The Zajonc Mere Exposure Effect
In 1968, Robert Zajonc asked a deceptively simple question: does familiarity breed liking? Folk wisdom said the opposite, that familiarity breeds contempt and that the more you encounter something, the more tired of it you become. Zajonc suspected the folk wisdom had it backwards. He showed volunteers nonsense words, Chinese-like characters, and yearbook photographs of strangers, varying how many times each item appeared. Some appeared once, others 25 times, and some never. Afterward, people rated the items. The pattern was consistent: the more often something had been shown, the more favorably it was rated. Nothing about the stimuli had changed, no rewards had been attached to them, and no information had been added. They had simply been seen more often.
Zajonc called this the "mere exposure effect," and twelve years later he and a student pushed it further, showing that shapes flashed too fast to be recognized were still preferred over shapes never shown. Together these studies made a claim that was radical for its time: preferences can form without reasons, and possibly without awareness.
The inspiration: a student in a bag
The story most often attached to the effect comes from Oregon State University. In 1967, speech instructor Charles Goetzinger arranged for one student to attend his class covered head to foot in a large black bag, with only bare feet showing. The figure sat at the back of the room three mornings a week and rarely spoke. Goetzinger knew who it was; the roughly 20 other students did not.
According to accounts of the episode, the class's reaction moved from hostility toward curiosity and eventually something like friendliness. The story attracted national news coverage, and Zajonc used the "Bag Man" as a vivid illustration of his idea: people can warm to something they initially dislike, with no new information about it, simply through repeated contact.
The anecdote was not a controlled experiment, so it illustrates the hypothesis rather than proving it.
Source: Oregon State University Foundation, "What was his bag, man?" (2022), on the 1967 Goetzinger class.
Who was Robert Zajonc?
Robert Zajonc (pronounced roughly "ZYE-onts") was a Polish-born American social psychologist who spent most of his career at the University of Michigan before moving to Stanford. He survived the Second World War as a teenager in Europe, emigrated to the United States, and became known for elegant, minimalist experiments that challenged dominant theories. Before the mere exposure work he was already known for research on social facilitation, the finding that the presence of others improves performance on easy tasks and impairs it on hard ones.
The 1968 monograph: what Zajonc actually did
The paper, "Attitudinal Effects of Mere Exposure," was published as a monograph supplement to the Journal of Personality and Social Psychology. It combined two kinds of evidence.
Correlational evidence. Zajonc first showed that, in natural language, words that occur more often tend to have more positive meanings. Across antonym pairs, the positive member is typically the more frequent one. Correlations like these cannot show direction of cause, since people may simply talk more about things they like, which is why he moved to experiments.
Experimental evidence. In the laboratory studies, Zajonc controlled exposure frequency directly:
- Unfamiliar foreign-looking words. Participants saw nonsense words that meant nothing to them. Different words were shown between 0 and 25 times, with frequency counterbalanced across words so that no particular word was always the frequent one. Participants then guessed whether each word meant something good or bad. Frequently shown words were rated as having more positive meanings.
- Chinese-like characters. Using ideographs that were meaningless to the American participants, a similar procedure produced the same pattern: more exposures, more positive guesses about meaning.
- Yearbook photographs. Participants viewed photographs of unfamiliar men's faces taken from a college yearbook, again at varying frequencies, then rated how much they thought they would like each person. Faces seen more often were rated as more likable.
Two features of the design are important. First, the counterbalancing ruled out the explanation that the frequently shown items were simply nicer to begin with. Second, the stimuli were chosen to be unfamiliar and essentially meaningless, so participants could not draw on prior associations.
One correction to a popular retelling is needed here. The 1968 studies were not subliminal. The stimuli were shown long enough to be seen clearly. The famous subliminal demonstration came in 1980, discussed below. Another frequently noted detail from the 1968 data is that the increase in liking was steepest across the first few exposures and flattened out at higher frequencies.
Zajonc also ran a field version: an ad-like box containing Turkish words such as kadirga and saricik appeared on the front page of the student newspapers at the University of Michigan and Michigan State University, with each word printed a different number of times (1, 2, 5, 10 or 25). Students later received questionnaires asking whether each word seemed to mean something good or bad, and the frequently printed ones again came out ahead.
