplay between
emotional gratifications provided by a narrative and individual differences in the need for affect.
We do not assume that the impact of need for affect on experiential states is limited to
transportation and its operationalization. Transportation has a substantial overlap with narrative
engagement and flow, so we expect similar relationships of these concepts with the need for
affect. Moreover, the affective component of transportation is closely linked to affective
responses such as identification, suspense, or enjoyment. Thus, future research may successfully
connect the need for affect with these more specific experiential states and associated theories.
A related question raised by the present experiments concerns the relationship of emotional,
attentional, and cognitive components of transportation. Our results on the relationship of need
for affect and transportation seem to highlight emotional components as a driving force of
narrative persuasion. However, this does not imply that the attentional and cognitive components
described by Gerrig (1993) and Green and Brock (2000) are irrelevant for narrative persuasion, or
that transportation should be conceptualized as a purely affective concept. The major theoretical
accounts (Gerrig, 1993; Green & Brock, 2002) describe transportation as a holistic experiential
state in which emotional, attentional, and cognitive components are closely interwoven with each
other (recent empirical work by Busselle & Bilandzic, 2009, supports this view). Provided that
this description is adequate, need for affect might not only facilitate emotional responses but also
more cognitive mechanisms of narrative persuasion such as reduced cognitive activity and vivid
mental simulation. At any rate, more research is needed on the dimensionality of transportation
and on the exact nature of the emotional and cognitive processes associated with it.
Our results complement recent findings on the impact of the need for affect on the
persuasion through affect-based messages (Haddock, Maio, Arnold, & Huskinson, 2008). These
authors found that the need for affect moderated the impact of an affect-based message about a
Transportation and need for affect
35
fictitious animal on participants’ attitudes (e.g., good-bad) and the recognition of message
information. Given that the affect-based message was a narrative, these effects may have been
mediated through the link from need for affect to transportation highlighted by our experiments.
Are Emotional Stories Always Persuasive?
Considering that emotion seems to be a key to narrative persuasion, one might suspect that
emotional stories per se are more persuasive than unemotional stories. Our model and results
indicate that such a broad conclusion would be premature. Descriptions of significant life events
or other emotionally appealing content do not always contribute to the persuasive impact of a
story. Rather, our mediated moderation model highlights the interplay of emotional content and
need for affect in narrative persuasion. According to our model, emotional content is supposed to
foster persuasion only if recipients are transported into the story world, which, in turn, is more
likely when the need for affect is high. The comparisons between the story versions with high
and low emotional content in Experiment 2 at different levels of need for affect indicated that
emotional content may lead to less story-consistent beliefs in those recipients who are not at all
inclined to experiencing emotions. In that sense, the role of emotional content in narrative
persuasion may differ from that of high-quality arguments in persuasion through non-narrative
texts. Unlike compelling arguments that never impede belief change, emotional content to a story
may have adverse effects for people who are not inclined to have intense emotional experiences.
At this point, the mechanisms underlying these effects are unclear. It might be, for example, that
readers experiencing less transportation not only show less intense emotional responses to the
narrative but also have more cognitive resources available to engage in active counterarguing.
Thus, future research should include direct measures of cognitive and emotional responses to sort
out these possibilities.
Transportation and need for affect
36
Our results on the moderating role of need for affect on narrative persuasion have important
implications for narrative persuasion in applied settings. Narratives are increasingly used in
campaigns to promote public health and social cooperation (Paluck, 2009; Singhal, Cody,
Rogers, & Sabido, 2004) and are part of product advertisements such as drama ads, or
transformational ads (Deighton, Romer, & McQueen, 1989). At the same time, it is becoming
more and more popular to tailor persuasive communication to individual characteristics of the
consumer, something that can be easily done with computer-based messages (e.g., Dijkstra,
2008). Our research emphasizes the importance to consider that an individual's need for affect is
critical for the success of narrative emotional messages. Given that background characteristics of
individuals (such as previous media choice) can be used to predict their need for affect (Maio &
Esses, 2001), our research points out a feasible strategy to increase communication effects and
marketing revenues.
Pathos in Narrative Persuasion
Our model of the relationships between need for affect, transportation, and narrative
persuasion rests on the argument that emotional responses crucially determine the effect of stories
on beliefs. The roots of this argument date back to Aristotle who distinguished three modes of
rhetorical proof in his Rhetoric (Aristotle, 367-322 B.C.E. / 2001; see also McGuire, 1969). In
addition to logical proof (logos) which is based on the arguments provided for or against a
position, and ethical proof (ethos), which lies in the speaker’s character and credibility, Aristotle
identified a third way to persuade the audience: Emotional proof or pathos. Successful persuasion
by means of pathos includes the experience of emotions (Aristotle, 367-322 B.C.E. / 2001, Book
1, Ch. 2, §1356a). Recipients’ emotions such as anger or pity are thought to influence judgments
about issues, people, or social groups. Narratives are a prototypical way to persuade through
pathos (e.g., Braet, 1992), an assertion that is in line with the results presented here. Of course, a
Transportation and need for affect
37
more stringent test would require direct and concurrent measures of emotional responses such as
think-aloud data or facial expressions of emotions.
Generally speaking, the role of affect that is caused by external factors (e.g., good mood
due to sunny weather) is well-documented in the persuasion literature (e.g., Bless, Bohner,
Schwarz, & Strack, 1990). The literature on affective messages and related processing, however,
is still sparse, with the research on fear-appeals being one exception (see Clore & Schnall, 2005,
and Petty, Fabrigar, & Wegener, 2003, for reviews). Our results suggest that persuasive attempts
through pathos may have a more pervasive impact than previously conceded. Edwards (1990), for
example, has argued that inducing affect is an effective means to change affect-based attitudes
but not cognition-based attitudes which require the presentation of arguments. However, from the
perspective of recent dual-process theories of social information processing (e.g., Gawronski &
Bodenhausen, 2007; Strack & Deutsch, 2004), it seems plausible to assume that emotional
narratives can change cognition-based attitudes and beliefs as well. For example, according to
Gawronski and Bodenhausen, affective reactions may be used as the basis for propositional
evaluative judgments if the affective reaction is consistent with currently active propositional
information.
