Emotion Review http://emr.sagepub.

com/

The Mere Exposure Phenomenon: A Lingering Melody by Robert Zajonc

Richard L. Moreland and Sascha Topolinski
Emotion Review 2010 2: 329
DOI: 10.1177/1754073910375479

The online version of this article can be found at:

http://emr.sagepub.com/content/2/4/329

Published by:

http://www.sagepublications.com

On behalf of:

International Society for Research on Emotion

Additional services and information for Emotion Review can be found at:

Email Alerts: http://emr.sagepub.com/cgi/alerts

Subscriptions: http://emr.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations: http://emr.sagepub.com/content/2/4/329.refs.html

>> Version of Record - Oct 12, 2010

What is This?

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

 Emotion Review
Vol. 2, No. 4 (October 2010) 329–339
© 2010 SAGE Publications and
The International Society
The Mere Exposure Phenomenon: A Lingering for Research on Emotion
ISSN 1754-0739

Melody by Robert Zajonc DOI: 10.1177/1754073910375479

er.sagepub.com

Richard L. Moreland
Department of Psychology, University of Pittsburgh, USA

Sascha Topolinski
Department of Psychology II—Social Psychology, University of Würzburg, Germany

Abstract
The mere exposure phenomenon (repeated exposure to a stimulus is sufficient to improve attitudes toward that stimulus) is one
of the most inspiring phenomena associated with Robert Zajonc’s long and productive career in social psychology. In the first part
of this article, Richard Moreland (who was trained by Zajonc in graduate school) describes his own work on exposure and learning,
and on the relationships among familiarity, similarity, and attraction in person perception. In the second part, Sascha Topolinski
(a recent graduate who never met Zajonc, but found his ideas inspirational) describes his own work concerning embodiment and
fluency in the mere exposure effect. Also, several avenues for future research on the mere exposure phenomenon are identified,
further demonstrating its continuing relevance to the field.

Keywords
affect, cognition, mere exposure

This article is a collaboration between two people who have old films on the television, and worrying about starting graduate
never met. But we do have something in common, namely an school in the fall. I had been admitted to the University of
interest in understanding the mere exposure phenomenon. That Michigan’s Department of Psychology and assigned to work
phenomenon is just one of many identified and analyzed with Dr. Robert Zajonc. I knew that he was a famous social
by Robert Zajonc during his impressive career as a social psychologist, and I was familiar with at least some of his work,
psychologist. We have written the article in a somewhat per- on such topics as social facilitation and the mere exposure phe-
sonal way, reflecting our fascination with the mere exposure nomenon. But there was much more to learn about Bob, as I
phenomenon and our admiration for Zajonc’s work on it. “Age later discovered.
before beauty,” the old saying goes, and so our article will begin I decided that my best option was to simply meet Bob and
with comments by Richard Moreland and end with comments hear what he wanted me to do in the fall. So I made an appoint-
by Sascha Topolinski. ment with him by telephone, and when the fateful day came, I
took the bus to the Institute for Social Research (ISR) and even-
tually found Bob’s office. I sat down, introduced myself, and
1973—The Melody Begins exchanged a few pleasantries. Then I told Bob that I wanted to
In the summer of 1973, my wife (Nancy) and I moved to Ann do good work for him in the fall, and so I thought it would be
Arbor, Michigan. She began to work in Detroit at a large helpful if he described whatever plans he had made for me. Bob
international accounting firm. Nancy also began to take said there were many things I might do, but he hoped I would
evening classes at the University of Michigan, working toward become involved in some animal research that he was doing,
a graduate degree in accounting. Meanwhile, I sat around in our mostly with chickens. I was disappointed, but tried to hide it. I
new apartment, keeping odd hours, eating too much, watching had never done any animal research, and in fact, I was secretly

Corresponding author: Sascha Topolinski, Department of Psychology II—Social Psychology, University of Würzburg, Roentgenring 10, 97070 Würzburg, Germany.
Email: sascha.topolinski@psychologie.uni-wuerzburg.de.

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

330   Emotion Review Vol. 2 No. 4

proud to be one of the first people in my family to escape life offered over the years, but at the time when I began my work
on the farm. I managed to keep smiling as Bob spoke, but I did with Bob, the leading contender involved the notion of
not commit myself to anything and left as soon as possible. I “response competition.” According to that explanation, people
told Bob that I would get back in touch with him before the fall are uncertain about how to deal with a novel stimulus. What
semester began. does it mean? How can (should) it be handled? At first, multiple
That night, at the apartment, I told Nancy what had hap- and often incompatible responses seem possible. This is an
pened and asked for advice about what to do next. I am not a unhappy state of affairs. With repeated exposure to the stimu-
very assertive person, and back then I was even meeker than I lus, however, some of the possible responses are abandoned,
am now. But Nancy and I decided that I should just tell Bob that because they are too difficult, fail to produce sufficiently posi-
I preferred to do research on humans and that I was reluctant to tive outcomes, or actually produce negative outcomes. The
do any animal research. If that made Bob angry with me, even resulting decline in response competition is pleasurable, and
angry enough to reject me as a student, then Nancy and I would that pleasure becomes associated with the stimulus itself, making
just try to live with those consequences. it seem more likeable.
Luckily, our worries were unfounded. Bob was very gra- As an aside, it seems to me that this explanation for the mere
cious when I clarified my research preferences. He offered me exposure phenomenon is not so different from the perceptual
opportunities to work on several topics, all of which involved fluency explanation that was proposed years later, the explana-
human research, and after a bit of discussion we decided that I tion favored by Sasha Topolinski and his colleagues. But at the
would work on the mere exposure phenomenon. I joined sev- time I was in graduate school, we were blissfully unaware of
eral other graduate students who were already working with what perceptual fluency means and how it might produce expo-
Bob on various topics. These students included Rick Crandall sure effects. Relevant work only began to appear in the 1980s
(who later became my officemate), Bill Wilson, and Hazel (e.g., Jacoby, 1983; Mandler, Nakamura, & Van Zandt, 1987),
Markus. They proved to be a wonderful group of colleagues— and by that time, both Bob and I had moved on to pursue other
interesting, intelligent, and fun. I could write a whole article interests.
about any one of them. Direct evidence for the response competition explanation
Bob had written a famous paper (Zajonc, 1968) on the mere first came from Harrison (1968), an earlier student of Bob’s
exposure phenomenon just a few years earlier. Although several who studied the mere exposure phenomenon in his doctoral dis-
psychologists had already noted that familiar objects tend to be sertation. Other researchers later used different methods to
likeable, Bob was the first person to formalize that observation study the role of response competition in the mere exposure
as a hypothesis, one that he stated in strong terms. (I later phenomenon; their results were generally supportive as well
learned that this was characteristic of all Bob’s work—he liked (e.g., Matlin, 1970, 1971; Saegert & Jellison, 1970). Attention
to present his ideas as simply and boldly as possible, and then then turned toward the identification of variables that might
search for supportive evidence wherever he could, sometimes in moderate the mere exposure phenomenon, making it stronger or
odd places.) Bob argued that the mere repeated exposure of an weaker. Although this research was certainly valuable, it some-
object to a person (“mere exposure” in the sense that the object times produced results that were hard to explain in terms of
is just accessible to perception) is sufficient to improve some- response competition. For example, some studies (Smith &
one’s attitude toward that object. Bob offered correlational and Dorfman, 1975) revealed an inverted-U relationship between
experimental evidence to support this hypothesis. The correla- mere exposure and liking—repeated exposure led first to an
tional evidence involved the fact that words used more often (not increase, but then later to a decrease, in evaluations of stimuli.
just in English, but in other languages too) tend to refer to more And some studies showed that though stimuli that were initially
positive things. But is this because repeated exposure to words liked became even more likeable with repeated exposure,
causes their referents to be liked better, or because people speak stimuli that were initially disliked became even less likeable
more often about likeable things? To help resolve that issue, Bob (e.g., Grush, 1976).
carried out several experiments in which novel stimuli of vari- Results like these opened the door for critics to offer new
ous kinds, such as nonsense syllables, “Turkish” words, Chinese explanations of their own for the results of mere exposure
ideographs, and photographs of male faces, were presented to research. Stang (1974, 1975, 1976), for example, argued that
participants at different frequencies. The participants were not repeated exposure initially increases liking for a stimulus because
required to respond to such stimuli in any way, but simply had it allows someone to learn more about that stimulus, but once
to watch them. After the exposure portion of each experiment such learning has occurred, further exposure to the stimulus is
ended, participants’ evaluations of the stimuli were assessed. boring and thus decreases the person’s liking for the stimulus.
The results showed that stimuli viewed more often were evalu- Stang also raised the issue of demand characteristics as a factor
ated more favorably (evaluations were actually a logarithmic in the mere exposure phenomenon. He provided some evidence
function of exposure frequency). Apparently, Zajonc was that participants in the typical experiment (a) are aware that
correct—mere exposure to an object was indeed sufficient for some stimuli are being shown to them more often than others,
the development of a positive attitude toward that object. (b) believe that familiar things are generally liked better than
Why should mere exposure to an object make a person like unfamiliar things, and (c) evaluate the stimuli they have seen
it more? Several explanations for this phenomenon have been accordingly.

