The term thin slice comes from a frequently cited article by Nalini Ambady and Robert Rosenthal, in which subjects evaluated thirty-second silent video clips of instructors teaching a class.’ Subsequent analysis found that these brief evaluations predicted the instructors’ end-of-semester student ratings. Their work built on earlier research that found a similar predictive power in job interviews,’ where the first impressions were critical for the eventual hiring decision.
Thin slices of behavioral data have been shown to predict a broad range of consequences, including therapist competency ratings, the personalities of strangers, and even courtroom judges’ expectations for criminal trial outcomes. One of the most impressive examples of thin slices predicting important, long-term consequences is the marital research conducted by John Gottman and his colleagues. Sybil Carrere and Gottman were able to predict marital outcomes over a six-year period based on human microcoding of just the first three minutes of a marital conflict.
Across a wide range of studies, Ambady and Rosenthal found that observations lasting up to five minutes had an average correlation of r = .39 with subsequent behavior, which corresponds to 70 percent accuracy at predicting outcomes…
Join over 100,000 readers. Get a free weekly update via email here.
I want to subscribe!