Knowledge Engine

Does seeing a painting often make it great?

·no read data

The idea

In 1968 the psychologist Robert Zajonc published a monograph in the Journal of Personality and Social Psychology showing that simply seeing something more often tends to make people like it more. He showed participants made-up Turkish-like words, Chinese-like characters and yearbook photographs of men, varying how many times each appeared, from zero up to 25. Ratings of how good or likeable each item seemed rose with exposure, roughly in step with the logarithm of how often it had been seen. This became known as the mere exposure effectmere exposure effect The tendency for people to like something more simply because they have seen or heard it more often..

In 2003 James Cutting, a perception psychologist at Cornell, asked whether the same process helps hold an artistic canon together. His case was Gustave Caillebotte, a painter and collector whose bequest of Impressionist works to the French state contains many now-famous paintings. Cutting built 66 pairs of images: each pair set a work linked to Caillebotte beside a similar painting by the same artist, of similar style, period and subject. To estimate cultural exposure he counted how often each image was reproduced in books in Cornell’s libraries, finding 4,232 reproductions across 980 books.

In his main study, 166 undergraduates saw each pair briefly and chose the one they liked more. They picked the more often reproduced image 59 percent of the time, while recognising only 2.7 percent of the images at all. Cutting concluded that mere exposure, much of it unnoticed, helps maintain the canon.

Why it might change a decision

The account makes a testable prediction about repetition. If exposure alone drives preference, showing people a weak painting many times should raise their liking for it, just as it would for a strong one. If quality matters too, liking for weak and strong works should move apart with repeated viewing. That difference matters for anyone judging whether a work’s reputation reflects its merits or its circulation, since the two accounts predict different results from the same repeated showings.

How solid is this?

Cutting’s paper reports six studies, and its claims rest on correlationalcorrelational Showing that two things vary together, without establishing that either one causes the other. data plus one classroom experiment. In a smaller sample of faculty and graduate students the more frequent image was again preferred, 57 percent of the time. Children aged 6 to 9 showed no such preference (51.2 percent), which fits the idea that the effect accumulates with years of exposure. Judgements of complexity and of how typical each painting was of Impressionism did not explain the preferences.

The obvious competing explanation is that frequency tracks quality: editors may reproduce better paintings more often, so people prefer them for the same reasons editors did. Cutting’s sixth study tried to address this. Across 21 class periods, 116 students were briefly shown the less often reproduced image of each pair four times and the more often reproduced one once. When later asked to choose, they preferred the more frequent image only 48 percent of the time, below the level in unexposed students. That comparison was between different groups of students, a weaker design than testing the same people before and after.

Aaron Meskin, Mark Phelan, Margaret Moore and Matthew Kieran, philosophers working with experimental methods, tested whether exposure helps bad art too. In a 2013 paper in the British Journal of Aesthetics they showed 57 third-year philosophy students 48 landscape paintings by Thomas Kinkade, whom they treated as a bad painter on the basis of critical opinion, and 12 by John Everett Millais, the Pre-Raphaelite. Images appeared for about two seconds mid-lecture across 14 classes; half of each set was shown five times and the rest once. Students then rated how much they liked each image on a 10-point scale, alongside a control groupcontrol group The people in a study who do not receive the treatment, so that the treated group can be compared against them. of 57 philosophy undergraduates who had not seen them.

Exposure lowered liking for the Kinkades: the exposed group’s mean was about 4.9 against 5.8 for controls, a statistically significant difference, and 47 of the 48 Kinkades were rated lower by the exposed group. Liking for the Millais paintings moved upwards (about 5.7 against 5.2), but not significantly; the authors estimated they would need around 200 participants to detect an effect of that size.

Their preferred reading is that repeated viewing helps people notice what is wrong with a bad painting, but they state plainly that their results do not establish this. They also considered boredom with the whole category of images, and the older finding that exposure can lower liking for stimuli people already dislike. They argued the Kinkades were not initially disliked, but these alternatives remain open. The sample was small, from one university, and the Millais result is a non-significant trend.

Neither study shows that exposure makes a canon on its own. Cutting’s evidence supports exposure as one influence on preference among non-experts; Meskin and colleagues suggest that influence depends on what is being shown. The popular reading that canons are mere familiarity outruns both.

Sources

  • Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9(2, Pt. 2), 1-27. doi:10.1037/h0025848
  • Cutting, J. E. (2003). Gustave Caillebotte, French Impressionism, and mere exposure. Psychonomic Bulletin & Review, 10(2), 319-343. doi:10.3758/BF03196493
  • Meskin, A., Phelan, M., Moore, M., & Kieran, M. (2013). Mere exposure to bad art. British Journal of Aesthetics, 53(2), 139-164. doi:10.1093/aesthj/ays060

Check yourself

Answers are checked in your browser. Nothing is recorded or sent anywhere.

  1. How did Cutting (2003) measure how much cultural exposure each Impressionist image had received?

  2. In Meskin and colleagues' 2013 study, what happened to liking for the Thomas Kinkade paintings after repeated classroom exposure?

  3. What did Cutting find when he tested children aged 6 to 9?

  4. Which is the fairest reading of the evidence on Cutting's main finding?

Questions without answers

  1. If editors and curators choose which works to reproduce partly on quality, how could anyone separate the effect of exposure from the effect of quality in real-world data on canons?
  2. Meskin and colleagues chose Kinkade as 'bad art' using critics' views. What would it mean for their argument if a different audience, say Kinkade's American fans, showed the opposite pattern?

Back to the archive