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

 Moreland & Topolinski   Mere Exposure   331

Stang’s arguments shaped my own early work on the mere phenomenon. In the role-playing study, few people realized that
exposure phenomenon. I was called to Bob’s office in early the effects of stimulus exposure on liking were being studied,
1973 to discuss research plans. There I experienced something and even the people who did realize that did not always believe
that would happen again many times during the years that that stimuli shown more often should be more likeable.1 And in
followed—Bob first identified a theoretical issue that he believed the experiment, the usual mere exposure phenomenon was found,
we should study, and then asked me to design and carry out a even though stimulus exposure frequencies were manipulated
study that could resolve that issue. I was given almost complete between rather than within subjects.
independence in both regards, which shocked me then (and still The next issue Bob thought was important was whether learn-
does now), because Bob knew so much more than I did about ing about a stimulus is really an important factor in the mere
research. I should say that Bob was always available to answer exposure phenomenon, as Stang (and others) had argued. But
questions about any project on which we were working, and he “learning” can mean many things. Bob reasoned that the most
was more than willing to offer advice about our projects when basic form of learning is the simple recognition of a stimulus—
I asked for help. On several occasions, he helped me to identify realizing that it is familiar. If so, then could the mere exposure
different ways of analyzing our data that eventually produced phenomenon occur if participants could not recognize the rele-
stronger results in support of our hypotheses. But Bob was also vant stimuli? Bill Wilson investigated this issue in a very direct
happy to let me just work on my own, checking in occasionally way, producing impressive findings that gained widespread
to report on any progress. attention (Wilson, 1979; see also Kunst-Wilson & Zajonc, 1980).
I’m not sure why Bob treated me this way. My favorite expla- Bill asked participants to perform a dichotic listening task. While
nation is that he trusted me to do a good job and thus believed that they listened attentively to material played in one channel of their
closer supervision was not really necessary. But in my darker earphones, participants also heard experimental stimuli (at differ-
moments, other explanations have sometimes occurred to me. It ent exposure frequencies) in the unattended channel. The partici-
may be that Bob didn’t believe the projects we worked on were pants were later unable to distinguish the experimental stimuli
very important, and so it mattered little to him whether they suc- from other, similar stimuli that they had never heard before.
ceeded or not. Or maybe Bob was more concerned with other Stimulus recognition, in other words, was minimal. And yet the
projects on which he was working, so he was willing to take some mere exposure phenomenon still occurred—stimuli that were
risks with our projects, if that freed time and energy for him. In presented more often were liked better.
any event, I like to think that I nearly always rose to the occasion My own approach to this same issue was more indirect
and produced projects that made both Bob and me proud. (Moreland & Zajonc, 1977). Two studies were designed. The
The first issue that Bob identified for me was whether first involved randomly formed “yoked” pairs of participants.
demand characteristics are an important factor in the mere Two of these pairs participated in each session of a standard mere
exposure phenomenon, as Stang had argued. Keeping that issue exposure experiment. The members of each pair were treated
in mind, I designed two related studies (Moreland & Zajonc, exactly the same during the experiment, except at the end, when
1976). The first was a role-playing study in which participants stimulus evaluations were obtained. At that time, one individual
were given descriptive materials that revealed events (in their made ratings of liking for the stimuli, while the other made rat-
usual sequence of occurrence) that someone might experience ings of familiarity. Neither person was aware that the other was
during a typical study of exposure effects. After each new set of making ratings of a different sort. In analyzing the data, I
events was described, people were asked what they thought this treated the liking and familiarity ratings from each pair as
hypothetical study was all about. I was especially interested in though they had come from a single individual. Complex statis-
whether they believed that the relationship between stimulus tical analyses (again unfamiliar to me at that time) were then
exposure and liking was being tested, and if so, then what they carried out. The results showed that when exposure frequencies
thought that relationship might be. The second study was an and familiarity ratings for the stimuli were used to predict how
experiment in which the frequency with which stimuli were well those stimuli were liked, each predictor had its own sig-
shown to participants was manipulated between rather than nificant positive effect. It thus appeared that repeated exposure
within subjects (the latter kind of manipulation is used in nearly to a stimulus could make it more likeable, regardless of how
all research on the mere exposure phenomenon). In such an familiar it seemed.
experiment, it would be difficult or impossible for participants In the second study, participants were first shown Japanese
to guess that the effects of repeated stimulus exposure were ideographs at different exposure frequencies. Afterwards, they
being studied, because all of the stimuli that any one person saw rated the familiarity of the stimuli and tried to guess their mean-
would appear the same number of times. I wondered whether ing (by reviewing a list of trait adjectives and selecting the words
the mere exposure phenomenon could still occur under these that seemed most likely to reflect each ideograph). The positivity
conditions. (A problem that I have often experienced over the of the words that were selected provided an unobtrusive measure
years arose here for the first time—complex statistical data of how well the ideographs were liked. Exposure frequencies and
analyses were needed for the data from this experiment, analyses familiarity ratings were then used to predict this measure of lik-
with which I was not yet familiar.) ing, and both predictors again had significant positive effects, just
The results from both studies showed that Stang had exag- as in the first study. Once again, repeated exposure to a stimulus
gerated the role of demand characteristics in the mere exposure made it more likeable, regardless of how familiar it seemed.