The present research highlights two potential moderators of these effects, need for affect
as the major personal disposition and emotional content as the major narrative characteristic. In
further research, it would be worthwhile to examine whether and how the persuasive effects of
narratives depend on the quality of emotional responses during narrative comprehension. Both of
our experimental texts involved a tragic, lethal incident (murder, accident). More research is
needed on different kinds of stories and different kinds of emotional responses (i.e., anger, fear,
surprise, but also enjoyment) because these might have distinct effects on propositional beliefs
Transportation and need for affect
38
(Dillard & Nabi, 2006; Nabi, 1999). Alternatively, narrative persuasion might be based on
broader affective states such as moods or on the arousal that accompanies emotional responses.
Transportation and need for affect
39
References
Appel M. (2008a). Fictional narratives cultivate just-world beliefs. Journal of Communication,
58, 62-83. doi:10.1111/j.1460-2466.2007.00374.x
Appel, M. (2008b). Manche mögen’s heiß: Ergebnisse der deutschsprachigen Version eines
Instruments zur Erfassung des Emotionsmotivs (Need for Emotion /Need for Affect)
[Some like it hot: Results of the German language version of an instrument for assessing
the need for affect/need for emotion]. Diagnostica, 54, 2-15.
doi: 10.1026/0012-1924.54.1.2
Appel, M., & Richter, T. (2007). Persuasive effects of fictional narratives increase over time.
Media Psychology, 10, 113-134. doi: 10.1080/15213260701301194
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions.
Thousand Oaks, CA: Sage.
Aristotle (367-322 B.C.E. / 2001). On Rhetoric (translated by W. Rhys Robert, edited by Lee
Honeycutt) [Online document]. Retrieved from
http://www.public.iastate.edu/%7ehoneyl/Rhetoric/
Bartsch, A., Appel, M. & Storch, D. (2010). Predicting emotions and meta-emotions at the
movies. The role of the need for affect in audiences' experience of horror and drama.
Communication Research. DOI: 10.1177/009365020356441
Bartsch, A., Vorderer, P., Mangold, R., & Viehoff, R. (2008). Appraisal of emotions in media
use: Toward a process model of meta-emotion and emotion regulation. Media
Psychology, 11, 7-27. doi:10.1080/15213260701813447
Transportation and need for affect
40
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173-1182. doi:10.1037/0022-3514.51.6.1173
Bless, H., Bohner, G., Schwarz, N., & Strack, F. (1990). Mood and persuasion: A cognitive
response analysis. Personality and Social Psychology Bulletin, 16, 331-345.
doi:10.1177/0146167290162013
Bloom, H. (1994). The Western canon: The books and school of the ages. New York: Riverhead
Books.
Bollen, K. A. (1993). Structural equations with latent variables. New York: Wiley.
Braet, A. C. (1992). Ethos, pathos, and logos in Aristotle’s rhetoric: A re-examination.
Argumentation, 6, 307-320. doi:10.1007/BF00154696
Busselle, R., & Bilandzic, H. (2008). Fictionality and perceived realism in experiencing stories:
A model of narrative comprehension and engagement. Communication Theory, 18, 255-
280. doi:10.1111/j.1468-2885.2008.00322.x
Busselle, R., & Bilandzic, H. (2009). Measuring narrative engagement. Media Psychology, 12,
321-347. doi:10.1080/15213260903287259
Chen, S., & Chaiken, S. (1999). The heuristic-systematic model in its broader context. In S.
Chaiken & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 73-96). New
York: Guilford.
Clore, G. L., & Schnall, S. (2005). The influences of affect on attitude. In D. Albarracín, B. T.
Johnson, & M. P. Zanna (Eds.), Handbook of attitudes and attitude change (pp. 437-492).
Mahwah, NJ: Erlbaum.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/
correlation analysis for the behavioral sciences. Mahwah, NJ: Erlbaum.
Transportation and need for affect
41
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper &
Row.
Dal Cin, S., Zanna, M. P., & Fong, G. T. (2004). Narrative persuasion and overcoming
resistance. In E.S. Knowles & J.A. Linn (Eds.), Resistance and persuasion (pp. 175-191).
Mahwah, NJ: Erlbaum.
Deighton, J., Romer, D., & McQueen, J. (1989). Using drama to persuade. Journal of Consumer
Research, 16, 335-343. doi:10.1086/209219
Dijkstra, A. (2008). The psychology of tailoring-ingredients in computer-tailored persuasion.
Social and Personality Psychology Compass, 2, 765-784. doi:10.1111/j.1751-
9004.2008.00081.x
Dillard, J. P., & Nabi, R. L. (2006). The persuasive influence of emotion in cancer prevention and
detection messages. Journal of Communication, 56, 123-139. doi:10.1111/j.1460-
2466.2006.00286.x
Edwards, K. (1990). The interplay of affect and cognition in attitude formation and change.
Journal of Personality and Social Psychology, 59, 202-216. doi:10.1037/0022-
3514.59.2.202
Engeser, S., & Rheinberg, F. (2008). Flow, performance and moderators of challenge-skill
balance. Motivation and Emotion, 32, 158-172. doi:10.1007/s11031-008-9102-4
Escalas, J. E. (2004). Imagine yourself in the product: Mental simulation, narrative
transportation, and persuasion. Journal of Advertising, 33, 37-48.
Escalas, J. E. (2007). Self-referencing and persuasion: Narrative transportation versus analytical
elaboration. Journal of Consumer Research, 33, 421-429. doi:10.1086/510216
Fazio, L. K., & Marsh, E. J. (2008). Slowing presentation time increases, rather than decreases,
errors learned from fictional stories. Psychonomic Bulletin and Review, 15, 180-185.
