After a Scandal, People Downgrade Their Memory of the Artist but Not the Art, Five Experiments Find

In 2014, lawsuits against Bill Cosby forced a lot of people who grew up with The Cosby Show to decide what to do with a sitcom they had loved. A study out this month suggests many settle it in an odd place: they keep the memory of the fun, and quietly rewrite the memory of the man.

The paper, published in the Journal of Experimental Social Psychology, comes from Sally MyungJin Shin, an assistant professor of marketing at the University of South Florida, with Paul Rozin of the University of Pennsylvania and Gal Zauberman of Yale. Its title includes the phrase "Moral surgery." The surgery is performed by memory, and it is selective.

What the experiments compared

According to the University of South Florida's summary of the work, participants rated creators and their creations, then rated them again after learning of a transgression, so the two sets of scores could be compared. The press coverage names Cosby, Ray Charles (parental negligence) and John Lennon (physical abuse) as examples of the pattern.

The abstract on Shin's own website adds the design details. The effect appeared with positive opinions the researchers induced and with ones people already held. It held across several creative domains and at different levels of transgression severity. The authors say they ruled out low effort, temporal proximity and demand characteristics as explanations.

The twist is in the tense

Here is the part that makes the finding more than a restatement of "separate the art from the artist." Phys.org's account of the release says that when people were asked how they feel now, admiration dropped for both the creator and the work. The split showed up only when they were asked to look backward.

Shin put it this way: people "selectively downgrade past views of the creator while keeping past enjoyment of the creation shielded." She also described the separation as something that "exists only when we look back in time." In other words, the present-day verdict is harsh on everything. The memory is where the protection happens.

The authors' reading, as relayed in the coverage, is that the edit protects something people value: the good time they had. It is a tidy account. It is also an interpretation of lab ratings, and the coverage does not say how well recalled ratings track what people actually do afterward, such as rewatching a show or skipping it.

What the coverage cannot tell us

Futurity and Phys.org both republished the same University of South Florida release, so they count as one description of the study rather than two independent ones. The author abstract page is a second, separate look at the paper itself. The publisher's page for the article blocked automated access, so this piece runs on the abstract and the release, not the full text.

That leaves real gaps. There are no participant numbers, no effect sizes and no statistics in anything reviewed. It is also unclear whether Cosby, Charles and Lennon were each experimental stimuli or simply illustrations the release chose. And there is no base rate: nothing in the sources says how often people behave this way in real life compared with the lab.

What the abstract does support is narrower and more interesting than the headline version. When a beloved creator falls, the argument about whether to keep watching and listening happens in the present, where both the person and the work take a hit. The memory of the afternoon you spent laughing at the show has already been quietly protected.