See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/273632922

Effects of Literature on Empathy and Self-Reflection: A Theoretical-Empirical

Framework

Article in Journal of Literary Theory · March 2015

DOI: 10.1515/jlt-2015-0005,

CITATIONS READS

94 12,305

2 authors:

Emy Koopman Frank Hakemulder

previously: Erasmus University Rotterdam; currently: independent Utrecht University
15 PUBLICATIONS 502 CITATIONS 59 PUBLICATIONS 1,170 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Emy Koopman on 16 March 2015.

The user has requested enhancement of the downloaded file.

[AUTHOR VERSION]

Eva Maria (Emy) Koopman & Frank Hakemulder

Effects of Literature on Empathy and Self-Reflection: A Theoretical-Empirical Framework

In a time when the biggest bestsellers are about crime and erotic sadomasochism, the idea that
reading literature can make us better human beings may seem farfetched. However, ever since
Aristotle’s Poetics (1987; orig. around 335 BCE), authors, critics, and academics have made
claims concerning the ethical potential of narrative drama and poetic language (e.g., Althusser
1983; Booth 1988; Boyd 2009; Bronzwaer 1986; De Botton 1997; Habermas 1983; Hunt
2007; Nussbaum 1995; 1997; 2001; 2010; Pinker 2011; Sontag 2007; Van Peer 1995). The
general claims are that reading literature may enhance self-knowledge, make people more
aware of the plights of those suffering, and more willing to take action to help them. In recent
years, the philosopher Martha Nussbaum has defended the ethical power of literature most
ardently. Reading literature, she says, triggers a type of imagination that is “an essential
ingredient of an ethical stance that asks us to concern ourselves with the good of other people
whose lives are distant from our own” (1995, xvi). Through this “literary imagination,”
readers learn to put themselves in the place of people they could not have known that
intimately in any other way, thus deepening their understanding and compassion. As
Nussbaum (2001, 2) further argues, literary texts can lead to the sort of “self-examination”
that is crucial to ethical decision-making. Often this is reflection on the self, in relation to
others. Knowing how we might respond to certain situations might help us understand how
others would feel as well (cf. Johnson’s 1993 conception of “moral imagination,” and
extensive use of the term elsewhere, e.g., Beran 1998; Guroian 1996; Hutchison 2004).
These are just a few of the reasons why empathy and self-reflection make an
interesting couple to focus on when studying the effects of reading literary texts. To various
scholars, the claims about the relevance of this duo and their relation with literary imagination
are far-reaching rather than farfetched. Hunt (2007), for instance, proposes that fiction has
contributed to a mindset that enabled people to think up a concept like human rights. Novels
such as Rousseau’s Julie (1761) and Richardson’s Pamela (1740) stimulated readers, she
says, to empathize across borders of class, sex, and nation. “As a consequence,” Hunt claims,
“they came to see others – people they did not know personally – as like them, as having the
same kinds of inner emotions. Without this learning process, ‘equality’ could have no deep
meaning and in particular no political consequence” (ibid., 40). She illustrates her statements
with examples of responses of contemporary readers. Pinker (2011) takes this line of
argument even further, suggesting that the spectacular increase in availability and
consumption of narratives in our history might have caused an increase in empathic ability
and, subsequently, a decline in violence.
The relevance of literature’s effect on self-reflection is maybe less self-evident, but
that does not make the claims less sweeping. Among others, Althusser (1980), Habermas
(1983), and Bronzwaer (1986) have argued that literature’s polyvalence can lead us to reflect

1
on our own norms, values, and prejudices, and that this would ultimately benefit society.
Indeed, awareness of self and others may be key to our happiness, social success (Cooper &
Sawaf 2003; Goleman 1995), and even productivity (Ibarra/Barbulescu 2010, see
Bal/Veltkamp 2013). But is it indeed the case that a seemingly idle pass-time activity like
literary reading can do all that? And if so, how can we explain such an influence? Would the
effects be particular to unique literary text qualities or to other aspects that literary texts share
with other genres (e.g., narrativity)?
The purpose of this article is to construct a model for such effects of literary
narratives. It will reveal those questions in the research that are not addressed as of yet, and
will synthesize the available approaches to literary effects. In our attempt to construct an
explanatory model for literary imagination, we first evaluate the empirical indications that
reading literature affects empathy in its various forms (Section 2.1). We will see that it
remains unclear how such effects come about, but that recent studies help us to make some
significant progress in understanding the workings of literary imagination (Section 2.2). In
Section 3 we will do the same for reflection. Finally, we will synthesize our findings in our
explanatory model in Section 4. We will make use of earlier overviews (Hakemulder 2000;
Keen 2007; Kimmel 1970; Klemenz-Belgardt 1981; Mar/Oatley/Djikic/Mullin 2011), adding
other and more recent empirical work that will help us to build a synthetic framework for
future research. Yet, before all this, we need to define the key terms (Section 1).

1. Key terms: Literature – empathy – self-refection

Defining the key terms of our present undertaking is tricky. Arguably the easiest (read: least
complicated) concept to define is reflection. It may be conceptualized in many different ways:
thoughts about the human condition, the faults in the personality of others, cultural
differences, author intentions or his or her biography, and things that seem totally unrelated.
Here we use the term “reflection” or “self-reflection” to designate thoughts and insights on
oneself, often in relation to others, and/or society (in the present context of course evoked by
reading). While we are thus speaking of a mostly cognitive process of generating (new)
thoughts, since the self is implicated, affect-loaded memories are likely to be involved. We
will discuss the most relevant studies (Section 3), looking at thoughts triggered by reading,
devoting special attention to idiosyncratic aspects of reading literature.
But what is “literature”? The choices we can make here are debatable. According to
the sociological approach, literature might be nothing more than what we were brought up to
believe it to be, a matter of convention, perception, and socially determined status (e.g.,
Bourdieu 1996; Corse/Westervelt 2002; Fish 1980; Janssen 1997; Van Rees 1989). When
theorizing about the effects of literature, however, two other, text-immanent, definitions are
used. For the purposes of the present paper, these are more useful. We distinguish (1) a broad,
inclusive definition of literature, and (2) a narrow, exclusive one. The first pertains to
“narrative texts”, that is, texts presenting a sequence of events in which one or more
characters are involved. The second pertains to particular text qualities that distinguish literary
texts from other genres (“literariness”). “Literariness” has been typically conceptualized as a
combination of the aesthetic and the unconventional. Some argue it can be pointed out in
textual features that depart from ordinary language use (“foregrounding”; Miall/Kuiken 1994,
1999; Mukařovský 1976; Shklovsky 1965). This perspective also acknowledges that

2
literariness is ultimately a historical and contextually determined concept rather than a
universal, unchangeable quality of texts. What are unconventional, novel, and deviating ways
of representing at one moment in time for a certain population of readers, may be cliché-like,
trivial, uninteresting at other times, or for other readers. Since theorists who argue for the
ethical effects of literature seem to use both the inclusive and the exclusive definition, we will
deal with both, but when we explicitly speak of “literary texts,” we refer to the narrow sense.
This distinction between narrativity and literariness is essential in our discussion of the
research.
The term “empathy” is equally difficult to conceptualize. The term “empathy” only
exists in English since the early twentieth century, introduced by Titchener (1909) as a
translation of the German “Einfühlung”. Before that time, the term “sympathy” was used to
denote processes of feeling the pain or joy of a fellow human being (see Keen 2007). It is thus
not surprising that “empathy” and “sympathy” are often used interchangeably. Recently,
however, scholars have pleaded to distinguish the two, with “empathy” designating
experiencing emotions perceived as similar to the character(s) (“feeling with”), and
“sympathy” designating feeling concern for another without feeling what the other feels
(“feeling for”; e.g., Busselle/Bilandzic 2009; Coplan 2004; Keen 2006; Mar/Oatley 2008; Mar
et al. 2011). The question is to what extent readers experience this theoretical difference in
practice. Nevertheless, where researchers have made the distinction between sympathy and
empathy with characters or made a clear choice for either one of the two phenomena, we will
make that clear in our own discussion.
We also need to be specific about the type of empathy we are talking about. Davis
(1980; 1983), who defined empathy as “the notion of responsivity to the experiences of
another” (1980, 3), has argued that there is both a cognitive, perspective-taking side and an
emotional reactivity side to empathy. Davis (ibid.) thus distinguishes between “cognitive
empathy” (the ability to understand a character’s perspective, cf. “theory of mind”, see
Leverage/Mancing/Schweickert/Marston William 2011) and “emotional empathy” (feeling
similar emotions to a character, cf. “emotional contagion,” see Hatfield/Cacioppo/Rapson
1994). Similar distinctions have been made by several other scholars (e.g., Cohen 2001;
Decety/Jackson 2006; Zillmann 1994). Furthermore, this distinction is supported by
evolutionary theory (De Waal 2007), as well as by neurological evidence (Shamay-
Tsoory/Aharon-Peretz/Perry 2009). The latter suggests that cognitive and emotional empathy
are mediated by different brain structures: the emotional and cognitive components of
empathy appear to be working autonomously. Even so, Shamay-Tsoory et al. (ibid., 625)
emphasize that “every empathic response will evoke both components to some extent”.
Likewise, Nathanson (2003) has argued for the interdependence of the affective and cognitive
aspects of empathy. In most cases it is unlikely to feel something similar to what someone
else is feeling without being able to take that person’s perspective. The possible distinction
between cognitive and affective empathy will become relevant later on in our discussion and
the construction of our synthetic model, as we will see different studies making use of
different measures assessing either the one or the other form of empathy.
A final distinction needs to be made between trait and state empathy, and empathic
ability as an after-effect. First, before starting to read a text, readers already have a certain
disposition when it comes to their empathic sensitivity to others. This disposition, “trait