Transportation and need for affect
42
Gallucci, M. (2005, October). The struggle of the dummies: Beyond ANOVA with categorical
variables. Presentation given at Kurt Lewin Instituut, Vrije Universiteit Amsterdam, The
Netherlands. Slides retrieved August 7, 2007 from http://evol.psico.unimib.it/courses/kli
/2005/Thestruggleofthedummies.ppt
Gawronski, B., & Bodenhausen, G. V. (2007). Unraveling the processes underlying evaluation:
Attitudes from the perspective of the APE model. Social Cognition, 25, 687-717.
doi:10.1521/soco.2007.25.5.687
Gerrig, R. J. (1993). Experiencing narrative worlds: On the psychological activities of reading.
New Haven, CT: Yale University Press.
Gerrig, R. J., & Prentice, D. A. (1991). The representation of fictional information. Psychological
Science, 2, 336-340. doi:10.1111/j.1467-9280.1991.tb00162.x
Gosling, S. D., Vazire, S., Srivastava, S. & John, O. (2004). Should we trust web-based studies?
A comparative analysis of six preconceptions about internet questionnaires. American
Psychologist, 59, 93-104. doi:10.1037/0003-066X.59.2.93
Green, M. C. (2004). Transportation into narrative worlds: The role of prior knowledge and
perceived realism. Discourse Processes, 38, 247-266. doi:10.1207/s15326950dp3802_5
Green, M. C. (2006). Narratives and cancer communication. Journal of Communication, 56, 163-
183. doi:10.1111/j.1460-2466.2006.00288.x
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
Green, M. C., & Brock, T. C. (2002). In the mind’s eye: Transportation-imagery model of
narrative persuasion. In M. C. Green, J. J. Strange, & T. C. Brock (Eds.), Narrative
impact: Social and cognitive foundations (pp. 315-342). Mahwah, NJ: Erlbaum.
Transportation and need for affect
43
Green, M. C., Brock, T. C., & Kaufman, G. F. (2004). Understanding media enjoyment: The role
of transportation into narrative worlds. Communication Theory, 14, 311-327.
doi:10.1111/j.1468-2885.2004.tb00317.x
Green, M. C., Garst, J., & Brock, T. C. (2004). The power of fiction: Persuasion via imagination
and narrative. In L. J. Shrum (Ed.), The psychology of entertainment media: Blurring the
lines between entertainment and persuasion (pp. 161-176). Mahwah, NJ: Erlbaum.
Green, M. C., Garst, J., Brock, T. C., & Chung, S. (2006). Fact versus fiction labeling: Persuasion
parity despite heightened scrutiny of fact. Media Psychology, 8, 267-285.
doi:10.1207/s1532785xmep0803_4
Haddock, G., Maio, G. R, Arnold, K., & Huskinson, T. (2008). Should persuasion be affective or
cognitive? The moderating effects of need for affect and need for cognition. Personality
and Social Psychology Bulletin, 34, 769-778. doi:10.1177/0146167208314871
Johnson, M. K., Hashtroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological
Bulletin, 114, 3-28. doi:10.1037/0033-2909.114.1.3
Jöreskog, K. G., & Sörbom, D. (1996). LISREL 8: User’s reference guide. Chicago, IL: Scientific
Software International.
Kopfman, J. E., Smith, S. W., Ah Yun, J. K., & Hodges, A. (1998). Affective and cognitive
reactions to narrative versus statistical evidence organ donation messages. Journal of
Applied Communication Research, 26, 279-300. doi:10.1080/00909889809365508
Lombard, M., & Ditton, T. B. (1997). At the heart of it all: The concept of presence. Journal of
Computer-Mediated Communication, 13(3). Available: http://www.ascusc.org/jcmc
/vol3/issue2/lombard.html
Transportation and need for affect
44
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A
comparison of methods to test mediation and other intervening variable effects.
Psychological Methods, 7, 83-104. doi:10.1037/1082-989X.7.1.83
Maio, G. R., & Esses, V. M. (2001). The Need for Affect: Individual differences in the
motivation to approach or avoid emotions. Journal of Personality, 69, 583-615.
doi:10.1111/1467-6494.694156
Mar, R. A., & Oatley, K. (2008). The function of fiction is the abstraction and simulation of
social experience. Perspectives on Psychological Science, 13, 173-192.
doi:10.1111/j.1745-6924.2008.00073.x
Marsh, E. J., & Fazio, L. K. (2006). Learning errors from fiction: Difficulties in reducing reliance
on fictional stories. Memory and Cognition, 34, 1140-1149.
Mayer, J. D., & Gaschke, Y. N. (1988). The experience and meta-experience of mood. Journal of
Personality and Social Psychology, 55, 105-111. doi:10.1037/0022-3514.55.1.102
McGuire, W. J. (1969). The nature of attitudes and attitude change. In G. Lindzey & E. Aronson
(Eds.), The handbook of social psychology (pp. 136-314). Cambridge, MA: MIT Press.
Meyer, D. (2005). Es lebe Marrakesch! [Long live Marrakech!] [Online document]. Short story
retrieved April 16, 2007, from http://www.online-roman.de/humor/humor-198.html
Muller, D., Judd, C. M., & Yzerbyt, V. Y. (2005). When moderation is mediated and mediation is
moderated. Journal of Personality and Social Psychology, 89, 852-863.
Nabi, R. L. (1999). A cognitive-functional model for the effects of discrete negative emotions on
information processing, attitude change, and recall. Communication Theory, 9, 292-320.
doi:10.1111/j.1468-2885.1999.tb00172.x
Transportation and need for affect
45
National Kidney Foundation (2008). 25 facts about organ donation and transplantation [Online
document]. Retrieved June 26, 2008, from
http://www.kidney.org/news/newsroom/fsitem.cfm?id = 30
Nell, V. (1988). Lost in a book: The psychology of reading for pleasure. New Haven, CT: Yale
University Press.
Nuland, S. (1994). How we die: Reflections on life's final chapter. New York: Knopf.