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

332   Emotion Review Vol. 2 No. 4

Taken together, these two studies suggested that the mere Bob and I received some vindication years later from a
exposure phenomenon could occur without learning (in the meta-analysis by Bornstein (1989) of research on the mere
form of stimulus recognition). For better or worse, however, the exposure phenomenon. Bornstein found that the mean effect
paper describing those studies attracted critical attention from size for the phenomenon was substantial, and he identified
two people at the University of Illinois, Michael Birnbaum and several interesting moderators of the phenomenon. One such
Barbara Mellers. They submitted a paper of their own (Birnbaum moderator was whether participants were aware of repeated
& Mellers, 1979a) to the same journal in which ours was pub- stimulus exposure or not. Bornstein found that not only could
lished, a paper in which they used the pattern of correlations the mere exposure phenomenon occur among participants who
(from each of our studies) among frequency of exposure, were unaware of such exposure, but it was actually stronger
familiarity, and liking to show that a single-factor model of the under those conditions! He went on to develop an “attribu-
results was possible. According to that model, repeated expo- tional discounting” explanation for this surprising finding.
sure created a factor (which they named “subjective recognition”), Bornstein argued that when people are aware that repeated
which then served as the basis for both familiarity and liking stimulus exposure has occurred, they feel greater liking for
ratings. The fact that this model fit the data fairly well led stimuli that were shown more often, but attribute some of that
Birnbaum and Mellers to conclude that the effects of repeated liking to the repeated exposure, and thus like the stimuli them-
stimulus exposure on liking were mediated by learning about selves somewhat less. To put this another way, people allow an
the stimuli. extrinsic factor (repeated exposure) to reduce the intrinsic
You can imagine my dismay, as a young graduate student, at value of the stimulus. But when people are not aware that
this public criticism of my research with Bob. To my surprise, repeated stimulus exposure has occurred, attributional dis-
however, Bob was not at all upset. In an effort to calm me, Bob counting of this sort does not happen, and so the (enhanced)
noted that controversy (aside from its scientific benefits) can value of the stimulus remains intact. Bornstein and others later
actually aid a person’s career (see Christensen-Szalanski & carried out experiments in which the awareness of stimulus
Beach, 1983). Bob suggested that we “fight back,” and so he sent exposure was manipulated directly (e.g., Bornstein &
me to Frank Andrews, an ISR colleague who was (among other D’Agostino, 1992). The results of this research were consistent
things) an expert on causal modeling via structural equations with the findings from the meta-analysis (where studies in
analyses. Frank was generous with his time, in part because our which participants were aware of stimulus exposure were com-
problem intrigued him—he had not yet seen any applications of pared to studies in which participants lacked such awareness).
structural equations analyses to experimental (versus correla- Bill Wilson’s research, along with my own research and that
tional) data. Frank introduced me to LISREL, a piece of software of several others who worked with Bob, eventually led Bob to a
for performing such analyses, and I spent many hours with broader conclusion, namely that affect does not always depend
LISREL, struggling to analyze my data in new ways. I ended up on cognition. Sometimes cognition can depend on affect, and
testing two causal models. We reported the results of those analy- affect and cognition can even operate independently at times.
ses in Moreland and Zajonc (1979). In our preferred model of the Bob offered and defended this conclusion in a series of famous
data, repeated exposure to a stimulus had a positive effect on two papers, most notably Zajonc (1980), but see also Zajonc (1984,
distinct (but correlated) factors, one associated with liking and one 2001), and Zajonc and Markus (1984). Although I had little to do
associated with familiarity. Both effects were significant, and the with these papers, they had an impact on my efforts to find an
overall model fit the data well. This suggested, as we had con- academic job after graduating from Michigan. Many of my invi-
cluded in our original paper, that learning was not necessary for tations for job interviews were based on the mistaken impression
exposure effects to occur. In the other model, which represented that I could contribute to the “affect versus cognition” debate.
the viewpoint of Birnbaum and Mellers, repeated exposure to a My only other research with Bob on the mere exposure phe-
stimulus had a positive impact on a single factor, one that could nomenon arose from an interest in research on exposure effects
represent learning (although it could represent other things too). in person perception. I believed that there might be a link
Familiarity and liking were both associated with that factor. between this work and Heider’s (1958) balance theory, espe-
Although this model fit the data reasonably well, it did not fit cially when it came to the relationships among familiarity,
them as well as our model did, and so we rejected it. similarity, and liking. (Bob had already done some work of his
Birnbaum and Mellers (1979b) later mounted a “counter- own on balance theory [e.g., Zajonc & Burnstein, 1965], and on
attack,” which appeared in the same issue of the journal. In that the possible role of mere exposure in the development of inter-
paper, they presented new structural equations analyses that dif- personal attraction [Saegert, Swap, & Zajonc, 1973].) According
fered from ours in several minor ways, but seemed to favor their to Heider, beliefs about people include unit relations (people
one-factor model of the data over our two-factor model. belong together or apart) and sentiment relations (people are
Whether they finally won or lost the “battle” as a result is hard liked or disliked), and we prefer such beliefs to be consistent or
to say—I invite energetic readers with time on their hands to “balanced.” Balance is achieved when people who have a posi-
read all of these papers for themselves and make up their own tive unit relationship (e.g., a married couple) have a positive
minds. I suspect that everyone was relieved when the matter sentiment relationship as well (e.g., love), or when people who
was finally allowed to rest, with no further publication by either have a negative unit relationship (e.g., a divorced couple) also
side. Speaking for myself, I was certainly happy to move on. have a negative sentiment relationship (e.g., hate). It seemed to

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

 Moreland & Topolinski   Mere Exposure   333

me that both familiarity and similarity implied unit relations, these effects were positive. Several causal models of the data
whereas liking implied a sentiment relation. So, balance theory were tested, including ones in which familiarity, likeability, and
suggested to me that as someone becomes more familiar, that similarity were treated as a single latent factor or as separate
person should not only seem more likeable, but also more factors. In the best-fitting model, where separate factors were
similar to the self. And as someone becomes more similar to assumed, repeated exposure affected likeability, which then
oneself, that person should not only seem more likeable, but affected both familiarity and similarity (which did not affect
also more familiar. one another).
I found evidence for several of these links in past research As my time at Michigan went on, I worked less often with
on person perception in social psychology. But to test my ideas Bob, due to my growing interest in small groups. I spent more
more directly, I designed two new experiments (Moreland & time talking to and working with Gene Burnstein instead, and
Zajonc, 1982). In one of these experiments, participants were I owe Gene a great deal for his thoughtful supervision and
first shown photographs of faces at different exposure frequen- collaboration. I did play a small role in Bob’s later research
cies. Afterward, they rated the people in those photographs for on the confluence model of intellection development, working
both likeability and similarity to themselves. As expected, peo- primarily with Mike Berbaum, who was another one of Bob’s
ple whose photographs appeared more often were rated as both students (Berbaum & Moreland, 1980, 1985; see also
more likeable and more similar, and there was some evidence Berbaum, Moreland, & Zajonc, 1986; Moreland, 1985;
that the effects of familiarity on perceived similarity were medi- Zajonc, Berbaum, Hamill, Moreland, & Akeju, 1980; Zajonc,
ated by changes in likeability. The same photographs were Markus, Berbaum, Bargh, & Moreland, 1991). After I gradu-
shown to the participants in the second experiment, but every ated from Michigan, my work with Bob was largely done,
photograph appeared at the same exposure frequency. though we certainly remained in contact and he was always a
Participants were given bogus information about how similar good friend and supporter. But although I no longer worked
the people in those photographs were to themselves. Afterward, with Bob, he was always with me in spirit, and will probably
the participants rated those persons for both likeability and remain with me in that way so as long as I am alive and work-
perceived familiarity. As expected, people who were viewed as ing as a social psychologist. He taught me most of what I
more similar to the self were rated as both more likeable and know about analyzing and studying social behavior. I was
more familiar. And there was strong evidence that the effects of very lucky to have Bob as a mentor.
similarity on perceived familiarity were mediated by changes in
likeability. The results of these experiments suggested that
2007—The Melody Lingers On
familiarity, similarity, and liking have strong connections with
one another, so strong that they might actually be parts of a Only a few things from my undergraduate studies in social
single, complex construct (dominated by liking), not three dif- psychology (during the early 2000s) remained vivid in my
ferent constructs. Bob said this construct reminded him of what memory later on. Among them was the mere exposure phenom-
is commonly called “affinity.” enon, something discovered by Robert Zajonc, a giant in the
Years later, when I was a faculty member myself, I studied field, several decades earlier. Yet the mere exposure phenome-
affinity again, this time with Scott Beach, a graduate student at non did not seem like something I should do research on
my university. We designed a field experiment (Moreland & myself. Instead, it was like the periodic system of elements in
Beach, 1992) in which several female confederates (all comparable chemistry: an established fact that might serve as a conceptual
in appearance) visited a large undergraduate course, pretending to tool, but could not be the target of my own research. Even my
be students in it. On selected days, one of these confederates early work on fluency (see below) did not initially involve the
entered the back of the classroom (arriving a bit late, so she mere exposure phenomenon. I was certainly aware of the phe-
could be seen by everyone who was already there), walked nomenon, but primarily it interested me because of its general
down to the front row, and took a seat. The confederate then sat relationship with fluency and affect, as described in an ingen-
quietly (apparently taking notes) until the lecture ended, when ious review paper by Rolf Reber, Norbert Schwarz, and Piotr
she left the room with everyone else, but without talking to Winkielman (2004), which had a deep impact on me.
anyone. Our goal in all this was to produce conditions of “mere Honestly, it was only by chance that I ever studied the mere
exposure.” None of the actual students in the course realized exposure phenomenon. In the summer of 2007, Fritz Strack and
that an experiment was in progress. I began to investigate the reading fluency of word triples for a
Each confederate visited the course a different number of project on intuition (e.g., Topolinski & Strack, 2009d). We asked
times. At the end of the semester, students viewed slides of the participants to chew gum while reading word triples, in order to
confederates and rated how familiar, likeable, and similar (to interrupt their feelings of reading fluency. After piloting the
themselves) each woman seemed. Structural equations analyses experimental session, which was meant to be an hour in length,
were then carried out to see whether frequency of exposure I realized that there were still five minutes left at the end of the
(number of classroom visits) had any effects on those three session. As is my habit, I looked for a small task to fill those
variables, and how the variables related to one another. The few minutes. What else could be hampered by chewing gum?
results revealed that repeated exposure had weak effects on Following a vague intuition, I connected the fluency explana-
familiarity, but strong effects on likeability and similarity. All of tion for exposure effects (see below) to reading fluency. Would