Source: Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology Monograph Supplement, 9(2, Pt. 2), 1–27.
Why this finding shocked the field
Most theories of attitude formation at the time assumed that liking needed a reason: an object was useful, beautiful, rewarding, or consistent with one's values. Zajonc's experiments stripped all of that away. The words carried no meaning to participants, no rewards were attached, and yet a systematic preference emerged from repetition alone.
Mere exposure effect — the tendency for repeated, unreinforced exposure to a stimulus to increase positive feelings toward it.
The word "unreinforced" carries a lot of weight. Classical learning theory could easily explain liking for something that had been paired with food, praise, or comfort. What made mere exposure distinctive was that the stimulus was paired with nothing. The only thing that changed was its familiarity.
In 1980, Zajonc extended the argument in a widely cited American Psychologist article titled "Feeling and Thinking: Preferences Need No Inferences." He argued that affective reactions can occur very quickly, before and independently of detailed cognitive analysis. This launched a long-running debate with Richard Lazarus, who held that emotion always depends on some form of cognitive appraisal. Much of that debate turned on how one defines "cognition," and the mere exposure data became one of Zajonc's main exhibits.
Source: Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35(2), 151–175.
The subliminal follow-up: Kunst-Wilson and Zajonc (1980)
The strongest version of Zajonc's claim needed a test in which people could not consciously know which items they had seen. William Kunst-Wilson and Zajonc provided it in a short paper in Science.
Participants viewed irregular octagons, each flashed for just 1 millisecond, with each shape presented five times. The exposures were so brief that the shapes could not be identified. Later, participants saw pairs of octagons, one old and one new, and made two kinds of judgments: which one they liked better, and which one they had seen before.
The results split cleanly. On the recognition question, participants were at chance, picking the old shape about 48% of the time. On the preference question, they chose the old shape about 60% of the time, reliably above chance. In other words, people preferred the shapes they had been exposed to while being unable to tell which ones those were. The paper's title summarized the conclusion: affective discrimination of stimuli that cannot be recognized.
This study is the reason the effect is often described as working "without awareness."
Source: Kunst-Wilson, W. R., & Zajonc, R. B. (1980). Affective discrimination of stimuli that cannot be recognized. Science, 207(4430), 557–558.
How big is the effect? Bornstein's 1989 meta-analysis
Two decades after the original paper, Robert Bornstein gathered the accumulated research into a meta-analysis covering 208 experiments published between 1968 and 1987. The overall relationship between exposure and liking was r = 0.26, a reliable effect of modest size.
The meta-analysis also identified moderators, several of which surprised researchers:
- Subliminal exposures produced larger effects than clearly visible ones. Stimuli that people did not consciously recognize generated stronger preference shifts.
- The effect was weaker for simple stimuli and stronger for more complex ones, which hold up better under repetition.
- Very long or very many exposures reduced the effect, consistent with boredom setting in.
- Effects were smaller in children than in adults, in the studies available at the time.
Bornstein's work was important not only for estimating the effect size but for showing that recognition was not required, and might even work against the effect.
Source: Bornstein, R. F. (1989). Exposure and affect: Overview and meta-analysis of research, 1968–1987. Psychological Bulletin, 106(2), 265–289.
The modern picture: Montoya et al. (2017)
A more recent meta-analysis by R. Matthew Montoya and colleagues revisited the shape of the relationship. Drawing 268 curve estimates from 81 articles, they found a positive linear component and a negative quadratic component, which together describe an inverted U. Liking rises with early exposures, levels off, and can decline with continued repetition.
This refinement matters for applied claims. "More exposure means more liking" is only true up to a point. Beyond it, people get tired of a stimulus, which is consistent with the everyday experience of a song that was charming on the tenth play and grating on the hundredth.
Source: Montoya, R. M., Horton, R. S., Vevea, J. L., Citkowicz, M., & Lauber, E. A. (2017). A re-examination of the mere exposure effect: The influence of repeated exposure on recognition, familiarity, and liking. Psychological Bulletin, 143(5), 459–498.
