Skip to main navigation Skip to search Skip to main content

Modeling the minds of friends: Dorsomedial prefrontal cortex encodes accurate dyadic meta-evaluations about how others perceive us

Research output: Working paperPreprint

Abstract

Accurate meta-evaluations of the self—or accurately understanding how other people perceive us—are crucial for successful interpersonal interactions and relationships. In general, people can make accurate meta-evaluations across many social contexts and for many attributes, such as personality traits, abilities, and preferences. Yet it is still poorly understood how the brain accurately encodes others’ evaluations of the self. In a dyadic study design, we recruited pairs of friends who completed trait evaluations about themselves and each other during functional neuroimaging. Participants evaluated themselves, their partner (other-evaluation), and also predicted their partner’s evaluation of them (meta-evaluation). We replicated prior behavioral findings demonstrating that close friends are accurate at making meta-evaluations, and univariate neural findings that the default mode network supports both other- and meta-evaluations. Using dyadic representational similarity analysis, we found high correspondence between neural representations of participants’ meta-evaluations and exemplar models of their partners’ actual other-evaluations in dorsomedial prefrontal cortex. Neural encoding accuracy in this region also increased as a function of behavioral prediction accuracy. These results suggest that dorsomedial prefrontal cortex encodes a model of others’ evaluations, which may facilitate the accurate insight into others’ minds. In other words, it provides an implicit map used to model the minds of our friends.
Original languageAmerican English
DOIs
StateSubmitted - 2024

Fingerprint

Dive into the research topics of 'Modeling the minds of friends: Dorsomedial prefrontal cortex encodes accurate dyadic meta-evaluations about how others perceive us'. Together they form a unique fingerprint.

Cite this