Oatley, K. (1994). A taxonomy of the emotions of literary response and a theory of identification
in fictional narrative. Poetics, 23, 53-74. doi:10.1016/0304-422X(94)P4296-S
Paluck, E. L. (2009). Reducing intergroup prejudice and conflict using the media: A field
experiment in Rwanda. Journal of Personality and Social Psychology, 96, 574-587.
doi:10.1037/a0011989
Parisi, N., & Katz, I. (1986). Attitudes toward posthumous organ donation: A critical review and
synthesis of individual and next-of-kin donation decisions. Health Psychology, 16, 183-
195.
Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and peripheral
routes to attitude change. New York: Springer.
Petty, R. E., & Wegener, D. T. (1999). The elaboration likelihood model: Current status and
controversies. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social
psychology (pp. 41-72). New York: Guilford.
Petty, R. E., Fabrigar, L. R., & Wegener, D. T. (2003). Emotional factors in attitudes and
persuasion. In R. J. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Handbook of
affective sciences (pp.752-772). Oxford: Oxford University Press.
Prentice, D. A., Gerrig, R. J., & Bailis, D. S. (1997). What readers bring to the processing of
fictional texts. Psychonomic Bulletin and Review, 4, 416-420.
Transportation and need for affect
46
Richter, T., Schroeder, S., & Wöhrmann, B. (2009). You don’t have to believe everything you
read: Background knowledge permits fast and efficient validation of information. Journal
of Personality and Social Psychology, 96, 538-598. doi:10.1037/a0014038
Rumelhart, D. E. (1975). Notes on a schema for stories. In D. G. Bobrow & A. Collins (Eds.),
Representation and understanding: Studies in cognitive science (pp. 211-236). New York:
Academic Press.
Schroeder, S., Richter, T., & Hoever, I. (2008). Getting a picture that is both accurate and stable:
Situation models and epistemic validation. Journal of Memory and Language, 59, 237-
259. doi:10.1016/j.jml.2008.05.001
Singhal, A., Cody, M. J., Rogers, E. M. & Sabido, M. (Eds.) (2004). Entertainment-Education
and Social Change: History, Research, and Practice. Mahwah, NJ: Erlbaum.
Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations models. In S.
Leinhart (Ed.), Sociological methodology 1982 (pp.290-312). San Francisco: Jossey-Bass.
Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior.
Personality and Social Psychology Review, 8, 220-247.
doi:10.1207/s15327957pspr0803_1
Strange, J. J., & Leung, C. C. (1999). How anecdotal accounts in news and in fiction can
influence judgments of a social problem's urgency, causes, and cures. Personality and
Social Psychology Bulletin, 25, 436-449. doi:10.1177/0146167299025004004
Tellegen, A., & Atkinson, G. (1974). Openness to absorbing and self-altering experiences
("absorption"): A trait related to hypnotic susceptibility. Journal of Abnormal Psychology,
83, 268-277. doi:10.1037/h0036681
Underwood, B. J. (1975). Individual differences as a crucible in theory construction. American
Psychologist, 30, 128-134. doi:10.1037/h0076759
Transportation and need for affect
47
Vaughn, L. A., Hesse, S. J., Petkova, Z., & Trudeau, L. (2009). “The story is right on”: The
impact of regulatory fit on narrative engagement and persuasion. European Journal of
Social Psychology, 39, 447-456. doi:10.1002/ejsp.570
Wheeler, S. C., Green, M. C., & Brock, T. C. (1999). Fictional narratives change beliefs:
Replications of Prentice, Gerrig & Bailis (1997) with mixed corroboration. Psychonomic
Bulletin and Review, 6, 136-141.
Wirth, W. (2006). Involvement. In J. Bryant & P. Vorderer (Eds.), Psychology of entertainment
(pp. 199-213). Mahwah, NJ: Erlbaum.
Wyer, R. S., Adaval, R., & Colcombe, S. J. (2002). Narrative-based representations of social
knowledge: their construction and use in comprehension, memory, and judgment. In M. P.
Zanna (Ed.), Advances in experimental social psychology. Vol. 34 (pp. 131-197). New
York: Academic Press. doi:10.1016/S0065-2601(02)80005-3
Zillmann, D. (1991). Empathy: Affect from bearing witness to the emotions of others. In J.
Bryant & D. Zillmann (Eds.), Responding to the screen: Reception and reaction processes
(pp. 135-168). Hillsdale, NJ: Erlbaum.
Zwaan, R. A., & Radvansky, G. A. (1998). Situation models in language comprehension and
memory. Psychological Bulletin, 123, 162-185. doi:10.1037/0033-2909.123.2.162
Transportation and need for affect
48
Footnotes
1
We are grateful to Marcello Gallucci (2005) for pointing out this type of mediated moderation
model to us.
2
Although not based on our theory, at least two alternative mediated moderation (or moderated
mediation) models are conceivable and statistically testable. The first of these alternative models
assumes an interaction between text and need for affect to predict transportation. The second one
assumes an interaction between need for affect and transportation to predict beliefs.
According to the first alternative model, the text factor would moderate the impact of
need for affect on transportation as the criterion variable. To test for this interaction,
transportation was regressed on text (effect-coded), need for affect (z-standardized), and the
product of both variables. Need for affect predicted transportation (B = 0.23, SE
B
= 0.08, p < .01,
R
2
= .06) irrespective of the text presented, as indicated by a non-significant interaction term (B
= 0.02, SE
B
= 0.09, p = .82, R
2
= .00). According to the second alternative model, need for
affect would moderate the impact of transportation on beliefs as the criterion variable. To test for
this interaction, the belief score was regressed on transportation (z-standardized) and need for
affect (z-standardized), and the interaction term of both variables. The interaction between both
variables did not have a significant effect on beliefs (B = -0.06, SE
B
= 0.06, p = .17, R
2
= .01).
Thus, neither the first nor the second alternative model was supported by our data.