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

334   Emotion Review Vol. 2 No. 4

chewing gum interfere with the mere exposure phenomenon? In The fluency notion also helped to connect the preference
that first session, it turned out that chewing gum destroyed the effects found in mere exposure research with other effects from
phenomenon, a finding that I later replicated numerous times. I research on (social) cognition. For instance, repeated exposure
was thrilled by this finding, especially because it connected to novel propositions (e.g., “Osorno is the capital of Chile”) causes
embodiment research with the mere exposure effect. Soon I them to seem more correct, the famous truth effect (Bacon,
realized that the literature was actually full of recent studies on 1979; Begg, Anas, & Farinacci, 1992; Hasher, Goldstein, &
the mere exposure phenomenon. Far from being an old and half- Toppino, 1977; Reber & Schwarz, 1999; Topolinski & Reber,
forgotten phenomenon, it was associated with many current 2010; Unkelbach, 2007). And repeated presentation of the
theoretical issues, all hotly debated. names of ostensible actors causes them to be perceived as more
As Richard Moreland has already described, researchers famous (Jacoby & Kelley, 1987; Jacoby, Kelley, Brown &
were interested right from the start in why the mere exposure Jasechko, 1989; Jacoby & Whitehouse, 1989). The common
phenomenon occurs. Some of the main contenders, during the denominator in all of these effects was fluency.
first generation of research on the phenomenon, were response In sum, this second generation of research on repeated stimulus
competition (Harrison, 1968), arousal (Berlyne, 1966, 1971, 1974; exposure produced a powerful explanatory tool, namely process-
Crandall, 1970), and habituation/boredom (Berlyne, 1970; ing fluency. However, a deeper riddle remained to be solved—
Stang, 1974). However, none of these early explanations could what was actually the driving force behind fluency gains?
account for all of the findings in the literature (see Bornstein,
1989). Thus, the “best” explanation was far from clear, even in
Substrates of Fluency
the 1980s, more than 10 years after the phenomenon was first
described by Zajonc. The fluency explanation for exposure effects seems plausible
and straightforward. Repetition leads to increased fluency,
which drives preference. However, another question remains,
Processing Fluency and Mere Exposure
namely, the fluency of what? In processing a stimulus, multiple
Toward the end of the 1980s, a new and powerful explanation for perceptual, cognitive, and affective processes often run in paral-
the mere exposure phenomenon arose from both the cognitive lel (see Borowsky & Besner, 2006, for example). When words
and social psychological literature on recognition and familiarity are presented to participants, for instance, visual encoding,
(Jacoby & Dallas, 1981; Jacoby & Kelley, 1987; Jacoby, lexical identification, and semantic retrieval can all occur (cf.,
Woloshyn, & Kelley, 1989; Mandler, 1980; Whittlesea, 1993). Borowsky & Besner, 2006), as well as automatic pronunciation
This explanation involved processing fluency, which is the speed (cf. Stroop, 1935). Which of these processes undergoes fluency
and efficiency of processing a stimulus (Reber, Schwarz, & gains during repeated exposure? A third generation of mere
Winkielman, 2004; Reber, Wurtz, & Zimmermann, 2004). This exposure research has recently begun to address this question
became the standard explanation for the mere exposure phenom- (Topolinski, in press; Topolinski & Strack, 2009a, 2010).
enon by the end of the 1990s and inspired a second generation of While investigating the procedural substrates of fluency
mere exposure research (e.g., Bornstein & D’Agostino, 1992; gains, Fritz Strack and I have adopted the notion of “embodi-
Phaf & Rotteveel, 2005; Reber, Winkielman, & Schwarz, 1998; ment” (Barsalou, 1999, 2008; Niedenthal, 2007; Niedenthal,
for earlier analyses of how perceptual fluency might influence Barsalou, Winkielman, Krauth-Gruber, & Ric, 2005; Niedenthal,
exposure effects, see Seamon, Brody, & Kauff, 1983; Seamon, Winkielman, Mondillon, & Vermeulen, 2009; Semin & Smith,
Marsh, & Brody, 1984). 2008; Schubert & Semin, 2010; Wilson, 2002). Embodiment
The fluency explanation is based on two simple assump- means that people represent the stimuli they encounter by cov-
tions, both supported by abundant evidence. First, repeated ertly simulating the sensorimotor responses associated with
processing of a stimulus results in greater processing fluency those stimuli (e.g., Barsalou, 2008; Niedenthal et al., 2005,
(Jacoby & Dallas, 1981; Jacoby & Whitehouse, 1989). For 2009; Semin & Smith, 2008). For example, whenever I see my
instance, words are read faster after repetition than before cup on the desk, I covertly simulate grasping it (Tucker & Ellis,
(Jacoby & Dallas, 1981; Scarborough, Cortese, & Scarborough, 1998). Moreover, the meaning of the cup to me is the sum of all
1977). Second, greater fluency automatically generates more the actions I do with it, as well as their sensory and affective
positive affect (e.g., Reber, Schwarz, et al., 2004), which can be consequences (Glenberg, 1997). Thus, the meaning of the world
demonstrated through both explicit self-reports of liking and is not sealed in dry, amodal concepts, but rather is deeply
mood (Monahan, Murphy, & Zajonc, 2000; Reber et al., 1998; grounded in modal representations. How things feel, look, and
Topolinski & Strack, 2009b, 2009c, 2009d), and through more smell, and especially how we act on them, are all very important
indirect measures of subtle affective responses that occur out- (Barsalou, 2008; Glenberg & Roberston, 2000).
side of awareness (e.g., Harmon-Jones & Allen, 2001; Evidence for this notion has emerged in recent years across
Topolinski, Likowski, Weyers, & Strack, 2009; Winkielman & various disciplines and domains. Let me consider just one brief
Cacioppo, 2001). Positive affect arising from fluency-gains in example. In their ingenious work, Van den Bergh, Vrana, and
processing is assumed to be the reason why stimuli that are Eelen (1990, see also Beilock & Holt, 2007; Yang, Gallo, &
exposed more often tend to be preferred (e.g., Phaf & Rotteveel, Beilock, 2009) presented letter dyads to participants. Some of
2005; Reber, Schwarz, et al., 2004). these dyads consisted of letters that are typed with the same