The field evidence: classroom faces and mirror images
Laboratory ratings of octagons are one thing. Two studies show the effect in more natural settings.
Moreland and Beach (1992). Four women of similar appearance posed as students in a large university personality psychology course. They did not interact with anyone. One attended no sessions, and the others attended 5, 10, and 15 sessions. At the end of the term, about 130 students viewed slides of the four women and rated them on traits such as attractiveness, warmth, and intelligence. Ratings of liking and similarity rose with the number of sessions attended. Repeated presence in the same room was enough.
Mita, Dermer, and Knight (1977). Researchers photographed women and printed each photo twice: once as a true image and once mirror-reversed. The women themselves tended to prefer the mirror-reversed version, which matches the face they see in a mirror every day. Their close friends tended to prefer the true image, which matches how they see her. Each group preferred the version it had seen more often.
The mirror study is a neat demonstration because the stimulus is the same face, and the only difference between the two prints is which version each viewer has been exposed to.
Sources: Moreland, R. L., & Beach, S. R. (1992). Exposure effects in the classroom: The development of affinity among students. Journal of Experimental Social Psychology, 28(3), 255–276. Mita, T. H., Dermer, M., & Knight, J. (1977). Reversed facial images and the mere-exposure hypothesis. Journal of Personality and Social Psychology, 35(8), 597–601.
Competing explanations: why does repetition increase liking?
Zajonc showed that the effect exists. Explaining it has proven harder, and several accounts compete.
Uncertainty reduction
Zajonc's own later view was rooted in evolutionary logic. A novel stimulus carries potential danger, so organisms respond to it with mild wariness. Each exposure that passes without harm signals that the stimulus is safe, and the wariness fades. On this account, liking reflects a learned absence of threat. It fits the "Bag Man" story well and is consistent with the observation that mere exposure effects appear in animals as well as humans.
Perceptual fluency and misattribution
The most influential cognitive account is the perceptual fluency/misattribution model associated with Bornstein and Paul D'Agostino (1994). Repeated exposure makes a stimulus easier to process: it is perceived more quickly and accurately. People notice this ease but do not always know where it comes from, so they misattribute it to a property of the stimulus, such as likability.
This model makes a distinctive prediction: if people know why a stimulus feels easy (because they recognize having seen it), they should discount the fluency, and the effect should shrink. Bornstein and D'Agostino's 1992 experiments reported that the effect was indeed larger for stimuli people did not recognize, which matches the meta-analytic finding that subliminal exposures produce stronger effects.
Hedonic fluency
Rolf Reber, Piotr Winkielman, and Norbert Schwarz showed in 1998 that fluency increases liking even without repetition. When they made pictures easier to process by other means, such as a matching prime or higher figure-ground contrast, participants rated the pictures as prettier. Winkielman and colleagues later argued that fluency is not merely misattributed but is itself mildly pleasant: easy processing feels good. On this view, mere exposure is one route among many to a general fluency-liking link.
Fluency — the subjective ease with which information is processed, which tends to produce more positive evaluations.
How the accounts relate
These explanations are not fully exclusive. Uncertainty reduction addresses why liking would be adaptive. Fluency accounts address the mechanism in the moment. The debate between misattribution and hedonic fluency concerns whether fluency must be mistaken for something else or is pleasant in its own right. Research has not crowned a single winner, and the best-supported position is that more than one process can contribute.
Sources: Bornstein, R. F., & D'Agostino, P. R. (1992). Stimulus recognition and the mere exposure effect. Journal of Personality and Social Psychology, 63(4), 545–552. Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual fluency on affective judgments. Psychological Science, 9(1), 45–48.
Replication status: what holds up
Social psychology's replication crisis has hit several famous findings hard. The mere exposure effect has fared comparatively well, but an honest summary needs some nuance.
What is solid. The basic finding that repeated exposure to neutral, unfamiliar stimuli modestly increases liking has been reproduced across hundreds of studies, many labs, and many kinds of stimuli, including words, shapes, faces, music, and real people in classrooms. Two large meta-analyses, nearly three decades apart, both find a reliable positive effect.