Transportation and need for affect
49
Table 1:
Means, Standard Deviations and Intercorrelations of All Variables in Experiments 1 and 2
M SD 1 2 3 4 5
Experiment 1
1 Story (control vs. experimental)
a
-0.07 1.00
2 Need for Affect 0.76 0.85 -.03
3 Transportation 3.83 0.95 .41*** .23*
4 Story X Need for Affect
b
-0.02 1.00 -.00 -.05 .02
5 Story X Transportation
b
0.41 0.91 .03 .02 -.20 .26**
6 Belief Index (Dependent Variable) 4.65 1.12 .32*** .00 .29*** .19** .23**
Experiment 2
1 Story (emotional content low vs. high)
a
-0.14 0.99
2 Need for Affect 0.85 0.91 .07
3 Transportation 3.51 0.95 .10 .25**
4 Story X Need for Affect
b
0.06 1.02 .06 -.13 -.01
5 Story X Transportation
b
0.12 0.99 -.09 -.02 .10 .24**
6 Belief Index (Dependent Variable) 3.76 0.64 -.06 .21* .28** .18* .27**
Note. Experiment 1: N = 181, Experiment 2: N = 133
a
Contrast coding (-1 vs. 1),
b
Need for Affect and Transportation were z-standardized for computing the
interaction terms.
*** p < .001, ** p < .01, * p < .05 (one-tailed).
Transportation and need for affect
50
Table 2:
Summary of Hierarchical Regression Analyses (main effect and interaction terms) including
Transportation (Model 2a), Need for Affect (Model 2b), and both Need for Affect and Transportation
(Model 3) as Moderators of Narrative Persuasion
Experiment 1 Experiment 2
B SE
B
R
2
B SE
B
R
2
Model 1
Intercept (B
0
) 4.67 0.08 3.75 0.06
Story (control vs. experimental /
low vs. high emotional content)
a
0.36*** 0.08 .10 -0.04 0.06 .00
Model 2a
Intercept (B
0
) 4.53 0.08 3.75 0.05
Story (control vs. experimental /
low vs. high emotional content)
a
0.23** 0.08 .04 -0.04 0.05 .00
Transportation
b
0.28** 0.09 .05 0.17** 0.05 .07
Story X Transportation
0.35*** 0.09 .07 0.15** 0.05 .06
Model 2b
Intercept (B
0
) 4.68 0.08 3.74 0.05
Story
a
0.36*** 0.08 .10 -0.06 0.05 .01
Need for Affect
b
0.02 0.08 .00 0.15** 0.05 .06
Story X Need for Affect
0.21** 0.08 .04 0.14* 0.05 .05
Model 3
Intercept (B
0
) 4.53 0.08 3.74 (0.05)
Story
a
0.24** 0.09 .03 -0.05 0.05 .00
Transportation and need for affect
51
Need for Affect
b
-0.14 0.08 .00 0.12* 0.05 .03
Transportation
b
0.27*** 0.09 .04 0.14* 0.05 .05
Story X Need for Affect
0.14 0.08 .01 0.10 0.05 .02
Story X Transportation
0.31*** 0.09 .05 0.13* 0.05 .04
Note. Model fit, Experiment 1: Model 1: R
2
= .10, F(1,179) = 20.6, p < .001; Model 2a: R
2
= .20,
F(3,177) = 14.7, p < .001; Model 2b: R
2
= .14, F(3,177) = 9.4, p < .001; Model 3: R
2
= .22, F(5,175) =
9.6, p < .001.
Model fit, Experiment 2: Model 1: R
2
= .00, F(1,131) = 0.4, p = .52; Model 2a: R
2
= .14, F(3,129) = 7.1,
p < .001; Model 2b: R
2
= .10, F(3,129) = 4.5, p < .01; Model 3: R
2
= .19, F(5,127) = 6.0, p < .001
a
Contrast-coded (Experiment 1: control story without emotional details = -1, experimental story with
emotional details = 1; Experiment 2: story with low emotional content = -1; story with high emotional
content = 1).
b
z-standardized.
*** p < .001, ** p < .01, * p < .05 (one-tailed).
Transportation and need for affect
52
Figure captions
Figure 1. Moderator model for the distal predictor (need for affect) (a), moderator model for the
proximal predictor (transportation) (b), and mediated moderation model (c).
Figure 2. Estimates of the simple slopes (with standard errors) of the effect of transportation (a)
and need for affect (b) in the groups who read the experimental story or the control story in
Experiment 1 (*** p < .001; * p < .05).
Figure 3. Structural equation model (hypothesized model, multi-sample-analysis, completely
standardized solution) for the data from Experiment 1 (Figure 3a) and Experiment 2 (Figure 3b).
The path from transportation to beliefs was allowed to vary between the experimental group and
the control group (Experiment 1) and the group who received the story with high emotional
content and the group who received the story with low emotional content (Experiment 2). Factor
loadings and the path from need for affect to transportation was constrained to be equal across
groups.
Figure 4. Estimates of the simple slopes (with standard errors) of the effect of transportation (a)
and need for affect (b) in the groups who read the story with high emotional content or the story
with low emotional content in Experiment 2 (*** p < .001).
