Cognitive and Information Sciences Professor Paul Smaldino is part of a team of researchers whose study titled "Psychologists update their beliefs about effect sizes after replication studies" was recently published in Nature Human Behaviour. The research focused on whether psychologists were able to predict how they would update their confidence in a result based on successful or unsuccessful replications; and whether they behaved as predicted by a model that relied solely on information from these studies to update their beliefs.
The study showed psychologists did change their minds in the proper direction, and they were also good at predicting how their beliefs would change in light of evidence. Importantly, the degree to which psychologists updated their beliefs was not affected by whether or not the results turned out differently than they expected. These results give hope that behavioral scientists may be less influenced by motivated reasoning than sometimes feared and are able to rely rationally on an accumulated body of evidence to learn true aspects of human behavior.