3
empathy,” is a personality variable, that could, as some have argued
(Mar/Oatley/Hirsch/Delapaz/Peterson 2006; Mar/Oatley/Peterson 2009), have been developed
more strongly because of one’s exposure to narrative, but it also could be independent of
reading behavior. Second, during reading, readers can have the cognitive and affective
empathic responses towards characters that have been discussed above. This type of empathy
can be called “narrative empathy” (after Keen 2007), and it is linked to the broader concept
“narrative emotions” (see Kneepkens/Zwaan 1994; Miall/Kuiken 2002; Tan 1996), which
consists of all emotions toward the narrative world, including empathy with characters. Third,
after reading, cognitive and affective empathic responses to living beings who are similar to
the depicted characters can occur (“real-life empathy” as an after-effect), or there could be
(for some period of time) a more general increase in one’s empathic ability. Differentiating
between these types of empathy may be useful, since claims about the effects of reading on
empathy would need to take into account how far-reaching and long-lasting these effects are,
if they occur at all. It does seem possible that we cry in response to the tragedies that befall
characters, but fail to sympathize with a neighbor in need. Obviously this might curb the
general enthusiasm about the relevance of literary imagination for society at large. Let us first
see whether previous research provides us with enough evidence to warrant the optimism of
the theorists quoted in our introduction.

2. Empathy

2.1 Effects: The evidence

When it comes to the impact of reading stories on empathy, several studies demonstrated
positive effects on various empathic measures, like role-taking ability, motivation for pro-
social behavior and altruistic conduct (Adler/Foster 1997; Bilsky 1989;
Djikic/Oatley/Moldoveanu 2013; Johnson 2012; 2013; Mar et al. 2006; Mar et al. 2009;
Shapiro/Morrison/Boker 2004). In Bilsky’s (1989) experiment, for instance, high school
students were randomly assigned to read one of two stories, or asked to do some quiet work
on their own (control group). Both stories presented a pro-social dilemma: a character has to
decide whether to offer help to another character and bear the personal costs. After reading,
participants were separately interviewed by an experimenter asking them to consider the
consequences of the alternative solutions of the dilemma they had read about. Then the
Awareness of Consequences Scale was administered, which measures the ability to put
oneself in the position of another person. Furthermore, participants completed a Prosocial
Motivation Questionnaire. Both story groups scored significantly higher on both measures
than the control group. In the context of the present article it is interesting that this study used
literary texts as stimulus materials; however, it is not clear whether it is essential that these
stories were literary.
Bilsky (1989) showed evidence for an effect of reading narratives on the cognitive
aspects of empathy, although affective aspects may be implied in the ability to take another’s
perspective as well as in pro-social attitudes. Other studies have used items on attitude change
as a measure of cognitive empathy (or: social reasoning). In these studies, researchers
typically attempt to change white school children’s attitudes toward (minority) outgroups
(from Native Americans to children with disabilities) with the help of stories (Beardsley 1979;

4
Brisbin 1971; Geiger 1975; Jackson 1944; Heldsworth 1968; Litcher/Johnson 1969;
Marlowe/Maycock 2001; Schwartz 1972; Tauran 1967; Zucaro 1972). Only two of these
studies showed no effects (Beardsley 1979; Schwartz 1972). From the remaining reports, we
can conclude that reading stories with positive portrayals of outgroup members results in a
positive change in attitude toward that group. Again, it is not clear what literariness
contributes to such effects; the literary value of the texts that were used is unknown. Also,
children and adolescents might be easier to influence than adults.
A few studies have found an effect of narratives on cognitive empathy measures with
adults. Hakemulder (2000) found that readers of a narrative text about a woman in a
fundamentalist Islamic country opposing traditional gender roles were more inclined than
readers of an expository text on the same subject to believe that women in such countries find
it hard to accept their secondary position in society. It seems that readers are more likely to
(over)generalize from the experience of one story character than from an essay recounting the
experiences of many (cf. the “identifiable victim-effect,” e.g., Kogut/Ritov 2005).
Other recent findings suggest that reading fiction is related to empathic ability while
reading non-fiction is not (Djikic et al. 2013; Mar et al. 2006; Mar et al. 2009). Mar et al.
(2006), for instance, found that readers who were more familiar with fiction, as attested to by
their correct recognition of the names of fiction authors (the Author Recognition Test, ART),
also had higher scores on the “Reading the Mind in the Eyes Test” (Eyes Test) in which
respondents have to infer emotional states from pictures of actors’ eyes. Mar et al. (2006, 706)
used the Eyes Test as a measure of “mentalizing”, which they see as “a cognitive form of
empathy”. How this particular capacity to recognize (visual) facial expressions would be
increased by reading fiction is unclear. Perhaps this is due to one’s imaginative abilities, but
then the relation can also be reversed: people who are more able to imagine themselves in
another’s position may be more attracted to fiction. Of course it is problematic to infer (mono-
directional) causal relations from correlation studies.
Recent causal evidence for a relation between reading literary fiction and empathic
ability (note: not actual empathic behavior or attitudes) comes from Kidd and Castano (2013).
Kidd and Castano (2013) compared reactions on multiple measures of Theory of Mind after
reading literary fiction versus popular fiction, and literary fiction versus expository non-
fiction. In both comparisons, comprising 5 experiments with different texts, they found higher
scores for the literary condition on the Eyes Test and the more intricate Yoni test (Shamay-
Tsoory/Aharon-Peretz 2007). Yet, it is unclear to what kind of “literary” feature or features
this can be attributed. In an additional analysis using frequencies of various words, Kidd and
Castano (2013) found that the frequency of negative emotion words was related to higher
scores on ToM, while the effect of condition remained significant. Since their main empathic
ability test, the Yoni test, consists for a large part of understanding “gloating” and “envy”, it
could be that literary texts are better at priming these “negative” emotions. The Eyes Test, on
the other hand, uses some rather complex terms to designate emotions (e.g. “despondent”),
which could partly make it a test of verbal knowledge, maybe leading readers who just read a
more complex text to score better.
While Kidd and Castano (2013) claimed to have measured both affective and
cognitive ToM, the Eyes Test and Yoni test can be seen as measuring very basic mentalizing
skills, which can be considered predominantly cognitive, since there seems to be little

5
emotion involved. As the authors of the Eyes Test, Baron-Cohen, Wheelwright, Hill, Raste
and Plumb (2001, 241), have stressed, attributing a correct mental state is just the first stage of
Theory of Mind: “It does not include the second stage: inferring the content of that mental
state”. Nor does it directly translate to actually felt compassion with someone else. Affective
empathy for others is not often specifically measured in empirical studies, and when it is,
effects of reading narratives appear limited. Djikic et al. (2013) randomly assigned
respondents to read an essay or a literary short story and measured both cognitive and
affective empathy, using Davis’ self-report measure, the Interpersonal Reactivity Index (IRI;
Davis 1980; 1983). They did not show any change in self-reported affective empathy. Bal and
Veltkamp (2013) did find that reading a fiction text (as opposed to a non-fiction newspaper
article) caused an increase in respondents’ self-reported affective empathy (measured again by
Davis), but this effect only held for those participants who felt emotionally transported in the
story (measured by the narrative engagement scale by Busselle/Bilandzic 2009).

2.2 Processes: Explaining the relation between reading and empathy

All in all, the above-mentioned research gives tentative support for the statement that reading
narratives affects empathy positively, at least in some of its cognitive aspects. Now let us turn
to the question how it might work. To do this we need to discuss under which conditions we
see the effects, hoping that these conditions will help us make informed speculations about the
underlying psychological mechanisms and construct our framework. We will look at the
following aspects: narrativity, fictionality, literariness, absorption, and personal reader
variables.
When a comparison was made between different types of texts in the above-mentioned
studies, this was a comparison between a type of narrative text (a “story”) and an expository
text, except for the Kidd and Castano (2013) study, which also compared “literary fiction” to
“popular fiction”. That narrative texts would have stronger effects on empathy than expository
texts makes sense theoretically. According to Oatley (1994; 1999; 2002; 2008), narrative
fiction constitutes a simulation that runs on our “planning-processor,” that is, the part of our
minds we use in daily life to plan actions in order to attain goals. When reading a story,
Oatley (ibid.) argues, we make mental models of the narrative world, take on the goals and
plans of the protagonist, and subsequently experience emotions according to our evaluation to
what extent these goals are accomplished (cf. Hakemulder 2000). This type of simulation, or
“role-taking,” Oatley (1999) suggests, does not only help us to experience emotions, but also
to understand these emotions better than in everyday life.
Oatley (1999, 101) speaks of narrative “fiction,” claiming that fiction is “twice as true
as fact”. He suggests that the low-threat context of reading fiction may cause an optimal
aesthetic distance (cf. Cupchik 2001), creating a safe environment to engage with characters.
Since they are not real, we can feel for them without actually having to come to their aid.
However, while the above-mentioned studies tell us something about the effects of narrativity
on empathy, they do not provide evidence for an effect of fictionality, as they did not
exclusively pertain to fictionality. In the studies of Bal and Veltkamp (2013) and Kidd and
Castano (2013), the texts the researchers compared differed on many more variables than just
fictionality or literariness. While the Mar et al. (2006; 2009) studies explicitly claimed to
compare the effects of fiction with those of non-fiction, the difference between the authors