What is more fragile. The effect is modest, not dramatic, and its size depends heavily on conditions. Some specific claims are less secure than the general one. The subliminal version has produced mixed results across labs, partly because ensuring that exposures are truly below awareness is technically difficult and partly because the effect is small enough to be missed in modest samples. Meta-analytic averages also come mainly from published studies, which can inflate estimates if null results went unpublished.
What has been revised. The early impression of a steadily rising curve has given way to the inverted-U picture: diminishing returns, then decline.
A fair summary: the mere exposure effect is one of the more durable findings in social psychology, but it is a small nudge rather than a powerful force, and its subliminal form deserves more caution than its everyday form.
Criticism and limitations
Beyond replication, researchers have raised several substantive objections.
Demand characteristics. In some designs, participants might infer that the frequently shown items are "supposed" to be preferred. The subliminal studies and field studies like Moreland and Beach help address this, since participants could not easily guess the hypothesis.
Initial valence matters. The effect is strongest for neutral or mildly positive stimuli. For stimuli that are initially disliked, repetition does not reliably increase liking and can intensify negative reactions.
Context shapes the outcome. Exposure does not happen in a vacuum. If repeated encounters are paired with unpleasant experiences, such as a noisy neighbor or an intrusive advertisement, liking can fall rather than rise.
Lab stimuli are artificial. Nonsense words and octagons are chosen precisely because they carry no meaning. Real products, people, and political candidates carry many meanings, and the modest lab effect may be swamped by other information in the real world.
Common misconceptions
"Zajonc's 1968 study was subliminal." It was not. The 1968 stimuli were clearly visible. The subliminal demonstration was Kunst-Wilson and Zajonc in 1980.
"More exposure always means more liking." The effect has diminishing returns and can reverse with overexposure, according to the 2017 meta-analysis.
"The effect can make people love anything." It works best on neutral, unfamiliar stimuli and is modest in size. It nudges preferences rather than overriding them.
"The name-letter effect is just mere exposure." People do prefer letters in their own names, as Belgian psychologist Jozef Nuttin showed in 1985. But Nuttin reported that this preference persisted after accounting for how frequently letters appear in language, which is why most researchers attribute it mainly to self-association (sometimes called implicit egotism) rather than familiarity alone.
Real-world applications
Advertising and branding
The most obvious application is advertising. Brand logos, jingles, and slogans are repeated precisely because familiarity seems to make them feel more comfortable. The inverted-U finding adds a practical caution that many marketers already know as "wear-out": past a point, extra repetition annoys more than it persuades. Rotating creative variations is one common response.
Politics and name recognition
Political campaigns invest heavily in name recognition through yard signs, ads, and media appearances. Mere exposure offers one reason this could help, especially in low-information races where voters know little else about candidates. It is only one factor among many, and the lab effect's modest size suggests it matters most at the margins.
Relationships and attraction
The Moreland and Beach classroom study ties the effect to everyday social life. Proximity and repeated contact are well established predictors of friendship and attraction, and mere exposure is part of the reason. For more on how this fits with other factors, see the psychology of attraction.
How the effect fits with other biases
The mere exposure effect sits alongside other fluency-related biases, such as the illusory truth effect, where repeated statements are judged more likely to be true. Both suggest that the mind uses ease of processing as a cue for value or validity. For an overview of how these shortcuts interact, see cognitive biases explained. For another demonstration that people process far less consciously than they assume, see the invisible gorilla experiment.
The takeaway
Zajonc's work showed that preference is not always the product of conscious evaluation. Repetition alone can make something feel a little more likable, and in some conditions this happens even when people cannot recognize what they have seen. Fifty years of follow-up research has confirmed the core finding while trimming the more dramatic claims: the effect is modest, works best on neutral and unfamiliar things, and fades or reverses with overexposure. It remains a durable reminder that some of our preferences are formed by processes we do not notice.
Frequently Asked Questions
Q: Does the mere exposure effect work if you actively dislike something? A: Not reliably. The effect is strongest for neutral or unfamiliar stimuli. For things that are initially disliked, repetition may make them tolerable, but it can also intensify the dislike, especially if each encounter is unpleasant.