Transportation and need for affect
53
Figure 1
a)
b)
c)
Beliefs
Need for
Affect
Fictional narrative
Beliefs
Transporta
tion
Fictional narrative
Beliefs
Need for
Affect
Transport
ation
Fictional narrative
Transportation and need for affect
54
Figure 2
a)
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
B=0.61 (SE
B
=0.13)***
B=-0.05 (SE
B
=0.11)
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
0.61 (0.13)
***
-0.05 (0.11)
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
B=0.61 (SE
B
=0.13)***
B=-0.05 (SE
B
=0.11)
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
B=0.61 (SE
B
=0.13)***
B=-0.05 (SE
B
=0.11)
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
3,5
4,5
5,5
M - 1SD M + 1 SD
Transportation
Belief Index
Experimental
Control
Story
0.61 (0.13)
***
0.61 (0.13)
***
-0.05 (0.11)-0.05 (0.11)
b)
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
B=0.23 (SE
B
=0.11)*
B=-0.19 (SE
B
=0.11)
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
0.23 (0.11)*
-0.19 (0.11)
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
B=0.23 (SE
B
=0.11)*
B=-0.19 (SE
B
=0.11)
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
B=0.23 (SE
B
=0.11)*
B=-0.19 (SE
B
=0.11)
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
3,5
4
4,5
5
5,5
M - 1SD M + 1 SD
Need for Affect
Belief Index
Experimental
Control
Story
0.23 (0.11)*
-0.19 (0.11)
Transportation and need for affect
55
Figure 3
Need for Affect Transportation Beliefs
Par-
cel 1
Par-
cel 2
.86 .84
(.30)(.27)
Par-
cel 1
Par-
cel 2
.90 .86
(.25)(.18)
Par-
cel 1
Par-
cel 2
.88 .87
(.24)(.22)
.28* .02/.56**
Need for Affect Transportation Beliefs
Par-
cel 1
Par-
cel 2
.90 .87
(.24)(.19)
Par-
cel 1
Par-
cel 2
.86 .90
(.19)(.26)
Par-
cel 1
Par-
cel 2
.76 .66
(.57)(.43)
.32* -.06/.67**
b) Experiment 2
a) Experiment 1
Transportation and need for affect
56
Figure 4
a)
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
B=0.01 (SE
B
=0.08)
B= 0.32 (SE
B
=0.07)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
0.01 (0.08)
0.32 (0.07)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
B=0.01 (SE
B
=0.08)
B= 0.32 (SE
B
=0.07)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
B=0.01 (SE
B
=0.08)
B= 0.32 (SE
B
=0.07)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
3
3.5
4
4.5
M - 1SD M + 1 SD
Transportation
Belief Index
High emotion
content
Low emotion
content
Story
0.01 (0.08)
0.32 (0.07)***
b)
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
B=0.02 (SE
B
=0.07)
B= 0.29 (SE
B
=0.08)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
0.02 (0.07)
0.29 (0.08)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
B=0.02 (SE
B
=0.07)
B= 0.29 (SE
B
=0.08)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
B=0.02 (SE
B
=0.07)
B= 0.29 (SE
B
=0.08)***
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
3
3.5
4
4.5
M - 1SD M + 1 SD
Need for Affect
Belief Index
High emotion
content
Low emotion
content
Story
0.02 (0.07)
0.29 (0.08)***
... Similar to personality traits, it is regarded as a relatively stable intrinsic character of human nature (1,2). NFA could help to explain individuals' differences in many critical mental aspects such as mental health (3)(4)(5)(6)(7)(8)(9), information preferences and decision making (10)(11)(12)(13)(14), social attitude [e.g., evaluation of self and others (15)(16)(17)(18), attitudes toward brands (19), attitudes toward job (20), attitudes toward the country (21), drugs (22)] and so on. It has also been found that NFA influences individual reactions to media and entertainment (23,24), political beliefs and ideology (16,25), legal decisions (26) and risk taking capacity (27). ...
Article
Full-text available
The need for affect (NFA), which refers to the motivation to approach or avoid emotion-inducing situations, is a valuable indicator of mental health monitoring and intervention, as well as many other applications. Traditionally, NFA has been measured using self-reports, which is not applicable in today's online scenarios due to its shortcomings in fast, large-scale assessments. This study proposed an automatic and non-invasive method for recognizing NFA based on social media behavioral data. The NFA questionnaire scores of 934 participants and their social media data were acquired. Then we run machine learning algorithms to train predictive models, which can be used to automatically identify NFA degrees of online users. The results showed that Extreme Gradient Boosting (XGB) performed best among several algorithms. The Pearson correlation coefficients between predicted scores and NFA questionnaire scores achieved 0.25 (NFA avoidance), 0.31 (NFA approach) and 0.34 (NFA total), and the split-half reliabilities were 0.66–0.70. Our research demonstrated that adolescents' NFA can be identified based on their social media behaviors, and opened a novel way of non-intrusively perceiving users' NFA which can be used for mental health monitoring and other situations that require large-scale NFA measurements.
... Empirical studies have confirmed the intuition that fictional stories can influence attitudes and behaviors (e.g., Appel & Richter, 2010Green & Brock, 2000;Gerrig & Prentice, 1991). For example, watching violence on television can cause people to become more fearful of being victimized (Shanahan & Morgan, 1999). ...
Article
Stories have a powerful ability to shape our beliefs, attitudes, opinions, and knowledge about the world. In the current work, we ask how readers evaluate the truth of facts embedded in fiction. In three experiments, we investigate the influence of the credibility of the story's narrator on the likelihood that readers encode and recall misinformation contained in the narrative. Participants read stories containing accurate real-world facts and misleading lures. The stories were narrated by either a credible or a non-credible narrator. Following the stories, participants were tested for the critical story information with a free response test of their general knowledge (Experiments 1 and 2) or with a speeded true-false test (Experiment 3). Overall, narrator credibility had no influence on readers' memory for accurate information. However, readers were more likely to reproduce and affirm misinformation when it was delivered by a credible than a non-credible narrator. The current studies suggest that the credibility and the expertise of the source of the information are critical in determining what readers remember and believe.
... In the area of persuasion, individuals with a strong NFA appear to be more sensitive and easily persuaded by messages with strong emotional content (Haddock et al., 2008; see also Sundar et al., 2015). Another study, by Appel and Richter (2010), showed that the extent to which people tended to be persuaded by a message depended on their motivation to approach emotional events, this effect being mediated by the extent people self-report transportation into narrative worlds (i.e. are focused on events occurring in the narrative). Building on this result, Appel and Malečkar (2012), showed that a high NFA led people to be more influenced by narrative persuasion than a low NFA. ...
Article
Introduction The way we interact with our environment depends on our spontaneous tendency to approach or to avoid emotional experiences triggered by that environment. This dimension of the emotional experience is called the need for affect, that is, the tendency of individuals to adopt approaching or avoidance behaviour with regard to emotional stimuli. Methods The Need For Affect (NFA) Scale has been the subject of numerous studies since the validation of the original version (Maio & Esses, 2001) and its short version (Appel et al., 2012). However, no validation of the latter scale has been conducted in French. We propose a French version of the short NFA scale on a student sample and a sample from the general population. Results We found the structure of the original scale in a French translation (of the English version). In addition, invariance tests showed that this structure remained the same for both samples. Conclusion We recommend the use of this version of the short NFA scale for studies conducted on French-speaking samples.