6
listed in the ART could also have been seen in terms of narrativity instead of fictionality.
Many of the authors of non-fiction texts will use some narrative aspects in their writing, but
never as much as any of their counterparts in the fiction. In addition, other studies comparing
the emotional effects of fiction and non-fiction found no clear evidence for a specific “fiction
effect”. Goldstein (2009), for example, found no difference between experienced levels of
sadness and anxiety in response to film clips presented as fictional or factual.
On the other hand, there is evidence from neuropsychology that texts that are
presented as fictional are processed differently than when the same texts are presented as
factual (Altman/Bohrn/Lubrich/Menninghaus/Jacobs 2012). In the “fiction” condition,
participants in Altman et al.’s study applied the “literary imagination”: they perceived the
events in the stories as possibilities of how something might have been, an imaginative
construction of hypothetical events or scenarios. These results seem to suggest that fiction
causes simulation processes, as Oatley proposed, focused on the motives of character’s
actions, that is, theory of mind. In the non-fiction condition, readers were reading to update
their world-knowledge (cf. the referential function of communication in the model of
Jakobson 1960), focusing on content (Altman et al. 2012). Thus, while non-fiction appears
unchangeable, readers of fiction seem to be involved in a process of constructive content
simulation, inclined to mind-wandering, considering what might have happened, or could
happen. The authors suggest that these simulation processes must involve perspective-taking
and the generating relational inferences, leading to co-activation in brain areas related to
theory of mind and empathy. Future research should further test whether fiction is indeed
“twice as true as fact” (Oatley 1999, 101), and examine to what extent believing a story is
based on facts has a positive (or negative) impact on empathy.
Regarding literariness, the studies discussed above provide little empirical evidence
that this is a relevant factor for empathy. The only causal empirical study that showed a bigger
impact of literary narratives than of popular narratives (Kidd/Castano 2013), was limited to
empathic ability and did not take into account which features of literary texts would be of
influence – the texts were selected on the basis of acclaim. We do have theoretical reasons to
believe literature is special. It could be argued that literary texts, with their “gaps” (Iser 1988),
polyphony (Bakhtin 1984) and often ambiguous and more complex characters, will give
readers complex psychological schemas to figure out, deviating schemas that we meet with
less frequently outside literature (cf. Hakemulder 2000; Kidd/Castano 2013; Mar/Oatley
2008). Hence, literary texts could give readers a greater challenge for their theory of mind,
and consequently a better training of the faculties involved than other narrative texts would.
Even though they did not look specifically at empathy, Kotovych, Dixon, Bortolussi and
Holden (2011) have provided some evidence that gaps and other ambiguities in literary texts
may be conducive to training one’s cognitive empathy. They showed that when readers had to
make inferences about characters due to ambiguities in the text, the characters became more
transparent to them, “closer” as it were. However, more empirical studies are needed.

[Insert Figure 1]

What we have found so far are mostly effects of narrativity. Figure 1 shows fiction
(novels and short stories not presented as “factual”) as a subset of narratives (which include,

7
besides novels and short stories: diaries, news reports, stories we tell each other in daily
conversations or on blogs, etc.), and literature as another subset of narratives that overlaps
partly with fiction. Literature, however, can also be explicitly presented as based on facts, as
in literary journalism or literary memoirs. Thus, if we find evidence that the degree to which a
text is literary increases certain responses (e.g., affect, cf. Miall/Kuiken 1994), this does not
mean that we found evidence for the effects of fiction, as fiction includes many non-literary
texts, and literature includes many non-fictional texts. If we find that reading fictional texts
causes more of a certain type of response (e.g., personal memories, cf. Seilman/Larsen 1989)
than expository texts, this again does not mean we found evidence for the effects of
fictionality: it might just as well be the lack of a narrative in the latter. A simple solution for
many of the methodological problems involved here, is using an instruction variable, leading
one group of participants to believe they will be reading a fictional text, while the other group,
reading the same text, thinks it is a “true story” (cf. Zwaan 1993, whose instruction to
respondents reading a newspaper text that they were reading a “literary” text lead to slower
reading, deeper processing, and better recall for text surface structure).
Apart from the distinction between narrativity, fictionality and literariness, other
factors need to be taken into account, as some of the above-mentioned studies bring to the
fore. First of all, the extent to which readers are transported into the narrative world seems
relevant in determining the effect of reading. In the Bal and Veltkamp (2013) study, namely,
effects of increased affective empathy only held for those who felt emotionally transported. If
readers do not engage with characters and events, they are also less likely to experience a
more general increase in empathy. The involvement with characters and events has been
given various names in the literature, including “narrative engagement”, “absorption”,
“transportation” and “narrative emotions” (e.g., Busselle/Bilandzic 2009; Green/Brock 2000;
Kneepkens/Zwaan 1994; Koopman 2011). In the present context we find it useful to make a
distinction between being drawn into the narrative world, feeling a sense of transitioning to
that world, on the one hand (“absorption” or “transportation” in the narrow sense), and having
specific character-directed emotions like empathy, sympathy and identification (“narrative
empathy,” Keen 2007) on the other.
A study by Johnson (2012), provides empirical evidence for an effect of transportation
as well as for sympathy. Johnson (ibid.) measured effects of reading a fiction story on what he
called “affective empathy” during reading (“sympathy” with the character, according to our
previous definitions), and on “pro-social behavior”. Johnson assumed that the degree of being
transported into a narrative indicates the extent to which one simulates the social experiences
depicted, and that this is the main way in which reading leads to empathic responses. Indeed,
in Johnson’s first study, participants who reported higher “transportation” also reported higher
sympathy with the character and were more likely to engage in pro-social behavior (picking
up a pen the experimenter had dropped). This effect was independent of trait empathy.
Johnson (2013) also found supporting evidence for the notion that transportation into a story
is responsible for changes in beliefs and attitudes toward out-groups. The story used in
Johnson’s (ibid.) experiments generated sympathy with Muslims and intrinsic motivation to
reduce prejudice; these variables were argued to explain how transportation leads to prejudice
reduction.

8
 While it is unclear whether transportation occurs before sympathy and empathy with
characters (or vice versa) or whether these aspects of narrative engagement occur
simultaneously, the Johnson studies (2012; 2013) do suggest that transportation and
sympathy/empathy with characters work together in influencing attitudes towards others who
are like the characters. Similarly, Hakemulder (2000) showed that readers who were
instructed to put themselves in the shoes of the character representing an outgroup to them,
were affected more strongly in their beliefs about that outgroup than the group who read the
story without such an instruction, and even more so than the control group who read a text
unrelated to the topic of the story. Hakemulder (2008) replicated this effect for other age
groups and target outgroups.
Of course, there is a reason why readers feel for characters and are transported into the
narrative world. This could partly be explained by textual characteristics (style, plot, subject
matter), but it is likely that no text ever has the exact same emotional impact on all readers
(cf. Koopman/Hilscher/Cupchik 2012). There is some empirical evidence that readers’
personal experiences (e.g., Green 2004; László/Larsen 1991) and their previous experience
with reading narrative texts (e.g., Andringa 1996; Thury/Friedlander 1995), can play a role
here. Personality factors, like trait empathy, could also be of importance. However, there has
not been much research into these factors.

2.3 Empathy: Summary

The metaphor of the moral laboratory (cf. Hakemulder 2000) comes close to a concise
summary of the research and theory on reading and empathy. Being absorbed in a narrative
can stimulate empathetic imagination. Readers go along with the author/narrator in a
(fictional) thought-experiment, imagining how it would be to be in the shoes of a particular
character, with certain motives, under certain circumstances, meeting with certain events.
That would explain why narrativity can result in a broadening of readers’ consciousness, in
particular so that it encompasses fellow human beings. Fictionality might stimulate readers to
consider the narrative they read as a thought experiment, creating distance between them and
the events, allowing them to experiment more freely with taking the position of a character
different from themselves, also in moral respects. Literary features, like gaps and ambiguous
characterization, may stimulate readers to make more mental inferences, thus training their
theory of mind.
However, apart from literature possibly being able to train basic cognitive ability, we
have little indication for the importance of literary imagination over narrative or fictional
imagination. While someone who just has read a literary text appears better at recognizing
when someone is “gloating” or “envious” than someone who just read a popular text
(Kidd/Castano 2013), this gives no guarantee of an enlarged compassion for fellow human
beings. When looking at potentially “beneficial” effects of literature, we should not limit
ourselves to empathy. Feeling empathy for characters, or even more generally after reading a
text, may not translate as readily into pro-social behavior as some of us tend to hope (cf. Keen
2007). Could it be that, indeed, all narratives (soaps, narratives in talk shows, reality
television, cf. Krijnen 2007) can function as a moral laboratory, but that some are more
effective in challenging and extending readers’ consciousness while others confirm
preexisting conceptions? Here we arrive at the importance of including reflection as a

9
potential ethical effect of literary reading. The multiple meanings in literary texts may trigger
readers to reflect on their own lives, including their views and behaviors (cf. Althusser 1980;
Bronzwaer 1986; Habermas 1983). Self-reflection might interact with empathy, as one could
relate feelings towards a character to one’s own previous experiences or future goals. The
reversed could also be true: thoughts about oneself could increase one’s understanding for
someone else. Yet, it also seems possible that reflection can occur and be ethically valuable
even without empathy, for example when one is irritated by a character, leading one to reflect
on the character’s actions and how one would act oneself.

