Sound-Bite Science

A recent Today Show segment on the U.S. network NBC featured an exposé on the concept of implicit bias. It presented the Implicit Association Test, a computer test of reactions to words and images with racial or other identity-oriented content as showing, definitively, that individuals are bigoted without even realizing it. Is the test accurate? Few viewers would have the expert knowledge to know; so they have to trust that it is. They have to believe science. But what does this mean? Is believing science about rationality, skepticism, and respect for the scientific method? Or is it an unquestioning adherence to expert proclamations, particularly if the experts in question have impressive affiliations?

As Jesse Singal points out, new results make it into the public eye, not by way of scholarly papers directly, but through media coverage, often based on press releases—a system can bring scientifically misleading declarations into public consciousness and, ultimately, common sense. Yet for Singal, media distortions of social-scientific findings are only a part of the problem; the science itself merits a more skeptical look. Singal, author of The Quick Fix, writes there and in his newsletter about popular behavioral science and its misinterpretations. He explains that the studies—mostly selected for catchiness rather than scientific urgency—are received in the general population as expertise, often despite serious methodological flaws. This, Singal thinks, has policy implications, from the military to human-resources departments, diverting resources away from practices that could more effectively address mental health, diversity, and other urgent issues.

Phoebe Maltz Bovy: How should we think about something like the Implicit Association Test, which the Today Show presented, as evidence?

Jesse Singal: I was surprised to see that segment, because it’s 2021. All sorts of questions have been raised about the accuracy of the Implicit Association Test, to the point where the test creators acknowledged a long time ago that it’s really too noisy to diagnose individuals. I thought it was telling that years after that, the Today Show is treating it as this exciting new thing everyone should know about, without any skepticism at all. Some of these ideas have only been debunked in the eyes of people who follow this stuff closely, or psychologists themselves, but they’re still out there, zombie ideas shambling around.