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

 Moreland & Topolinski   Mere Exposure   335

finger (using the traditional 10-finger typing method). For simulations. If people encounter stimuli repeatedly, but the
instance, F and G are both typed by the left index finger, which sensorimotor simulations associated with those stimuli are
implies motor interference. Other dyads consisted of letters blocked by some means, then exposure effects should weaken.
that are typed with different fingers, such as F and J. F is typed My colleagues and I have tested this prediction in several
by the left index finger and J is typed by the right index finger, experiments involving verbal, vocal, and visual stimuli
which implies no motor interference. The task for participants, (Topolinski, in press; Topolinski & Strack, 2009a, 2010).
who were all skilled typists, was simply to indicate their liking
for the letter dyads. No actual typing (or other manual
Blocking Sensorimotor Simulations During Mere
response) was required. The researchers found that the typists
Exposure
preferred letter dyads with no motor interference over dyads
with such interference. Fritz Strack and I first chose words as stimuli because (a) their
This astonishing result was explained in terms of covert associated dominant response (pronunciation) is well-established,
motor simulation. Merely seeing the letter dyads triggered and (b) the motor system responsible for that response (the mouth)
(among these skilled typists) the overlearned response of actu- can easily be identified (e.g., Inoue, Ono, Honda, & Kurabayashid,
ally typing the letters. This covert simulation ran more fluently 2007), and blocked (e.g., Campbell, Rosen, Solis-Macias, &
for non-interfering letter dyads than for interfering letter dyads, White, 1991; Miyake, Emerson, Padilla, & Ahn, 2004).
resulting in more positive affect and preference for the former In Topolinski and Strack (2009a, Experiment 1) we
dyads. Beilock and Holt (2007) recently demonstrated that this presented participants with nonsense words (e.g.,
effect indeed depends on covert motor simulations. In a similar PANTOKRATOR), and also with Chinese characters, which
experimental paradigm, they introduced a secondary manual served as control stimuli. Participants were later asked how
motor task for some of the participants, one that was designed much they liked each stimulus. Some of the stimuli were
to block the covert typing simulations. The preference for repeatedly presented (old), and others were presented just once
non-interfering over interfering letter dyads disappeared for (new). This was a conceptual replication of classic mere expo-
participants who executed the secondary manual task. Related sure experiments by Stang (1974) and others. Crucially, we
phenomena are the findings that individuals prefer letters that also implemented two secondary motor tasks. One task was
are entailed in their own name over other letters (Nuttin, 1985); meant to block covert pronunciation of the words. For that
or that individuals prefer target persons who have the same task, we asked participants to chew gum while evaluating the
birthday as they do over other target persons (Finch & Cialdini, stimuli (oral group). Another task served as a control condi-
1989). Both these effects may also draw on motor fluency of tion, one in which the pronunciation was allowed to occur
simulating typing the letters and dates, since the letters of the naturally. For that task, we asked participants to continuously
own name and the own birthday had been typed and written squeeze a soft foam ball with their nondominant hand (manual
more frequently by an individual than other letters and dates, group). Although both the oral and the manual participants
although this hypothesis was never tested. performed a concurrent motor task, we predicted that the oral
This research illustrates the more general notion that we task would interfere with covert pronunciation of the words,
covertly simulate the sensorimotor responses that were associ- but not the characters, whereas the manual task would not inter-
ated with a particular stimulus in the past, and that the fluency fere with covert simulations of the words or the characters.
of these embodied simulations shapes our experiences and This is exactly what occurred. In the manual group, we
preferences (Beilock & Holt, 2007; Yang et al., 2009). found reliable exposure effects for both the words and the char-
Applying this notion to the mere exposure phenomenon acters—old stimuli were liked better than new stimuli. In the
leads to a straightforward hypothesis concerning the procedural oral group, however, we found exposure effects only for the
substrates of fluency. Encountering a stimulus, we simulate the characters, not for the words. It is important to note that block-
specific sensorimotor responses associated with that stimulus. ing pronunciation of the words did not just weaken the mere
If we encounter the stimulus again, then we repeat that covert exposure phenomenon, but completely destroyed it—the mean
simulation, which runs more fluently because it has been liking ratings were virtually the same for old and new stimuli.
“trained.” This fluency gain triggers positive affect, which In a further demonstration, we used two different types of
makes the stimulus preferable to others that we have encoun- stimuli, namely verbal (words) and vocal (tunes), and manipu-
tered less often. The crucial question is which response is lated their respective motor systems, namely the mouth and
dominant for a given stimulus. For words, the dominant, over- the voicebox (Topolinski & Strack, 2009a, Experiment 3).
learned response is generally pronunciation, as in the well- Words are spoken, tunes are sung. We predicted that blocking
known Stroop-effect (Stroop, 1935; for a review, see MacLeod, oral motor simulations would interfere with exposure effects
1991). We covertly pronounce each word that we encounter. for words, whereas blocking vocal motor simulations would
When we encounter a word repeatedly, this covert pronuncia- interfere with exposure effects for tunes. To test this hypoth-
tion is also repeated and thus runs more fluently (e.g., esis, we presented words and tunes (some of these were again
Scarborough et al., 1977). repeated, whereas others were not) to participants and asked
The litmus test for this embodied account of the mere exposure one group to move their tongues in a concurrent oral task (oral
phenomenon would be to simply block covert stimulus motor group), and another group to continuously hum the sound