... As a consequence, readers experience emotional reactions to story events (even when readers know the events are fictional; Busselle & Bilandzic, 2009;Green & Brock, 2000). Still, research in this tradition often focuses on the general emotional intensity of the reader's experience rather than the nature of the emotional experience (e.g., that of anger or joy; Appel & Richter, 2010;Green et al., 2012). We wish to shift more attention to the discrete nature of those experiences. ...
Article
Emotional responses are a central feature of readers’ narrative experiences. Situations in which readers adopt characters’ goals and experience similar emotional reactions to story events are often the focus of research on readers’ experiences of stories. However, readers may understand (or appraise) story events in a way that differs from the main character, and may consequently feel different emotions. In the current work, we leverage an appraisal theory perspective to clarify conditions under which readers experience emotions that mirror characters’ emotional responses to story events, as well as conditions under which readers experience distinct emotions. Study 1 examined readers’ experience of anger toward different story characters. Study 2 examined readers’ experience of sadness or fear for one story character. Results suggest that readers appraise the event from both a character’s perspective as well as their individual view, which generally translates into the experience of emotions that correspond with both the character’s appraisals and their own.
... Pioneering work on transportation and persuasion promises that the theory can also be applied to the digital ad avoidance field. To illustrate, transportation can reduce underlying cognitive and argumentative tasks that counteract persuasion (Appel & Richter, 2010;Moyer-Gusé & Nabi, 2010). Consumers experiencing transportation also feel strong emotional arousal and become identified with the story characters (Green, 2006;Ratcliff & Sun, 2020;Wang & Calder, 2006). ...
Article
Full-text available
The recent growth in digital marketing investments and revenues has attracted the attention of both marketing practitioners and scholars. However, this growth has dramatically increased users' exposure to ad messages, encouraging consumers to avoid them. Therefore, ad avoidance has become a major problem for marketing practitioners. Although researchers have become much more interested in this subject over the past two decades, the body of knowledge on ad avoidance in the digital environment remains fragmented due to the lack of a comprehensive review. Therefore, a holistic overview study is needed that focuses on the big picture and can help researchers to understand the literature comprehensively. This study aims to provide a comprehensive understanding of the topic using a systematic literature review approach on digital ad avoidance. To this end, we provide an in-depth content analysis of 56 relevant articles published in 31 peer-reviewed scientific journals up to December 31, 2021. Based on a theories, contexts, characteristics, and methods (TCCM) framework, the study results shed light on “what do we know, how do we know, and where should research about digital ad avoidance research be heading?” Additionally, drawing on the content analysis, we have presented an integrative framework that considers antecedents, outcomes, mediators, and moderators, which can help develop the field systematically and guide future research. By doing so, we think this review meets the need to give an overview of the state-of-the-art scientific body of knowledge on digital ad avoidance and makes important and solid contributions to the literature, practical implications, and future research directions based on the findings.
Article
Purpose The increasing popularity of immersive activities (e.g. immersive performing art (IPA) at tourism destinations calls for the need to understand customers’ immersive experiences and its impact on their attitudes and behavioral intentions. This study aims to conceptualize customers’ interactive experience in IPA and systematically examine how interaction in IPA experience shapes customers’ word-of-mouth (WOM) intention and thus generates business values for destination managers. Design/methodology/approach A two-step mixed-methods approach was used. The qualitative study was adopted to identify the salient dimensions of interaction in IPA experience and develop a context-specific measurement scale. Structural equation modeling approach was applied to test the theoretical model using partial least squares 3.3.0. Findings The results indicated that the three dimensions of interaction in IPA experience (i.e. parasocial interaction, narrative interaction and environmental interaction) are positively related to customers’ experiences of narrative transportation and self-expansion related to the IPA, which further influence their WOM intention. Originality/value This study considers IPA as an emerging form of immersive tourism activities and conceptualizes its interactive nature, and contributes to the understanding of how customers’ interactive experience helps them construct meanings through narrative transportation, and offers valuable guidance for IPA designers and destination management organizations.
Chapter
When we think of everyday language use, the first things that come to mind include colloquial conversations, reading and writing e-mails, sending text messages or reading a book. But can we study the brain basis of language as we use it in our daily lives? As a topic of study, the cognitive neuroscience of language is far removed from these language-in-use examples. However, recent developments in research and technology have made studying the neural underpinnings of naturally occurring language much more feasible. In this book a range of international experts provide a state-of-the-art overview of current approaches to making the cognitive neuroscience of language more 'natural' and closer to language use as it occurs in real life. The chapters explore topics including discourse comprehension, the study of dialogue, literature comprehension and the insights gained from looking at natural speech in neuropsychology.
Article
This study investigated the efficacy of narratives in countering vaccine-related narrative misinformation on Facebook. Two narrative counter strategies to address misinformation about the safety of the MMR vaccine (i.e., personal stories of uneventful vaccinations where a parent talks about how their child was vaccinated and no injury happened to them, and narratives of anti-vax to pro-vax conversions, where a former anti-vaccine parent narrates their experience of their child getting sick with a vaccine-preventable disease and thus realizing the importance of vaccines) were compared to a non-narrative counter message (an infographic, in the case of the present study) in increasing belief in the counter, reducing vaccine misinformation, and increasing MMR vaccination intentions. Second, we tested a theoretically grounded model predicting MMR vaccination intentions and misinformation amplification likelihood, combining pre-existing issue-specific situational motivation in problem solving with perceived counter and misinformation source credibility and message effects. Implications for public relations theory are discussed.
Article
This project tests the ability of narrative and gameplay to persuade in a casual health game, through two experimental studies. Study 1 (N = 212) explores how independently manipulated narrative and gameplay factors can persuade people to have healthier attitudes and behavioral intentions; Study 2 (N = 353) also investigates how narrative and gameplay (difficulty) factors might improve attitudes and behavioral intentions toward an in-game brand. Both studies consider the role of flow, an immersive and inherently rewarding psychological state, as potential mediators between game factors and persuasive outcomes. In both studies, results show improvement in both attitudes and behavioral intentions toward health and the brand; however, the cause of those changes is not clear, as the manipulated narrative and gameplay factors do not influence the persuasive outcomes, and the mediating variable flow influences the persuasive outcomes inconsistently.