3. Self-reflection

3.1 Effects: The evidence

We now turn to evidence that reading literature stimulates self-reflection (e.g., Culp 1985;
Levitt/Rattanasampan/Chaidaroon/Stanley/Robinson 2009; Shirley 1969; Sorensen 1999;
Waxler 2008). In a qualitative self-report study by Shirley (1969), for example, participants
provide comments that show that reading literature can stimulate moral self-evaluation. For
example, one reader of Dostoevsky’s Crime and Punishment reported: “After reading the
book I discovered how self-centered I was and how quick I was to form my opinions” (ibid.,
407).
This type of retrospective self-reports suggests that readers are at least familiar with
the ethical potential of literature. Their reports are important because they convey experiences
that seem more likely to occur outside than inside a controlled laboratory setting. However,
there are reasons to remain skeptical. Informants’ responses may have been subject to social
desirability, and their reconstructions of their past experiences are not necessarily reliable.
Moreover, it is unclear whether literariness is a precondition for these experiences.
When we focus on experimental evidence, there appear to be only a few studies that
support the notion that reading increases reflection, or related concepts, like critical thinking
(Bird 1984; Block 1993). Bird (1984) demonstrated that literature programs (a Junior Great
Books Program) can enhance scores on the Worden Critical Thinking Test. Scoring high on
this test indicates that participants actively process written and spoken information, they are
more likely to be involved in questioning, activation of background knowledge, divergent
thinking, exploring the relations among ideas, and grappling with real-life issues. It may be
that reading literary texts trains such mental activities.
However, Halász (1991) did not find more reflection in response to a literary than to
an intermediate and an expository text. In this study the researcher conceptualized reflection
as ideas related to the text itself. He did find that “remindings” (or: memories) in response to
the literary text were more personal, more affective, and more detailed (cf. Seilman/Larsen
1989). Now, memories are not necessarily related to deeper reflection; a story can make us
think about a certain personal experience, but that does not mean we reflect on it, or come to
other, meaningful insights about ourselves. But it does seem that the more room there is for
personal remindings, the more likely it is that readers will indeed reflect on their lives.
Indeed, it could very well be the case that the reflection triggered by texts higher in
literariness is qualitatively different. Multiple content analysis studies of reader responses
conducted by a research group led by Miall and Kuiken arrived at the concept of “self-

10
modifying feelings” as feelings specific to literary reading (e.g., Kuiken/Miall/Sikora 2004;
Miall/Kuiken 2002; Sikora/Kuiken/Miall 2010; 2011). “Self-modifying feelings” designate a
deeper or changed self-understanding. However, self-modifying feelings are not automatically
evoked by literary texts. As Miall and Kuiken (2002, 229) emphasize: “self-modifying
feelings are evident only among certain readers – and among them only some of the time.”
The question then becomes how and when this and similar types of reflection occur. Even
though only a minority of readers may experience it, the possibility of reflection causing
changes in the way one sees oneself and others makes it worthwhile to study the processes
that bring this reaction about.

3.2 Processes: Explaining the relation between reading and reflection

Let us now look at the conditions under which the effects of reading on reflection seem to
occur. First of all, it can be argued that literary texts are more complex to read, more complex
than the generally more straightforward expository texts (which want to inform or convince)
and popular narrative texts (which do not include as many ambiguities as literary texts; cf.
Van Peer 1986b). Beach and Hynds (1991, 461) have shown that understanding literary texts
requires well-developed problem solving strategies as well as “asking questions”. The
motivation to understand a text may stimulate readers to consider their own questions, to
bring their problem solving skills into action, and thus to train their critical thinking. The
more complex the text, the more thoughts may be provoked (although there may be a turning
point where the text becomes incomprehensible). Indeed, in an experiment, Van Peer,
Hakemulder and Zyngier (2007) found higher cognitive reflection in response to a poetic
sentence when it contained more deviating linguistic features (“foregrounding”).
Further theoretical argumentation that literary texts would trigger deeper reflection
than non-literary texts comes from Miall and Kuiken (1994; 1999). In line with Russian
Formalism, Miall and Kuiken (ibid.) propose that what makes a text “literary” is the extent of
foregrounding, and that foregrounding leads readers to become unsettled and to start looking
at things differently (“defamiliarization”). This would provide an explanation for the process
of reflection. Miall and Kuiken (2002) offer the theoretical argument that the deep type of
reflection that they call self-modifying feelings results from the following sequence: striking
features in the texts evoke narrative and aesthetic emotions; these are linked to personal
experiences (memories), and used to reflect upon oneself. Narrative emotions would include
empathy, sympathy, and identification, while aesthetic feelings would include surprise,
admiration and appreciation. The feeling of things becoming “strange,” i.e. defamiliarization,
can be seen as an aesthetic emotion, the most important one within this theory. Empirical
evidence for the connection between foregrounding, aesthetic emotions and reflection comes
from Sikora, Kuiken and Miall (2010). Sikora et al. (2010) found that emotions while reading
were related to changes in self-concept and they explained these processes through a focus on
style. Students in the Sikora et al. study who had experienced a loss, sensed the presence of
the lost one in certain striking stylistic moments of a poem, leading them to identify
metaphorically with the text and experiencing deepened self-perception.
The above suggests that self-reflection is most likely to occur when, (i) previous
personal experiences are evoked by descriptions of characters, places and events, (ii) readers
experience emotional responses to the characters, and (iii) readers perceive the text itself, the

11
artefact, as striking. But when do people experience such narrative (i and ii) and aesthetic
emotions (iii) and do these responses co-occur? Hypotheses on narrative and aesthetic
emotions during reading have been proposed by Kneepkens and Zwaan (1994). They have
argued that literary texts (i.e., texts with unconventional syntactic and semantic features) will
evoke more aesthetic emotions than non-literary texts. In addition, they suggested that
experiencing more aesthetic emotions would diminish the experience of being absorbed in the
fictional world, and thus limit narrative emotions. However, studies by Andringa (1996) and
Koopman (2011), which explicitly measured narrative and aesthetic emotions, indicate that
narrative emotions like sympathy and absorption tend to correlate with aesthetic emotions like
finding the text beautiful. These interrelations appear dependent on text and reader
characteristics: the correlation between sympathy and perceiving aesthetic attractiveness was
stronger for those with extensive reading experience in Andringa’s (1996) study, and only
these experienced readers demonstrated a correlation between sympathy and perceiving
novelty. It takes some experience with reading literature, Andringa proposes, to notice and
appreciate unconventional stylistic features.
It does seem likely that both strong aesthetic emotions and strong narrative emotions
can lead to reflection, however, empirical evidence mainly pertains to the relation between
narrative emotions and reflection. In the context of studying responses to film, Igartua (2010)
has found that a higher involvement with characters (empathy and sympathy) is related to
cognitive elaboration and a more complex reflective process (cf. Vorderer 1993). It can be
hypothesized that similar processes hold for (narrative) texts. As Levitt et al. (2009) argued,
based on in-depth interviews with six readers, identification with characters’ experiences
enables readers to reflect on threats and experiment with new possibilities and perspectives to
deal with personally difficult situations. In an experimental study by Koopman, Hilscher and
Cupchik (2012) on responses to literary rape scenes, narrative emotions (absorption,
identification, sympathy and empathy) were found to correlate positively with an intellectual
response. Likewise, and suggesting a role for aesthetic emotions as well, a study by Cupchik,
Leonard, Axelrad, and Kalin (1998) showed that literary excerpts with emotional subject
matter did not only evoke more emotions than literary excerpts with descriptive content, but
also evoked more thoughts on the text, and were found more challenging and interesting (cf.
Cupchik/László 1994). Yet, their respondents were relatively experienced readers. For less
experienced readers, narrative and aesthetic emotions may not be as strongly related, as the
Andringa (1996) study suggested, and the subsequent reflection may also be less.

3.3 Self-reflection: Summary

While there is no convincing evidence that literary texts are generally more thought-
provoking than non-literary texts (either narrative or expository), there is tentative indication
for a relation between reading literary texts and self-reflection. However, as was the case for
the studies on empathy, there is a lack of systematic comparisons between literary narratives
and non-literary narratives. There are some suggestions regarding the processes that can lead
to self-reflection. Empirical and theoretical work indicates that the combination of
experiencing narrative and aesthetic emotions tends to trigger self-reflection. Personal and
reading experience may influence narrative and aesthetic emotions.