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

336   Emotion Review Vol. 2 No. 4

(“umm hmm”) that people make when they agree with a namely the vascular theory of facial expression (Zajonc, 1985).
speaker (Gardner, 2001) in a concurrent vocal task (vocal group). This theory, aimed at facial motor efference in emotions, holds
Note that both tasks are quite specific—moving the tongue that emotional facial muscle reactions are not only epiphenomenal
engages the oral motor system and leaves the voice box unaf- expressions of emotion, but may appear before the emotional
fected, whereas humming engages the voice box and leaves experience and may even causally shape it by regulating cere-
the mouth unaffected. bral blood flow (Zajonc, 1985; see also Adelman & Zajonc,
This paradigm created a double dissociation between stimu- 1989; Zajonc, Murphy, & Inglehart, 1989). This notion of motor
lus type and motor interference. In the oral group, we found efference as a causal determinant of affective experience con-
exposure effects for words, but not for tunes. In contrast, we verges with the idea underlying an embodied notion of the mere
found exposure effects for tunes, but not for words, in the vocal exposure effect, in which motor efference also shapes affective
motor group. These results strongly suggest that the fluency responses. This conceptual homology is a fascinating example
of covert stimulus motor simulations is what drives the mere of how Robert Zajonc’s work anticipated current theorizing.
exposure phenomenon.
As I mentioned earlier, repetition-induced fluency influences Future Directions
not only preferences, but also other judgments, such as ratings
An embodied notion of exposure effects opens the door to many
of truth (Begg et al., 1992; Unkelbach, 2007), and fame
new research questions. For instance, if sensorimotor fluency deter-
(Bornstein & D’Agostino, 1992; Jacoby, Kelley, et al., 1989;
mines the feeling of familiarity, then does it also influence other
Jacoby, Woloshyn, et al., 1989). To investigate whether these
experiences that indicate a previous encounter, such as recognition
fluency effects also involve sensorimotor simulations, we repli-
(cf., Yang et al., 2009)? However, other mere exposure effects
cated the classical false-fame paradigm with an oral task
are less likely to be based on covert sensorimotor simulations.
(Topolinski & Strack, 2010). In a first demonstration, we pre-
Consider, for instance, the classical finding that pre-exposed
sented the names of Bollywood actors (e.g., AISHWARYA
Chinese ideographs, being purely visual stimuli, are more liked
RAI) to participants and asked them to judge how famous these
than novel ones (Zajonc, 1968), even when they are exposed out-
actors were. Some of the names were repeated (old), whereas
side of individuals’ awareness (Kunst-Wilson & Zajonc, 1980). If
others appeared just once (new). We also asked some of the
the fluency hypothesis still holds for this effect, the sources of flu-
participants to eat popcorn (oral), and others to knead a soft
ency gain due to repeated exposure are still to be explored.
foam ball (manual), while rating the names. We found the clas-
Although there are most recent hints that motor fluency is a causal
sical false-fame effect for the manual interference group—old
determinant in this phenomenon (Topolinski, in press), other
names were judged as more famous than new names. However,
sources such as visual processes may also play a role. If this is true,
we did not find that effect for the oral motor group.
can interference with visual processing, such as mental imagery,
In a further demonstration, we investigated whether the most
block mere exposure effects for visual stimuli? This example also
common effect of stimulus repetition, namely increased famili-
illustrates that future research is still needed to elucidate the causal
arity, also depends on sensorimotor simulations (Bornstein &
architecture of these classical findings, which again proves how
D’Agostino, 1992; Garcia-Marques, Mackie, Claypool, &
stimulating Robert Zajonc’s work still is after decades.
Garcia-Marques, 2004; Whittlesea, 1993). For this purpose, we
Aside from these new research questions, older questions are
presented participants with the names of Asian stocks (e.g.,
still being asked. For example, does the mere exposure phenom-
UNITIKA) and the brand-names of drugs (e.g., DURAGESIC).
enon require awareness that some stimuli have been presented
Again, some of the names were repeated and others were not.
more often than others? Older work by Moreland and others
This time, we asked participants to rate how familiar the names
suggests that the answer to this question is “no,” but recent work
seemed to them. Furthermore, we asked some of the participants
by Newell and his colleagues (see Newell & Bright, 2001, 2003;
to eat a cereal bar (oral interference), and other participants to
Newell & Shanks, 2007) suggests otherwise. Newell and
knead a foam ball (manual interference), while rating the
Shanks, for instance, found that stimulus recognition plays a
names. Once again, manual interference left exposure-induced
more important role in producing exposure effects than the
familiarity unaffected, while oral interference destroyed that
older work by other researchers, reviewed earlier, suggested.
effect—old names were not rated as more familiar than new
Future research will presumably continue to address this issue.
ones in the oral group. These findings show that the effects of
stimulus repetition on judgments of fame and familiarity also
Conclusion
depend to some extent on covert sensorimotor simulations.
In conclusion, many mere exposure phenomena seem to Recent developments in research on the mere exposure phe-
draw heavily on embodied simulations. In general terms, this nomenon show that it is not a dusty antique that belongs in the
embodied view on the mere exposure effect reveals that it is not corner of some museum, but rather is something vital that still
the stimulus itself that drives these effects, but stimulus- commands broad interest from the scientific community. Most
triggered bodily reactions: the dynamics of motor efference can of my own current research involves that phenomenon, in one
create preferences (see also Beilock & Holt, 2007). This in turn way or another. Unfortunately, I never met Robert Zajonc; he
connects to yet another fascinating idea that was promoted by passed away in the same week that the galley proofs for my first
Robert Zajonc independently of his work on mere exposure, paper on the phenomenon arrived in the mail. But standing on

