The Science
Liking is not earned; it is accumulated. Every time a human eye lands on a stimulus, an invisible ledger updates — not a ledger of facts, but of feeling. Robert Zajonc proved it in 1968 in his landmark monograph Attitudinal Effects of Mere Exposure (Journal of Personality and Social Psychology), showing across Chinese ideographs, nonsense words, and human faces that the more often people simply saw a thing — with no reward, no argument, no persuasion — the more they liked it. He called it "mere" exposure because the exposure was all that was required. The affection arrived for free.
The effect is arguably the most robust finding in social psychology. Bornstein's (1989) foundational meta-analysis in Psychological Bulletin synthesized 208 experiments spanning 1968–1987 and found the effect reliable and substantial, with a pooled correlation around r = 0.26 — and far larger, up to r = 0.63, in Zajonc's own word-frequency data. Montoya, Horton, Vevea, Citkowicz, and Lauber's (2017) re-examination in Psychological Bulletin (268 curve estimates across 81 articles) confirmed the effect is real, directional, and — critically — strongest for visual stimuli. A visual collection built from the most-seen images in human memory sits squarely in the literature's strongest condition.
The mechanism beneath the effect makes it stranger still. Reber, Winkielman, and Schwarz (1998, Psychological Science; Reber, Schwarz, & Winkielman, 2004, Personality and Social Psychology Review) established that mere exposure works through processing fluency: repeatedly seen stimuli become easier for the brain to process, and that ease is itself experienced as a small jolt of pleasure — the "hedonic marking" hypothesis (Winkielman et al., 2003). When a viewer's eye hits a moon-landing image, the visual system processes it almost effortlessly because it has done so thousands of times before. The effortlessness feels good, and the brain — unable to locate the true source of the feeling — attributes it to the object. The viewer concludes, I like this, never suspecting that the liking was manufactured by a lifetime of prior exposure.
Repetition has a second-order effect as well: it breeds believed importance. Hasher, Goldstein, and Toppino (1977, Journal of Verbal Learning and Verbal Behavior) first documented the illusory-truth effect, and Dechêne, Stahl, Hansen, and Wänke's (2010, Personality and Social Psychology Review) meta-analysis fixed it at d = 0.53 — repeated statements are not just better liked but more readily believed. Exposure is also a measured quantity where digital collectibles trade. Kapoor, Guhathakurta, Mathur, Yadav, Gupta, and Kumaraguru (2022), in TweetBoost (Companion Proceedings of The Web Conference 2022), built a dataset of 245,159 tweets from 17,155 users linked to 62,997 OpenSea assets and showed that adding social-exposure features improved NFT valuation prediction by 6% over on-chain-only baselines — with raw exposure signals (likes, retweets, list membership) among the most predictive variables. Related research finds roughly 70% of marketplace social traffic originates on Twitter/X. In this market, exposure is not a soft metric; it is a priced one.
There is a known limit — and it is a revealing one. Bornstein (1989) and Montoya et al. (2017) document that the exposure-liking curve eventually bends downward under tedium, but only for simple, meaningless stimuli, and typically only after roughly 35 monotonous repetitions. Meaningful, complex, emotionally loaded stimuli sustain the effect far longer (Berlyne's two-factor model, 1970). Historical monuments are the most meaning-saturated stimuli imaginable; they live on the rising side of the curve, where each additional exposure still adds affection.
The same engine has run at commercial scale before. Coca-Cola spent the twentieth century proving Zajonc's thesis in the largest field experiment in commercial history: relentless visual repetition of a single script logo until the entire planet felt affection for tinted sugar water — accumulated exposure converted into preference, exactly the fluency-to-liking pathway later formalized by Reber et al. (2004). CryptoPunks won on the same mechanism: crude 24×24-pixel faces that succeeded not on aesthetics but on being seen first and seen most, their relentless circulation across crypto-Twitter generating a fluency that the audience read as intrinsic appeal. And pop radio is the purest applied test of Berlyne's inverted-U — songs nobody requested become beloved through sheer rotation, an industry built on managing the exposure window. The lesson each case teaches: pair high exposure with high meaning, and liking keeps climbing.
Key Findings
- Exposure works through fluency (Reber, Winkielman & Schwarz, 1998; Reber et al., 2004). Repeatedly seen images are processed near-effortlessly; the effortlessness is felt as pleasure and attributed to the image itself — liking on first view, for free.
- The effect is strongest for visual stimuli (Montoya et al., 2017, Psychological Bulletin; 268 curve estimates). The literature's strongest condition — visual material — is a purely visual collection's default condition.
- Exposure is a priced variable on-chain (Kapoor et al., 2022, TweetBoost; 245,159 tweets, 62,997 assets). Social-exposure features lifted NFT valuation prediction by 6% over on-chain-only baselines, with raw engagement among the top predictors.
- Repetition raises believed importance, not just liking (Hasher et al., 1977; Dechêne et al., 2010, d = 0.53). Familiar claims and familiar images are more readily accepted as significant — one reason culturally rehearsed events feel self-evidently monumental.
- Shared exposure is the most social form of familiarity. The events in question were seen together — through broadcasts, textbooks, and anniversaries — making the familiarity collective rather than private.
- Meaning immunizes against tedium (Bornstein, 1989; Berlyne, 1970). Emotionally loaded, complex stimuli sustain rising liking far past the ~35-exposure threshold that wears out simple ones; historical images sit permanently on the climbing arm of the curve.
Why This Matters for Meme-orial
Meme-orial does not generate familiarity; it inherits it. The generative-PFP era was built on the opposite premise — invent 10,000 brand-new faces nobody had ever seen and hope exposure would accrue after mint, through community grinding and influencer spend. That is exposure built from zero: expensive, slow, and fragile. Meme-orial's 104 subjects already sit at the saturation ceiling of human visual memory, encountered thousands of times per person across textbooks, television, documentaries, anniversaries, and memes. The collection was designed around a decades-validated affect mechanism that almost no other project uses, applied to the one subject matter where the exposure has already happened.
The construction follows from the science. One token per event keeps each image's accumulated familiarity undiluted. The violet/pink meta-layer encodes the shared dimension of that exposure — the part of cultural memory everyone experienced together — making the collective act of remembering itself visible in the artwork. Decade, country, and topic traits let collectors organize memory they already possess rather than decode invented lore. And because the subjects are maximally meaningful, the inverted-U research says continued circulation adds affection rather than fatigue: history, unlike a jingle, does not wear out. The result is a fixed set of 104 monuments to images the species has already logged — with the liking, as Zajonc showed, arriving on its own.