12
4. Empirical-theoretical models
In this section, we will contrast two theoretical positions within the field of reader response
studies on underlying processes that lead to empathy and reflection. These positions have
been briefly explained before: the idea of reading literature as a form of role-taking proposed
by Oatley (e.g., 1994; 1999) and the idea of defamiliarization through deviating textual and
narrative features proposed by Miall and Kuiken (1994; 1999). We argue that these positions
are in fact complementary. While the role-taking concept seems most adequate to explain
empathic responses, the defamiliarization concept seems most adequate in explaining
reflective responses. Supplemented by further empirical studies, the discussion of these two
theoretical explanations leads to the construction of a theoretical framework (and model) that
offers useful suggestions which texts could be considered to have which effects on empathy
and reflection. (Note that the “role-taking model” and “defamiliarization model” explained
below are our simplifications and interpretations of what respectively Oatley and colleagues,
and the Miall and Kuiken research group have claimed.)

4.1 Role-taking model

As indicated before, according to Oatley (1994; 1999; 2002; 2008), during reading we make
mental models of the narrative world and protagonists’ goals and plans. Mar and Oatley
(2008) suggest that human interaction is the most central aspect to literary reading, since
“literary narratives fundamentally deal with relationships among individuals and the
navigation of conflicting desires” (ibid., 174). Making mental inferences about characters,
according to Mar and Oatley (2008), is likely to result in identification, empathy and/or
sympathy with these characters. Further theoretical explanation of this perspective is
provided, we believe, by Kotovych et al. (2011), who argue that reading is like a conversation
between narrator and reader: when we try to understand a character in a book, we make
similar inferences about what the other is thinking and feeling as in conversation, and making
such inferences would increase our understanding of and identification with the character.
Reading narratives could, in this way, function as practice for inferring emotions and
taking the perspectives of others in real life (Mar/Oatley 2008; Nussbaum 1995). Through the
process of simulating others’ experiences, readers might eventually feel more empathy for
others outside of the narrative world. Figure 2 presents a visual summary of this role-taking
model. Within this model, a reader (with at least a basic level of trait empathy) encounters a
narrative text (of at least a basic quality) depicting “the vicissitudes of human intentions”
(Bruner 1986, 13). Subsequently, perspective-taking will occur: the reader makes inferences
of the character’s mental states and may even adopt the character’s goals and plans
temporarily (cf. Aarts/Gollwitzer/Hassin 2004). This process, we assume, is partly a function
of narrative qualities (such as narrative perspective; Hakemulder/Van Peer in press). As the
practice of making mental inferences, perspective-taking may lead to an increased ability to
form an adequate Theory of Mind, although that ability is a reader variable that will already
differ before one starts to read. Role-taking may also lead to narrative emotions:
transportation into the narrative world, and sympathy and/or empathy with the character. In
the model, we have focused on the character-centered emotions, using Keen’s term “narrative
empathy”. Both the increased ability to form a theory of mind and the experience of narrative
empathy could subsequently lead to empathy towards others outside the narrative world:

13
changed attitudes towards others and possibly increased pro-social behavior. Of course, for
pro-social behavior to occur, the situational context also plays an important role (e.g., the
personal costs involved). The exact conditions under which empathetic attitudes towards
others and pro-social behavior occur remain to be investigated (cf. Loewenstein/Small 2007).
The relationship between experiencing narrative emotions and empathic attitudes and
behavior will thus not be as straightforward as this model suggests.
The central place of role-taking in this model would explain why reading narrative
texts is more likely to lead to certain empathic responses than reading non-narrative
(expository) texts: without characters to make inferences about, there are no processes of
perspective-taking, sympathy and empathy. Both life-long exposure and short-time exposure
to narratives have been associated with measures of (mainly cognitive) empathy (theory of
mind), indicating an increased ability to take in another person’s perspective. While Keen
(2007) has argued that the ability to feel for characters has no direct relation to our actions
towards actual people, feeling narrative emotions in response to stories appears to be able to
have modest pro-social effects (see Johnson 2012).

[INSERT FIGURE 2]

There remain a number of questions to be solved for future research, indicated in

Figure 2. Although it seems likely that regularly forming a theory of mind about fictional
characters leads to training of that ability, we still lack conclusive evidence for this.
Similarly, we do not know for sure that empathy toward characters will lead to empathy for
people. Also, we believe that we do not fully understand yet under which conditions role-
taking leads to narrative empathy. Nor do we know when theory of mind (cognitive empathy
as ability) leads to real-world empathy and prosocial behavior.

4.2 Defamiliarization
We believe that the role-taking model only tells part of what may be going on. If we want to
understand responses to literary narrative, it is relevant to take other stimuli factors into
account than just narrativity. An attempt to create a theory of affective response particular to
literary texts has been made by Miall and Kuiken (1994; 1999; 2002), who follow Van Peer
(1986a) in his focus on “foregrounding” and “defamiliarization”. While Oatley (1994; 1999)
suggests that empathic and reflective responses arise mostly from role-taking, from
identifying with a character, seeing things and events from his or her perspective, Miall and
Kuiken (1994; 1999) place the potential power of literature mainly in style. They suggest that
the striking features in a literary text (“foregrounding”) lead readers to become unsettled and
start looking at familiar things in a different way (“defamiliarization”). While they also see a
role for narrative emotions, and for other aesthetic emotions, we simplify their theory in a
model focusing on the aesthetic emotion defamiliarization.
In addition to the previously mentioned (scarce) support for the idea that
foregrounding triggers reflection (e.g., Hakemulder/Van Peer in press), there is some indirect
empirical support for the defamiliarization hypothesis. In a study on phonetic, semantic, and
grammatical foregrounding (Miall & Kuiken 1994), literary excerpts with a higher rate of
these types of foregrounding, provoked longer reading times and stronger affect (cf. Hunt &

14
Vipond 1985; Van Peer 1986a), which at least suggests the possibility of providing
opportunity (time) and cause (affect) for reflection. Readers also seem to aesthetically
appreciate foregrounding (e.g., Hakemulder 2004), although the moderating influence of
reading experience has to be further explored (see Hakemulder/Van Peer in press, for mixed
results).

[INSERT FIGURE 3]

The defamiliarization model as represented in Figure 3 focuses on the proposed effects of

literariness on reflection. Here it is not the narrativity but the literariness (foregrounding) of
the text that is deemed most important. Foregrounded features in the text (e.g., novel
metaphors, rhyme) can lead to aesthetic feelings of perceived beauty, but particularly surprise
and defamiliarization, the latter of which is given the central position in our simplified
visualization of the defamiliarization theory. Together with the longer reading times it evokes,
defamiliarization could subsequently lead to self-reflection. Again, we see a number of
questions emerging. We have talked about the possible relation between defamiliarization and
reflection; it is still unclear under which conditions this effect occurs. Also, how self-
reflection relates to self-awareness and self-change is uncertain.

4.3 Multifactor model of literary reading

While both of the separate models are valuable in formulating predictions about reader
responses, to get a full picture of the way literary reading could lead to empathy and
reflection, they need to be combined in one model that takes into account both narrativity and
literariness on the stimulus side, and thus narrative empathy and defamiliarization as
processes. Moreover, we propose to include a third factor, fictionality. Even though empirical
evidence that fictionality would be crucial is minimal (but not nonexistent, see Altman et al.
2012), there are important theoretical reasons to assume it could be relevant. By including
fictionality, the stimuli side of the multifactor model reflects the distinctions made in Figure
1, and speaks to all the claims made regarding the effects of literary narrative fiction.
Figure 4 includes most of the elements of Figures 2 and 3. Central in our model,
however, is a new term, “stillness”. We borrow the term stillness from the Canadian author
Yann Martel (2009), who suggests reading certain literary texts will help to stimulate self-
contemplation (and appreciation for art), moments that are especially valuable in times when
life seems to be racing by, and we are enveloped by work and a multitude of other activities.
Other literary authors propose similar ideas. Rushdie (1991) said it is a psychological need for
everyone to have, in the “house” of their minds, at least one “room” in which they can be
quiet. Kundera (1995) proposes that literature is a realm in which readers can suspend their
moral judgment, rather than giving in to the habit of judging instantly, before one even
understands what one is judging. Stillness is related to, or overlaps with the more commonly
used term “aesthetic distance,” an attitude of detachment, allowing for contemplation to take
place (cf. Cupchik 2001).
In our model, stillness is an empty space or time that is created as a result of reading
processes: the slowing down of readers’ perceptions of the fictional world, caused by
defamiliarization (compare the increased reading times in Figure 3). The model proposes both