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

 Moreland & Topolinski   Mere Exposure   337

the shoulders of this giant, I am proud to carry forward the Crandall, J. E. (1970). Preference and expectancy arousal. Journal of
research enterprise that he began. General Psychology, 83, 267–268.
Finch, J. F., & Cialdini, R. B. (1989). Another indirect tactic of (self-) image
management: Boosting. Personality and Social Psychology Bulletin,
Note 15, 222–232.
1 The role-playing study was not included in Moreland and Zajonc Garcia-Marques, T., Mackie, D. M., Claypool, H. M., & Garcia-Marques,
(1976), because the journal editor did not like it enough. However, a L. (2004). Positivity can cue familiarity. Personality and Social
brief description of the study later appeared in a book on attitudes by Psychology Bulletin, 30, 585–593.
Rajecki (1982). Gardner, R. (2001). When listeners talk. Response tokens and listener
stance. Amsterdam: Benjamins.
References Glenberg, A. M. (1997). What memory is for. Behavioral & Brain Sciences,
Adelmann, P. K., & Zajonc, R. B. (1989). Facial efference and the 20, 1–55.
experience of emotion. Annual Review of Psychology, 40, 249–280. Glenberg, A. M., & Robertson, D. A. (2000). Symbol grounding and mean-
Bacon, F. T. (1979). Credibility of repeated statements: Memory for trivia. ing: A comparison of high-dimensional and embodied theories of
Journal of Experimental Psychology: Human Learning and Memory, 5, meaning. Journal of Memory & Language, 43, 379–401.
241–252. Grush, J. E. (1976). Attitude formation and mere exposure phenomena: A
Barsalou, L. W. (1999). Perceptual symbol systems. Behavioral and Brain nonartifactual explanation of empirical findings. Journal of Personality
Sciences, 22, 577–660. and Social Psychology, 33, 281–290.
Barsalou, L. W. (2008). Grounded cognition. Annual Review of Psychology, Harmon-Jones, E., & Allen, J. B. (2001). The role of affect in the mere
59, 617–645. exposure effect: Evidence from psychophysiological and individual dif-
Begg, I. M., Anas, A., & Farinacci, S. (1992). Dissociation of processes in ferences approaches. Personality & Social Psychology Bulletin, 27,
belief: Source recollection, statement familiarity, and the illusion of 889–898.
truth. Journal of Experimental Psychology: General, 121, 446–458. Harrison, A. A. (1968). Response competition, frequency, exploratory
Beilock, S. L., & Holt, L. E. (2007). Embodied preference judgments: Can behavior, and linking. Journal of Personality and Social Psychology, 9,
likeability be driven by the motor system? Psychological Science, 18, 363–368.
51–57. Hasher, L., Goldstein, D., & Toppino, T. (1977). Frequency and the confer-
Berbaum, M. L., & Moreland, R. L. (1980). Intellectual development ence of referential validity. Journal of Verbal Learning and VerbaI
within the family: A new application of the confluence model. Behavior, 16, 107–112.
Developmental Psychology, 16, 506–515. Heider, F. (1958). The psychology of interpersonal relations. New York:
Berbaum, M. L., & Moreland, R. L. (1985). Intellectual development John Wiley & Sons.
within transracial adoptive families: Retesting the confluence model. Inoue, M. S., Ono, T., Honda, E., & Kurabayashid, T. (2007). Characteristics
Child Development, 56, 207–216. of movement of the lips, tongue and velum during a bilabial plosive—A
Berbaum, M. L., Moreland, R. L., & Zajonc, R. B. (1986). Contentions over noninvasive study using a magnetic resonance imaging movie. The
the confluence model: A reply to Price, Walsh, and Vilburg. Angle Orthodontist, 77, 612–618.
Psychological Bulletin, 100, 270–274. Jacoby, L. L. (1983). Perceptual enhancement: Persistent effects of an
Berlyne, D. E. (1966). Curiosity and exploration. Science, 153, 23–33. experience. Journal of Experimental Psychology: Learning, Memory,
Berlyne, D. E. (1970). Novelty, complexity and hedonic value. Perception and Cognition, 9, 21–38.
and Psychophysics, 8, 279–286. Jacoby, L. L., & Dallas, M. (1981). On the relationship between autobio-
Berlyne, D. E. (1971). Aesthetics and psychobiology. New York: Appleton- graphical memory and perceptual learning. Journal of Experimental
Century-Crofts. Psychology: General, 110, 306–340.
Berlyne, D. E. (1974). Novelty, complexity and interestingness. In D. E. Jacoby, L. L., & Kelley, C. M. (1987). Unconscious influences of memory
Berlyne (Ed.), Studies in the new experimental aesthetics (pp. 175–181). for a prior event. Personality and Social Psychology Bulletin, 13, 314–336.
Washington, DC: Hemisphere. Jacoby, L. L., Kelley, C. M., Brown, J., & Jasechko, J. (1989). Becoming
Birnbaum, M. H., & Mellers, B. A. (1979a). Stimulus recognition may not famous overnight: Limits on the ability to avoid unconscious influences
mediate exposure effects. Journal of Personality and Social Psychology, of the past. Journal of Personality and Social Psychology, 56, 326–338.
37, 391–394. Jacoby, L. L., & Whitehouse, K. (1989). An illusion of memory: False
Birnbaum, M. H., & Mellers, B. A. (1979b). One-mediator model of expo- recognition influenced by unconscious perception. Journal of
sure effects is still viable. Journal of Personality and Social Psychology, Experimental Psychology: General, 118, 126–135.
37, 1090–1096. Jacoby, L. L., Woloshyn, V., & Kelley, C. M. (1989). Becoming famous
Bornstein, R. F. (1989). Exposure and affect: Overview and meta-analysis without being recognized: Unconscious influences of memory produced
of research, 1968–1987. Psychological Bulletin, 106, 265–289. by dividing attention. Journal of Experimental Psychology: General,
Bornstein, R. F., & D’Agostino, P. R. (1992). Stimulus recognition and the 118, 115–125.
mere exposure effect. Journal of Personality and Social Psychology, Kunst-Wilson, W. R., & Zajonc, R. B. (1980). Affective discrimination of
63, 545–552. stimuli that cannot be recognized. Science, 207, 557–558.
Borowsky, R., & Besner, D. (2006). Parallel distributed processing and MacLeod, C. M. (1991). Half a century of research on the Stroop Effect: An
lexical-semantic effects in visual word recognition: Are a few stages integrative review. Psychological Bulletin, 109, 163–203.
necessary? Psychological Review, 113, 181–195. Mandler, G. (1980). Recognizing: The judgment of previous occurrence.
Campbell, R., Rosen, S., Solis-Macias, V., & White, T. (1991). Stress in Psychological Review, 87, 252–271.
silent reading: Effects of concurrent articulation on the detection of Mandler, G., Nakamura, Y., & Van Zandt, B. J. S. (1987). Nonspecific effects
syllabic stress patterns in written words in English speakers. Language of exposure on stimuli that cannot be recognized. Journal of Experimental
and Cognitive Processes, 6, 29–47. Psychology: Learning, Memory, and Cognition, 13, 646–648.
Christensen-Szalanski, J. J. J., & Beach, L. R. (1983). Publishing opinions: Matlin, M. W. (1970). Response competition as a mediating factor in the
A note on the usefulness of commentaries. American Psychologist, 48, frequency–affect relationship. Journal of Personality and Social
1400–1401. Psychology, 16, 536–552.

