Learning from Others: Effects of Described Demonstrator Ability on Brain and Behavior

Observational learning allows us to learn to make decisions by observing the behaviors of others. The quality of such learning depends not only on the ability of the observed other, the demonstrator, but also on our beliefs about their abilities. We have previously demonstrated that observers learned to avoid an aversive outcome better from demonstrators described as high in ability, regardless of the demonstrator’s actual performance. Building on these findings, our study aimed to replicate these findings, and to investigate the neural mechanisms underlying description-sensitive observational learning. Forty-five participants engaged in an observational aversive learning task while undergoing functional magnetic resonance imaging. We manipulated the descriptions of the abilities of the demonstrators, while keeping their actual abilities low. We hypothesized that participants would perform better when demonstrators were described as having a high vs. low ability. We further investigated if brain activity in regions associated with observational learning, social impression formation and mentalizing were sensitive to the demonstrator’s described ability. Contrary to expectations, participants performed equally well regardless of the description of demonstrator ability. Subsequent analyses revealed however, that described ability influenced the level of copying, such that participants copied the choices of the demonstrators more when these were described as having high rather than low ability. Our findings point to the involvement of mentalizing processes combined with more general learning and decision-making processes in driving the behavioral effects of biased observational learning.