Q: Was Zajonc's original 1968 experiment subliminal? A: No. The 1968 studies used clearly visible nonsense words, Chinese-like characters, and yearbook photographs. The famous subliminal study, in which octagons were flashed for 1 millisecond, was published by Kunst-Wilson and Zajonc in Science in 1980.
Q: How large is the mere exposure effect? A: Modest. Bornstein's 1989 meta-analysis of 208 experiments found an average correlation of about r = 0.26 between exposure and liking. A 2017 meta-analysis found that the relationship follows an inverted U, with liking rising and then leveling off or falling with overexposure.
Q: Is the effect stronger for conscious or subliminal exposures? A: Bornstein's meta-analysis found larger effects for exposures people did not consciously recognize. The fluency/misattribution account explains this: when people know why something feels familiar, they discount that feeling. Subliminal results vary across labs, though, so this conclusion should be held with some caution.
Q: Has the mere exposure effect survived the replication crisis? A: Largely, yes. The basic effect has been reproduced across hundreds of studies and two major meta-analyses. Its size is modest, the curve is an inverted U rather than a steady rise, and the subliminal version is less consistent than the everyday version.
Q: Can advertisers use the mere exposure effect to make me like things? A: They can nudge preferences, particularly for unfamiliar brands or when products are otherwise similar. They cannot turn a bad product into a good one through repetition, and too much repetition tends to backfire through wear-out.
Q: Why does the effect sometimes reverse? A: Overexposure produces boredom and irritation. The 2017 meta-analysis supports an inverted U: early exposures increase liking, later ones add little, and heavy repetition can reduce it.
Sources
- Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology Monograph Supplement, 9(2, Pt. 2), 1–27.
- Kunst-Wilson, W. R., & Zajonc, R. B. (1980). Affective discrimination of stimuli that cannot be recognized. Science, 207(4430), 557–558.
- Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35(2), 151–175.
- Bornstein, R. F. (1989). Exposure and affect: Overview and meta-analysis of research, 1968–1987. Psychological Bulletin, 106(2), 265–289.
- Bornstein, R. F., & D'Agostino, P. R. (1992). Stimulus recognition and the mere exposure effect. Journal of Personality and Social Psychology, 63(4), 545–552.
- Bornstein, R. F., & D'Agostino, P. R. (1994). The attribution and discounting of perceptual fluency: Preliminary tests of a perceptual fluency/attributional model of the mere exposure effect. Social Cognition, 12(2), 103–128.
- Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual fluency on affective judgments. Psychological Science, 9(1), 45–48.
- Montoya, R. M., Horton, R. S., Vevea, J. L., Citkowicz, M., & Lauber, E. A. (2017). A re-examination of the mere exposure effect. Psychological Bulletin, 143(5), 459–498.
- Moreland, R. L., & Beach, S. R. (1992). Exposure effects in the classroom. Journal of Experimental Social Psychology, 28(3), 255–276.
- Mita, T. H., Dermer, M., & Knight, J. (1977). Reversed facial images and the mere-exposure hypothesis. Journal of Personality and Social Psychology, 35(8), 597–601.
- Nuttin, J. M. (1985). Narcissism beyond Gestalt and awareness: The name letter effect. European Journal of Social Psychology, 15(3), 353–361.
- Monahan, J. L., Murphy, S. T., & Zajonc, R. B. (2000). Subliminal mere exposure: Specific, general, and diffuse effects. Psychological Science, 11(6), 462–466.
- Oregon State University Foundation (2022). "What was his bag, man?"
Read next:
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

The Psychology of Attraction: What Science Says About Why We're Drawn to Each Other
The psychology of attraction explained: proximity, similarity, the mere-exposure effect, physical symmetry, Helen Fisher's three stages of love, and the science of adult attachment styles.
Complete Science Experiments Guide 2026 — Every Landmark Experiment by Field
Every landmark science experiment in one hub: Pasteur's swan-neck flask, Galvani's frog legs, Eratosthenes, Michelson–Morley, Hafele–Keating and more than 140 others, grouped by field and era.