Article
Full-text available
A Monte Carlo study compared 14 methods to test the statistical significance of the intervening variable effect. An intervening variable (mediator) transmits the effect of an independent variable to a dependent variable. The commonly used R. M. Baron and D. A. Kenny (1986) approach has low statistical power. Two methods based on the distribution of the product and 2 difference-in-coefficients methods have the most accurate Type I error rates and greatest statistical power except in 1 important case in which Type I error rates are too high. The best balance of Type I error and statistical power across all cases is the test of the joint significance of the two effects comprising the intervening variable effect.
Article
Full-text available
In this article, we attempt to distinguish between the properties of moderator and mediator variables at a number of levels. First, we seek to make theorists and researchers aware of the importance of not using the terms moderator and mediator interchangeably by carefully elaborating, both conceptually and strategically, the many ways in which moderators and mediators differ. We then go beyond this largely pedagogical function and delineate the conceptual and strategic implications of making use of such distinctions with regard to a wide range of phenomena, including control and stress, attitudes, and personality traits. We also provide a specific compendium of analytic procedures appropriate for making the most effective use of the moderator and mediator distinction, both separately and in terms of a broader causal system that includes both moderators and mediators. (46 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Manche mögen's heiß. Ergebnisse der deutschsprachigen Version eines Instruments zur Erfassung des Emotionsmotivs (need for emotion / need for affect) Some like it hot. Results of the German language version of an instrument aimed at the assessment of the need for affect/need for emotion Bitte zitieren als / Please cite as Appel, M. (2008). Manche mögen´s heiß. Ergebnisse der deutschsprachigen Version eines Instruments zur Erfassung des Emotionsmotivs (need for emotion / need for affect) [Some like it hot. Results of the German language version of an instrument aimed at the assessment of the need for affect/need for emotion.] Diagnostica, 54, 2-15.
Article
Drawing from cognitive response persuasion theories, functional emotion theories, and theoretical and empirical work on the influence of message-relevant and message-irrelevant affect on attitudes, this dissertation presents a cognitive-functional model of persuasion (CFM) designed to broaden the limited theorizing available on the persuasive effects of emotions. In general, the CFM suggests that discrete negative emotions influence attitudes through a process driven by three initial responses: motivated attention, motivated processing, and expectation of reassurance.^ To test the model, a 3 (Emotion Type: fear, anger, or none) x 2 (Expectation of Message Reassurance: certain or uncertain) between subjects x 2 (Story topic: domestic terrorism and juvenile crime) within subjects full-factorial experiment was conducted. 236 undergraduates read fictitious news stories designed to evoke either anger or fear about domestic terrorism and juvenile crime along with persuasive news stories on those topics. Emotional arousal was measured after each emotion story, and thought-listing, attitude, message evaluation, and information recall measures were taken after each persuasive story. The study concluded with an information recognition and reaction time task. Follow-up recall and attitude measures were obtained two weeks later.^ The domestic terrorism results offered some promising support for the model. Most notable, fear and anger prompted different levels of persuasion in directions consistent with hypotheses as did the two levels of certainty in expectation of message reassurance. The juvenile crime results were less encouraging. The effects of fear and anger on attitudes were inconsistent with hypotheses, and certainty in expectation of message reassurance was not influential. The contradictions between the two topics' results are explained by differences in message topic familiarity. In response to these and other findings, a revised version of the CFM is presented.^ In sum, the CFM attempts to bridge the theoretical gap between "emotional" and "rational" approaches to persuasion. This research offers evidence that discrete negative emotions can lead to different persuasive outcomes and that motivated attention, motivated processing, expectation of message reassurance, and issue familiarity can be influential in that process. The notion that emotions act as frames for new stories and its implications for information processing and recall is also discussed.
Article
We compare the relative contribution of bacteria in four suspended and benthic habitats to total system bacterial standing stocks, production, and respiration in the Ogeechee River, a sixth order, low-gradient, blackwater river. The habitats considered are water column, sandy main channel sediment, backwater sediment, and snags. Total system bacterial biomass ranged widely with a mean of 3240 mg C m-2. It was greatest during winter and early summer, and peaks were dominated by bacteria in sandy sediments where from 21 to 98% of total bacterial biomass was found. Bacterial biomass in both backwater and snag habitats remained consistently low. Dye studies revealed considerable water movement into sandy sediments, which were aerobic to the greatest depth sampled (60 cm). System bacterial production was less variable than biomass, ranging from 0.84 to 8.54 mg C m-2 n-1 >80% of which occurred in main channel sediments. System bacterial production could not account for system biomass; hence allochthonous sources of bacteria appear to be important. Bacterial respiration was calculated from rates of production and growth efficiencies measured previously in laboratory studies. The calculated contribution of bacterial respiration to measured system respiration had a modal value of 97%, but varied widely (from 9 to 161%). Benthic bacteria appear to be a major driver of system respiration, although uncertainties in these calculations point to the critical need for a better understanding of temporal variability in bacterial growth efficiency in nature.
Article
The present article provides an analysis of the attitude construct from the perspective of the Associative-Prepositional Evaluation Model (APE Model). It is argued that evaluative responses should be understood in terms of their underlying mental processes: associative and propositional processes. Whereas associative processes are characterized by mere activation, independent of subjective truth or falsity, propositional reasoning is concerned with the validation of evaluations and beliefs. Associative processes are claimed to provide the basis for primitive affective reactions; propositional processes are assumed to form the basis for evaluative judgments. Implications of this conceptualization for a variety of questions are discussed, such as automatic features of attitudes, processes of attitude formation and change, attitude representation in memory, context-sensitivity and stability of attitudes, and the difference between personal and cultural evaluations.