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

338   Emotion Review Vol. 2 No. 4

Matlin, M. W. (1971). Response competition, recognition, and affect. Scarborough, D. L., Cortese, C., & Scarborough, H. S. (1977). Frequency
Journal of Personality and Social Psychology, 19, 293–300. and repetition effects in lexical memory. Journal of Experimental
Miyake, A., Emerson, M. J., Padilla, F., & Ahn, J. (2004). Inner speech as a Psychology: Human Perception and Performance, 3, 1–17.
retrieval aid for task goals: The effects of cue type and articulatory sup- Schubert, T. W., & Semin, G. R. (2010). Embodiment as a unifying pers­
pression in the random task cuing paradigm. Acta Psychologica, 115, pective for psychology. European Journal of Social Psychology, 39,
123–142. 1135–1141.
Monahan, J. L., Murphy, S. T., & Zajonc, R. B. (2000). Subliminal mere Seamon, J. G., Brody, N., & Kauff, D. M. (1983). Affective discrimination
exposure: Specific, general and diffuse effects. Psychological Science, of stimuli that are not recognized: Effects of shadowing, masking, and
11, 462–467. cerebral laterality. Journal of Experimental Psychology: Learning,
Moreland, R. L. (1985). Birth order. In A. Kuper & J. Kuper (Eds.), The Memory, and Cognition, 9, 544–555.
social science encyclopedia. London: Routledge & Kegan Paul. Seamon, J. G., Marsh, R. L., & Brody, N. (1984). Critical importance of
Moreland, R. L., & Beach, S. R. (1992). Exposure effects in the classroom: exposure duration for affective discrimination of stimuli that are not recog-
The development of affinity among students. Journal of Experimental nized. Journal of Experimental Psychology: Learning, Memory, and
Social Psychology, 28, 255–276. Cognition, 10, 465–469.
Moreland, R. L., & Zajonc, R. B. (1976). A strong test of exposure effects. Semin, G. R., & Smith, E. R. (2008). Introducing embodied grounding. In
Journal of Experimental Social Psychology, 12, 170–179. G. R. Semin & E. R. Smith (Eds.), Embodied grounding: Social, cogni-
Moreland, R. L., & Zajonc, R. B. (1977). Is stimulus recognition a neces- tive, affective, and neuroscientific approaches (pp. 1–8). New York:
sary condition for the occurrence of exposure effects? Journal of Cambridge University Press.
Personality and Social Psychology, 35, 191–199. Smith, G. F., & Dorfman, D. D. (1975). The effect of stimulus uncertainty
Moreland, R. L., & Zajonc, R. B. (1979). Exposure effects may not depend on the relationship between frequency of exposure and liking. Journal
on stimulus recognition. Journal of Personality and Social Psychology, of Personality and Social Psychology, 31, 553–558.
37, 1085–1089. Stang, D. J. (1974). Methodological factors in mere exposure research.
Moreland, R. L., & Zajonc, R. B. (1982). Exposure effects in person percep- Psychological Bulletin, 81, 1014–1025.
tion: Familiarity, similarity, and attraction. Journal of Experimental Stang, D. J. (1975). Effects of mere exposure on learning and affect. Journal
Social Psychology, 18, 395–415. of Personality and Social Psychology, 31, 7–12.
Newell, B. R., & Bright, J. E. H. (2001). The relationship between the struc- Stang, D. J. (1976). A critical examination of the response competition
tural mere exposure effect and the implicit learning process. Quarterly hypothesis. Bulletin of the Psychonomic Society, 7, 530–532.
Journal of Experimental Psychology, 54, 1087–1104. Stroop, J. R. (1935). Studies of interference in serial verbal reactions.
Newell, B. R., & Bright, J. E. H. (2003). The subliminal mere exposure Journal of Experimental Psychology, 12, 242–248.
effect does not generalize to structurally related stimuli. Canadian Topolinski, S. (in press). Moving the eye of the beholder: Motor compo-
Journal of Experimental Psychology, 57, 61–68. nents in vision determine aesthetic preference. Psychological Science.
Newell, B. R., & Shanks, D. R. (2007). Recognizing what you like: Topolinski, S., Likowski, K. U., Weyers, P., & Strack, F. (2009). The face of
Examining the relation between the mere-exposure effect and recogni- fluency: Semantic coherence automatically elicits a specific pattern of
tion. European Journal of Cognitive Psychology, 19, 103–118. facial muscle reactions. Cognition & Emotion, 23, 260–271.
Niedenthal, P. M. (2007). Embodying emotion. Science, 316, 1002–1005. Topolinski, S., & Reber, R. (2010). Immediate truth—Temporal contiguity
Niedenthal, P. M., Barsalou, L. W., Winkielman, P., Krauth-Gruber, S., & between a cognitive problem and its solution determines experienced
Ric, F. (2005). Embodiment in attitudes, social perception, and emotion. veracity of the solution. Cognition, 114, 117–122.
Personality and Social Psychological Review, 9, 184–211. Topolinski, S., & Strack, F. (2009a). Motormouth: Mere exposure depends
Niedenthal, P. M., Winkielman, P., Mondillon, L., & Vermeulen, N. (2009). on stimulus-specific motor simulations. Journal of Experimental
Embodiment of emotional concepts: Evidence from EMG measures. Psychology: Learning, Memory, and Cognition, 35, 423–433.
Journal of Personality and Social Psychology, 96, 1120–1136. Topolinski, S., & Strack, F. (2009b). Scanning the “fringe” of conscious-
Nuttin, J. M. (1985). Narcissism beyond Gestalt and awareness: The name ness: What is felt and what is not felt in intuitions about semantic
letter effect. European Journal of Social Psychology, 15, 353–361. coherence. Consciousness and Cognition, 18, 608–618.
Phaf, R. H., & Roteveel, M. (2005). Affective modulation of recognition Topolinski, S., & Strack, F. (2009c). The analysis of intuition: Processing
bias. Emotion, 5, 309–318. fluency and affect in judgements of semantic coherence. Cognition &
Rajecki, D. W. (1982). Attitudes: Themes and advances. Sunderland, MA: Emotion, 23, 1465–1503.
Sinauer. Topolinski, S., & Strack, F. (2009d). The architecture of intuition: Fluency
Reber, R., & Schwarz, N. (1999). Effects of perceptual fluency on judg- and affect determine intuitive judgments of semantic and visual coher-
ments of truth. Consciousness and Cognition, 8, 338–342. ence, and of grammaticality in artificial grammar learning. Journal of
Reber, R., Schwarz, N., & Winkielman, P. (2004). Processing fluency and Experimental Psychology: General, 138, 39–63.
aesthetic pleasure: Is beauty in the perceiver’s processing experience? Topolinski, S., & Strack, F. (2010). False fame prevented—avoiding fluency-
Personality and Social Psychology Review, 8, 364–382. effects without judgmental correction. Journal of Personality and Social
Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual Psychology, 98(5), 721–733.
fluency on affective judgments. Psychological Science, 9, 45–48. Tucker, M., & Ellis, R. (1998). On the relations between seen objects and
Reber, R., Wurtz, P., & Zimmermann, T. D. (2004). Exploring “fringe” components of potential actions. Journal of Experimental Psychology:
consciousness: The subjective experience of perceptual fluency and its Human perception and performance, 24, 830–846.
objective bases. Consciousness and Cognition, 13, 47–60. Unkelbach, C. (2007). Reversing the truth effect: Learning the inter­
Saegert, S. C., & Jellison, J. M. (1970). Effects of initial level of pretation of processing fluency in judgments of truth. Journal of
response competition and frequency of exposure on liking and Experimental Psychology: Learning, Memory and Cognition, 33,
exploratory behavior. Journal of Personality and Social Psychology, 219–230.
16, 553–558. Van den Bergh, O., Vrana, S., & Eelen, P. (1990). Letters from the heart:
Saegert, S., Swap, W. C., & Zajonc, R. B. (1973). Exposure, context, and Affective categorization of letter combinations in typists and nontypists.
inter­personal attraction. Journal of Personality and Social Psychology, Journal of Experimental Psychology: Learning, Memory, and Cognition,
25, 234–242. 16, 1153–1161.

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012

 Moreland & Topolinski   Mere Exposure   339

Whittlesea, B. W. A. (1993). Illusions of familiarity. Journal of Experimental Zajonc, R. B. (1985). Emotion and facial efference: A theory reclaimed.
Psychology: Learning, Memory, and Cognition, 19, 1235–1253. Science, 228, 15–21.
Wilson, M. (2002). Six views of embodied cognition. Psychonomic Bulletin Zajonc, R. B. (2001). Mere exposure: A gateway to the subliminal.
and Review, 9, 625–636. Psychological Science, 10, 224–228.
Wilson, W. R. (1979). Feeling more than we can know: Exposure effects Zajonc, R. B., Berbaum, M. L., Hamill, R., Moreland, R. L., & Akeju, S. A.
without learning. Journal of Personality and Social Psychology, 37, (1980). Family factors and the intellectual performance of Nigerian
811–821. eleven‑year‑olds. West African Journal of Educational and Vocational
Winkielman, P., & Cacioppo, J. T. (2001). Mind at ease puts a smile on the Measurement, 5, 19–26.
face: Psychophysiological evidence that processing facilitation leads to Zajonc, R. B., & Burnstein, E. (1965). The learning of balanced and
positive affect. Journal of Personality and Social Psychology, 81, unbalanced social structures. Journal of Personality, 33, 153–163.
989–1000. Zajonc, R. B., Markus, G. B., Berbaum, M. L., Bargh, J., & Moreland, R. L.
Yang, S., Gallo, D., & Beilock, S. L. (2009). Embodied memory judgments: (1991). One justified criticism plus three flawed analyses equals two
A case of motor fluency. Journal of Experimental Psychology: Learning, unwarranted conclusions: A reply to Retherford and Sewell. American
Memory, & Cognition, 35, 1359–1365. Sociological Review, 56, 159–165.
Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Zajonc, R. B., & Markus, H. (1984). Affect and cognition: The hard interface.
Personality and Social Psychology Monographs, 9, 1–27. In C. E. Izard, J. Kagan & R. B. Zajonc (Eds.), Emotions, cognition, and
Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. behavior (pp. 73–102). New York: Cambridge University Press.
American Psychologist, 35, 151–175. Zajonc, R. B., Murphy, S. T., & Inglehart, M. (1989). Feeling and facial
Zajonc, R. B. (1984). On the primacy of affect. American Psychologist, 39, efference: Implications of the vascular theory of emotion. Psychological
117–123. Review, 96, 395–416.

Downloaded from emr.sagepub.com at PONTIFICIA UNIV JAVERIANA on January 31, 2012