15
narrativity and literariness influence stillness indirectly, as it would result from
defamiliarization but also from role-taking: both processes are associated with longer reading
times. As Zillmann (1994) argues, role-taking requires cognitive effort, and hence time. Miall
and Kuiken (2001) showed that reading time increases with the degree to which texts present
character perspective. It might be argued that this is due to readers’ role-taking efforts.
Fictionality may also contribute to stillness: reading a text one knows to be fictional allows
one to postpone judgments, to suspend disbelief, follow the (implied) author in his/her
representations. It is stillness, we propose, that gives readers the opportunity to reflect: to
reflect on what the events really mean to the characters, time to consider several options for
appropriate inferences (theory of mind), and time to let empathy emerge to its full extent.
Zillmann (1994) discusses evidence to suggest that it is the fast paced media that cause
emotional confusion, because a full empathic response requires some 30 seconds to emerge.
Now, extending these suggestions to the present subject, it can be argued that slowness in
perception of characters will help readers to develop their empathic responses in full. Also,
the cognitive process of forming a theory of mind will profit from slowness.
Stillness, we propose, allows a space in which slow thinking (Kahneman 2011) can
take place. It takes a pause, or break, a moment of quietness, for readers to be able to
contemplate, for instance, a new way of looking at the (over)familiar, or to consider several
narrative perspectives of different characters at once. Stillness is not reflection itself, but a
precondition for reflection.
The model shows how two seemingly competitive approaches are actually
complementary. However, instead of answering all questions, the model also reveals new
problems. As it would make the model unclear, no question marks are added in the
visualization, but most of the relations represented in this model need to be explored. First of
all, the exact order is a matter of discussion: does defamiliarization cause stillness, does
stillness allow for defamiliarization to occur, or does foregrounding cause stillness and
subsequently defamiliarization? Second, the poetic term “stillness” obviously needs to be
specified and operationalized further. In the present context it must suffice that we might find
indicators in a combination of online measures (N400 responses and longer reading times)
and offline measures (open responses suggesting readers also experience the text as allowing
for mind-wandering). This is crucial, we propose, to our understanding of the effects on
empathy and reflection, especially if we want to compare the value of literary reading to, for
instance other media and genres.
A third matter for discussion is the exact placing of the line between reading
experience and after-effects. We propose here that the mental processes of forming a theory of
mind and reflection on self are part of both the realm of reading experience and that of the
after-effects. It seems to us that these processes do not have two separate pendants. When a
text inspires readers to self-reflection, is that a reading experience or an after-effect? It might
be that the first thoughts occur during the reading (for instance during mind-wandering, when
readers’ attention is less focused on the ongoing perceptual information, and more on
internally generated thoughts and feelings), or during moments the reading is actually
interrupted. Scholes (1989) argues that such breaches in the flow of reading may be essential
in bringing life to our reading, and the other way around (see also Vlad 2009). These thoughts
may later be elaborated upon when talking to others, reading about the text, or writing in

16
diaries or book reports. As to the exact placing of the box for “theory of mind,” we think there
may be good reasons to propose that many of the processes involved in social cognition are
similar for fictional characters and human beings (cf. Culpeper 1996). On the other hand, it
may also be that there are additional attribution processes for fiction directed by literary
conventions. Finally, by reducing the narrative emotions to the character-centered “narrative
empathy” and the aesthetic emotions to “defamiliarization,” the model may not sufficiently
specify the emotional process during reading. We deemed this necessary to not make the
model overly complex.

[INSERT FIGURE 4]

Outlook
By proposing a multi-factor model of literary reading, we hoped to give an impulse to current
reader response research, which too often conflates narrativity, fictionality and literariness.
Our multi-factor model suggests that while role-taking can take place for of all types of
narratives, literary and fictional narratives may evoke the type of aesthetic distance (stillness)
that leads to a suspension of judgment, adding to a stronger experience of role-taking and
narrative empathy.
However, we need to stress that we do not expect this process of stillness to occur for
every reader and for every text. Currently, the model needs to further specify reader and
textual characteristics as factors, because these are probably of influence. As to which text
factors stimulate role-taking, a number of suggestions were made by researchers. Oatley
(1999) for instance, proposes that narrative techniques such as the use of point-of-view,
stream of consciousness, but also conversation, and presenting characters with a coherent set
of goals will lead to the type of emotional involvement that in the end leads to insight (cf.
Hakemulder/Van Peer in press, for an overview). It seems that some of these techniques are
not exclusive for literary narratives. Also, it might well be that some literary techniques are
aimed at or at least have the effect of distancing readers from the character’s perspective
rather than bringing them closer to it (defamiliarization). Future research might look into the
interplay of these two textual effects, that of puling readers into the narrative world versus
distancing them from it. It may well be key to understanding the effect of literature.
Another issue is the degree to which the “human vicissitudes” presented in the
narratives should be familiar to readers (helping them to recognize a common humanity);
what if the text presents characters (e.g., from a far-removed outgroup) with goals that seem
strange and irrelevant to the readers due to a cultural and/or moral distance between them and
the characters? The model proposes that role-taking leads to narrative empathy. An empirical
question is at what level of role-taking we will find such effects, and when does it actually
become “simulation”? Fictionality might be an important factor. Future researchers may
examine whether the “safe environment for reflection”-hypothesis can be upheld.
The style and structure of the text are also likely to impact stillness and role-taking. It
can be reasoned that readers’ ability to take someone else’s perspective is trained to a larger
extent when they are presented with multiple or otherwise complex perspectives. This would
mean that literary texts are more likely to train readers’ ability to take the perspective of
others. However, for a proportion of readers, complex perspectives (i.e. an unreliable narrator,

17
the polyphony or double voice in Free Indirect Discourse, multiple interpretable metaphors,
etc.) may interfere too much with enjoyment of the text, leading them to stop reading
altogether. Researching such interactions between text features and reader variables remains
crucial for future empirical studies (cf. McManus/Furnham 2006).
To the extent that effects on empathy and reflection are found, the question remains, of
course, how long these effects last. Some preliminary attempts have been made to establish
the duration of effects. Appel and Richter (2007) suggest that persuasion through reading
narrative fiction needs an incubation period, some time to sink in. This so-called sleeper effect
could also apply to other effects than persuasion, like empathy and reflection. In their study
on reading and empathy, Bal and Veltkamp (2013) indeed found such a sleeper effect: the
effects did not occur right after the reading but one week later. Yet, as said before, Bal and
Veltkamp did not systematically compare different genres, so it still remains to be seen under
which conditions the sleeper effect occurs. Clarifying these kinds of issues will not only help
to substantiate the claims made by scholars in the field of Literary Studies, but will also
inform our general psychological knowledge of the processes of empathy and reflection (e.g.,
of its preconditions). In the end, this might give some extra weight to arguments about the
uses of literature in, for instance, the context of moral education, development of self-
knowledge in several social contexts, such as management trainings), reducing prejudice,
sensitivity training for certain professional groups (e.g., physicians and nurses, lawyers and
judges), and bibliotherapy; reasons enough to put the problems we discussed high on the
research agenda.

Acknowledgements
The authors would like to thank Els Andringa, Susanne Janssen, Gerald Cupchik and Keith
Oatley for providing them with useful comments and further references.

References

Aarts, Henk/Peter Max Gollwitzer/Ran Hassin, Goal Contagion: Perceiving is for Pursuing,
Journal of Personality and Social Psychology 87 (2004), 23–37.
Adler, Emily/Paula Foster, A Literature-Based Approach to Teaching Values to Adolescents:
Does it Work?, Adolescence 32 (1997), 275–286.
Althusser, Louis, Drie Opstellen over Kunst en Ideologie, Nijmegen 1983.
Altmann, Ulrike/Isabel Bohrn/Oliver Lubrich/Winfried Menninghaus/Arthur Jacobs, Fact vs
Fiction: How Paratextual Information Shapes Our Reading Processes, Social Cognitive
and Affective Neuroscience 9 (2012), 22–29, doi:10.1093/scan/nss098.
Andringa, Els, Effects of ‘Narrative Distance’ on Readers’ Emotional Involvement and
Response, Poetics 23 (1996), 431–452.
Appel, Markus/Tobias Richter, Persuasive Effects of Fictional Narratives Increase over Time,
Media Psychology 10 (2007), 113–134.
Aristotle, Poetics, Cambridge, MA 1987.
Bahktin, Michail, Problems of Dostoevsky’s Poetics, Minneapolis 1984.

18
Bal, Matthijs/Martijn Veltkamp, How Does Fiction Reading Influence Empathy? An
Experimental Investigation on the Role of Emotional Transportation, Plos One
8(2013), e55341, doi: 10.1371/ journal.pone.0055341.
Baron-Cohen, Simon/Sally Wheelwright/Jacqueline Hill/Yogini Raste/Ian Plumb, The
“Reading the Mind in the Eyes” Test Revised Version: A Study With Normal Adults,
and Adults with Asperger Syndrome or High-Functioning Autism, Journal of Child
Psychology and Psychiatry 42 (2001), 241–251.
Beach, Richard/Hynds, Susan, Research on Response to Literature, in Rebecca Barr/Michael
Kamil/Peter Mosenthal/David Pearson (eds.), Handbook of reading research, New
York 1991, 453 – 489.
Beardsley, Donna, The Effects of Using Fiction in Bibliotherapy to Alter the Attitudes of
Regular Third Grade Students Toward Their Handicapped Peers, Columbia 1979.
Beran, Michael, Lincoln, Macbeth, and the Moral Imagination,
http://www.nhinet.org/beran.htm, (12.09.2014) 1998.
Bilsky, Wolfgang, Angewandte Altruismusforschung: Analyse und Rezeption von Texten über
Hilfeleistung, Bern 1989.
Bird, Jan, Effects of Fifth Graders’ Attitudes and Critical Thinking/Reading Skills Resulting
from a Junior Great Books Program, New Brunswick 1984.
Block, Cathy, Strategy Instruction in a Literature-Based Reading Program, The Elementary
School Journal 94 (1993), 139–151.
Booth, Wayne, The Company We Keep: An Ethics of Fiction, Berkeley 1988.
Bourdieu, Pierre, The Rules of Art: Genesis and Structure of the Literary Field, Cambridge
1996.
Boyd, Brian, On the Origin of Stories. Evolution, Cognition, and Fiction, London 2009.
Brisbin, Charles, An Experimental Application of the Galvanic Skin Response to the
Measurement of Effects of Literature on Attitudes of Fifth Grade Students toward
Blacks, Detroit 1971.
Bronzwaer, Wim, De Vrije Ruimte, Baarn 1986.
Bruner, Jerome, Actual Minds, Possible Worlds, Cambridge 1986
Busselle, Rick/Helena Bilandzic, Measuring Narrative Engagement, Media Psychology 12
(2009), 321–347.
Cohen, Jonathan, Defining Identification: A Theoretical Look at the Identification of
Audiences with Media Characters, Mass Communication and Society 4 (2001), 245–
264.
Cooper, Robert/Ayman Sawaf, Emotional Intelligence in Leadership, Istanbul 2003.
Coplan, Amy, Empathic Engagement With Narrative Fictions, The Journal of Aesthetics and
Art Criticism 62 (2004), 141–152.
Corse, Sarah/Saundra Westervelt, Gender and Literary Valorization: ‘‘The Awakening’’ of a
Canonical Novel, Sociological Perspectives 45 (2002), 139–161.
Culp, Mary, Case Studies of the Influence of Literature on the Attitudes, Values, and
Behavior, 1975 and 1984, English Journal 74 (1985), 31–35.
Cupchik, Gerald, The Evolution of Psychical Distance as an Aesthetic Concept, Culture &
Psychology 8 (2001), 155–187.

