Getting to know you: general and specific neural computations for learning about people

Area of Science:

• Neuroscience
• Social Cognition
• Computational Psychiatry

Background:

• Understanding others' traits is key to social cognition.
• Cortical networks activate during social processing, but their specific computations remain unclear.
• Investigating domain-general versus domain-specific computations in social learning is crucial.

Purpose of the Study:

• To identify brain regions involved in learning social attributes.
• To differentiate domain-general from social-specific computations in the brain.
• To explore computational approaches in social cognition.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used on 26 participants.
• Participants learned about individuals' generosity or lotteries' win/loss rates.
• A control condition contrasted social learning with non-social learning.

Main Results:

• General prediction error signals (social and non-social) were observed in the dorsomedial/dorsolateral prefrontal and right lateral parietal cortices.
• Socially specific prediction error signals were identified in the precuneus.
• Social prediction errors localized to a distinct subregion within the precuneus and posterior cingulate cortex.

Conclusions:

• The findings highlight both domain-general and social-specific computations in learning about others.
• This research demonstrates the utility of computational methods in understanding social cognition.
• Specific brain regions, like the precuneus, show specialized roles in social learning.