19
Cupchik, Gerald/Janos László, The Landscape of Time in Literary Reception: Character
Experience and Narrative Action, Cognition and Emotion 8 (1994), 297–312.
Cupchik, Gerald/Garry Leonard/Elise Axelrad/Judith Kalin, The Landscape of Emotion in
Literary Encounters, Cognition & Emotion 12 (1998), 825–847.
Culpeper, Jonathan, Inferring Character From Texts; Attribution Theory and Foregrounding
Theory, Poetics 23 (1996), 335–361.
Davis, Mark, A multidimensional approach to individual differences in empathy, JSAS
Catalogue of Selected Documents in Psychology 10 (1980), 85.
–, Measuring Individual Differences in Empathy: Evidence for a Multidimensional
Approach, Journal of Personality and Social Psychology 44 (1983), 113–126.
De Botton, Alain, How Proust Can Change Your Life, New York 1997.
De Waal, Frans, Putting the Altruism Back into Altruism: The Evolution of Empathy, Annual
Review of Psychology 59 (2007), 1–22.
Decety, Jean/Philip Jackson, A Social Neuroscience Perspective on Empathy, Current
Directions in Psychological Science 15 (2006), 54–58.
Djikic, Maja/Keith Oatley/Mihnea Moldoveanu, Reading Other Minds: Effects of Literature
on Empathy, Scientific Study of Literature 3 (2013), 28–47.
Fish, Stanley (1980), Is There a Text in This Class? The Authority of Interpretive
Communities, Cambridge, MA 1980.
Geiger, Klaus, Jugendliche lesen Landser Hefte. Hinweise auf Lektürefunktionen und –
wirkungen, in Gunter Grimm (ed.), Literatur und Leser; Theorien und Modelle zur
Rezeption literarische Werke, Stuttgart 1975, 324–266.
Goldstein, Thalia, The Pleasure of Unadulterated Sadness: Experiencing Sorrow in
Fiction, Nonfiction, and ‘in Person’, Psychology of Aesthetics, Creativity, and the Arts
3 (2009), 232–237.
Goleman, Daniel, Emotional Intelligence: Why It Can Matter More Than IQ, New York 1995.
Green, Melanie, Transportation into Narrative Worlds: The Role of Prior Knowledge and
Perceived Realism, Discourse Processes 38 (2004), 247–266.
Green, Melanie/Timothy Brock, The Role of Transportation in the Persuasiveness of Public
Narratives, Journal of Personality and Social Psychology, 79 (5), 701–721.
Guroian, Vigen, Awakening the Moral Imagination: Teaching Virtues Through Fairy Tales.
http://catholiceducation.org/articles/arts/al0134.html (10.11.2014), 1996.
Habermas, Jürgen, Moralbewusstsein und kommunikatives Handeln, Frankfurt am Main
1983.
Hakemulder, Jemeljan, The Moral Laboratory. Experiments Examining the Effects of Reading
Literature on Social Perception and Moral Self-Concept, Amsterdam 2000.
–, Foregrounding and Its Effect on Readers, Discourse Processes 38(2004), 193–218.
–, Imagining What Could Happen: Effects of Taking the Role of a Character on Social
Cognition, in Sonia Zyngier/Marisa Bortolussi/Anna Chesnokova/Jan Auracher (eds.),
Directions in Empirical Literary Studies, Amsterdam 2008, 139–160.
Hakemulder, Frank/Willie Van Peer, Empirical Stylistics, in Violeta Sotirova (ed.),
Companion to Stylistics, Oxford in press.

20
Halász, Laszlo, Emotional Effect and Reminding in Literary Processing, Poetics 20 (1991),
247–272 .
Hatfield, Elaine/John Cacioppo/Richard Rapson, Emotional Contagion, New York 1994.
Heldsworth, Janet, Vicarious Experience of Reading a Book in Changing Nursing Students'
Attitudes, Nursing Research 17 (1968), 135–139.
Hunt, Lynn, Inventing Human Rights; A History, New York 2007.
Hunt, Russell/Douglas Vipond, Crash-Testing a Transactional Model of Literary Reading.
Reader: Essays in Reader-Oriented Theory, Criticism, and Pedagogy 14 (1985), 23–
39.
Hutchison, Fred, The Moral Imagination, Politics, and Wisdom,
http://www.renewamerica.com/columns/hutchison/120726 (10.09.2014), 2004.
Ibarra, Herminia/Roxana Barbulescu, Identity as Narrative: Prevalence, Effectiveness, and
Consequences of Narrative Identity Work in Macro Work Role Transitions, Academy
of Management Review, 35 (2010), 135–154.
Igartua, Juan-Jose, Identification With Characters and Narrative Persuasion through Fictional
Feature Films, Communications—The European Journal of Communication Research
35 (2010), 347–373.
Iser, Wolfgang, The Reading Process: A Phenomenological Approach, in David Lodge (ed.),
Modern Criticism and Thought: A Reader, London 1988.
Jackson, Evalene, Effects of Reading Upon Attitudes Toward the Negro Race, The Library
Quarterly 14 (1944), 47–54.
Jakobson, Roman, Linguistics and Poetics, in Thomas Sebeok (ed.), Style in Language, New
York 1960, 350–377.
Janssen, Susanne, Reviewing as Social Practice: Institutional Constraints on Critics’ Attention
for Contemporary Fiction, Poetics 24 (1997), 275–297.
Johnson, Dan, Transportation Into a Story Increases Empathy, Prosocial Behavior, and
Perceptual Bias Toward Fearful Expressions, Personality and Individual Differences
52 (2012), 150–155.
–, Transportation into Literary Fiction Reduces Prejudice Against and Increases Empathy for
Arab-Muslims, Scientific Study of Literature, 3 (2013), 77–92.
Johnson, Mark, Moral Imagination; Implications of Cognitive Science for Ethics, Chicago
1993.
Kahneman, Daniel, Thinking, Fast and Slow, London 2011.
Keen, Susan, A Theory of Narrative Empathy, Narrative 14 (2006), 207–237.
–, Empathy and the Novel, New York 2007.
Kidd, David/Emanuele Castano, Reading Literary Fiction Improves Theory of Mind, Science,
342 (2013), 377–380.
Kimmel, Eric, Can Children's Books Change Children’s Values?, Educational Leadership 38
(1970): 209–214.
Klemenz-Belgardt, Edith, American Research on Response to Literature: The Empirical
Studies, Poetics 10 (1981), 357–380.
Kneepkens, Leonore/Rolf Zwaan, Emotions and Text Comprehension, Poetics 23 (1994),
125–138.

21
Kogut, Tehila/Ilana Ritov, The “Identified Victim” Effect: An Identified Group, Or Just a
Single Individual?, Journal of Behavioural Decision-Making 18 (2005), 157–167.
Koopman, Eva Maria (Emy), Predictors of Insight and Catharsis Among Readers Who Use
Literature as a Coping Strategy, Scientific Study of Literature 1 (2011), 241–259.
Koopman, Eva Maria (Emy)/Michelle Hilscher/Gerald Cupchik, Reader Responses to
Literary Depictions of Rape, Psychology of Aesthetics, Creativity, and the Arts 6
(2012), 66–73.
Kotovych, Maria/Peter Dixon/Marisa Bortolussi/Mark Holden, Textual Determinants of a
Component of Literary Identification, Scientific Study of Literature 1 (2011), 260–291.
Krijnen, Tonny, There is More(s) In Television, Amsterdam 2007.
Kuiken, Don/David Miall/Shelley Sikora, Forms of Self-Implication in Literary Reading,
Poetics Today, 25 (2004), 171–203.
Kundera, Milan, Testaments Betrayed, London 1995.
László, Janos/Steen Larsen, Cultural and Text Variables in Processing Personal Experiences
While Reading Literature, Empirical Studies of the Arts 9 (1991), 23–34.
Leverage, Paula/Howard Mancing/Jennifer William/Richard Schweickert (eds.), Theory of
Mind and Literature, West Lafayette 2011.
Levitt, Heidi/Woraporn Rattanasampan/Suwichit Chaidaroon/Caroline Stanley/Tamara
Robinson, The Process of Personal Change Through Reading Fictional Narratives:
Implications for Psychotherapy Practice and Theory, The Humanistic Psychologist 37
(2009), 326–352.
Litcher, John/David Johnson, Changes in Attitudes Toward Negroes of White Elementary
School Students After Use of Multiethnic Readers, Journal of Educational Psychology
60 (1969), 148–152.
Loewenstein, George/Deborah Small, The Scarecrow and the Tin Man: The Vicissitudes
of Human Sympathy and Caring, Review of General Psychology 11 (2007), 112–126.
Mar, Raymond/Keith Oatley, The Function of Fiction is the Abstraction and Simulation of
Social Experience, Perspectives on Psychological Science 3 (2008), 173–192.
Mar, Raymond/Keith Oatley/Jordan Peterson, Exploring the Link Between Reading Fiction
and Empathy: Ruling Out Individual Differences and Examining Outcomes,
Communications: The European Journal of Communication 34 (2009), 407–428.
Mar, Raymond/Keith Oatley/Maja Djikic/Justin Mullin, Emotion and Narrative Fiction:
Interactive Influences Before, During, and After Reading, Cognition & Emotion 25
(2011), 818–833.
Mar, Raymond/Keith Oatley/Jacob Hirsh/Jennifer dela Paz/Jordan Peterson, Bookworms
versus Nerds: The Social Abilities of Fiction and Non-Fiction Readers, Journal of
Research in Personality 40 (2006), 694–712.
Marlowe, Michael/George Maycock, Using Literary Texts in Teacher Education to Promote
Positive Attitudes toward Children with Disabilities, Teacher Education and Special
Education 24 (2001), 75–83.
Martel, Yann, What Is Stephen Harper Reading? Yann Martel's Recommended Reading for a
Prime Minister and Book Lovers of All Stripes, Toronto 2009.

22
McManus, Chris, & Furnham, Adrian, Aesthetic Activities and Aesthetic Attitudes: Influence
of Education, Background and Personality on Interest and Involvement in the Arts,
British Journal of Psychology 97 (2006), 555–587.
Miall, David/ Don Kuiken, Foregrounding, Defamiliarization, and Affect: Response to
Literary Stories, Poetics 22 (1994), 389–407.
–, What is Literariness? Three Components of Literary Reading, Discourse Processes 28
(1999), 121–138.
–, Shifting Perspectives: Readers’ Feelings and Literary Response, in Willie van
Peer/Seymour Chatman (eds.), New Perspectives on Narrative Perspective, Albany,
NY 2001, 289–301.
–, A Feeling for Fiction: Becoming What We Behold, Poetics 30 (2002), 221–241.
Mukařovský, Jan, On Poetic Language, John Burbank/Peter Steiner (eds), Lisse, 1976.
Nathanson, A. I., Rethinking Empathy, in J. Bryant/D. Roskos-Ewoldsen/J. Cantor (eds.),
Communication and Emotion – Essays in Honor of Dolf Zillmann, Mahwah, NJ 2003,
107–130.
Nussbaum, Martha, Poetic Justice: The Literary Imagination and Public Life, Boston 1995.
–, Cultivating Humanity: A Classical Defense of Reform in Liberal Education, Cambridge
1997.
–, Upheavals of Thought. The Intelligence of Emotions, Cambridge 2001.
–, Not for Profit. Why Democracy Needs the Humanities. Princeton, NJ 2010.
Oatley, Keith, A Taxonomy of the Emotions of Literary Response and a Theory of
Identification in Fictional Narrative, Poetics 23 (1994), 53–74.
–, Meetings of Minds: Dialogue, Sympathy, and Identification in Reading Fiction, Poetics 26
(1999), 439–454.
–, Emotions and the Story Worlds of Fiction, in Melanie Green/Jeffrey Strange/Timothy
Brock (eds.), Narrative Impact: Social and Cognitive Foundations, Mahwah, NJ 2002,
39–69
–, The Science of Fiction, The New Scientist 198 (2008), 42–43.
Pinker, Steven, The Better Angels of Our Nature. The Decline of Violence in History and Its
Causes, New York 2011.
Rushdie, Salman, Imaginary Homelands: Essays and Criticism 1981-1991, London 1991.
Scholes, Robert, Protocols of Reading, New Haven 1989.
Schwartz, Caroll, The Effect of Selected Black Poetry on Expressed Attitudes towards Blacks
of Fifth and Sixth Grade White Suburban Children, Detroit 1972.
Seilman, Uffe/Larsen, Steen, Personal Resonance to Literature: A Study of Remindings While
Reading, Poetics 18 (1989), 165–177 .
Shamay-Tsoory, Simone/Judith Aharon-Peretz, Dissociable Prefrontal Networks for
Cognitive and Affective Theory of Mind: A Lesion Study, Neuropsychologia, 45
(2007), 3054 – 3067.
Shamay-Tsoory, Simone/Judith Aharon-Peretz/Daniella Perry, Two Systems for Empathy: A
Double Dissociation between Emotional and Cognitive Empathy in Inferior Frontal
Gyrus Versus Ventromedial Prefrontal Lesions, Brain 132 (2009), 617–627.
Shapiro, Johanna/Elisabeth Morrison/John Boker, Teaching Empathy to First Year Medical
Students: Evaluation of an Elective Literature and Medicine Course, Education for
Health: Change in Learning & Practice 17 (2004), 73–84.
23
Shirley, Fehl, The Influence of Reading on Concepts, Attitudes, and Behavior, Journal of
Reading 12 (1969), 369–372, 407–413.
Shklovsky, Viktor, Art as Technique, in L. T. Lemon & M. J. Reis (eds.), Russian Formalist
Criticism: Four Essays, Lincoln, NE 1965.
Sikora, Shelley/Don Kuiken/David Miall, An Uncommon Resonance: The Influence of Loss
on Expressive Reading, Empirical Studies of the Arts 28 (2010), 135–153.
–, Expressive Reading: A Phenomenological Study of Readers’ Experience of Coleridge’s
“Rime of the Ancient Mariner,” Psychology of Aesthetics, Creativity, and the Arts 5
(2011), 258–268.
Sontag, Susan, Literature is Freedom, in Paulo Dilonardo/Anne Jump (eds.), At the Same
Time: Essays and Speeches, New York, 2007 192–209.
Sorensen, Nathalie, Making Sense for Our Lives: Women's Collaborative Reading of Fiction,
Toronto 1999.
Tan, Ed, Emotion and the Structure of Narrative Film. Film as an Emotion Machine,
Mahwah, NJ 1996.
Tauran, Rouland, The Influences of Reading on the Attitudes of Third Graders toward
Eskimos, Baltimore, MD 1967.
Thury, Eva/Alexander Friedlander, The Impact of Expertise: The Role of Experience in
Reading Literary Texts, in Gerhard Rusch (ed.), Empirical Approaches to Literature:
Proceedings of the Fourth Conference of the International Society for the Empirical
Study of Literature, Siegen 1995, 61–71.
Titchener, Edward, Experimental Psychology of the Thought Processes, London 1909.
Van Peer, Willie, Stylistics and Psychology: Investigations of Foregrounding, London 1986
(Van Peer 1986a).
–, Pulp and purpose: Stylistic Analysis as an Aid to a Theory of Texts, in Theo D'Haen (ed.),
Linguistics and the Study of Literature, Amsterdam 1986, 268–286 (Van Peer 1986b).
–, Literature, Imagination, and Human Rights, Philosophy and Literature 19 (1995), 276–291.
Van Peer, Willie/Jemeljan Hakemulder/Sonia Zyngier, Lines on Feeling: Foregrounding,
Aesthetics and Meaning, Language and Literature 16 (2007), 197–213.
Van Rees, Kees, The Institutional Foundation of a Critic’s Connoisseurship, Poetics 18
(1989), 179–198.
Vlad, Alexandra, Brüche im Hierarchischen Verarbeitungsprozess von Literatur und Film:
Eine Kognitionspsychologische Untersuchung, Erlangen-Nürnberg 2009.
Vorderer, Peter, Audience Involvement and Program Loyalty, Poetics 22 (1993), 89–98.
Waxler, Robert, Changing Lives Through Literature, PMLA 123 (2008), 678–682.
Zillmann, Dolf, Empathy: Affect From Bearing Witness to the Emotions of Others, in
Jennings Bryant/Dolf Zillmann (eds.), Responding to the Screen: Reception and
Reaction Processes, Hillsdale 1994, 135–167.
Zucaro, Blase, The Use of Bibliotherapy Among Sixth Graders to Affect Attitude Change
Toward American Negroes, Philadelphia 1972.
Zwaan, Rolf, Aspects of Literary Comprehension, Amsterdam 1993.

24
Appendix) Figures

Figure 1) The distinction between narrativity, fictionality and literariness

25
Figure 2) Role-taking model.

26
Figure 3) Defamiliarization model.

27
 Figure 4) Multifactor model of literary reading.

28

View publication stats