Beliefs about bad people are volatile.

Area of Science:

• Cognitive Science
• Social Psychology
• Moral Psychology

Background:

• Humans form moral impressions quickly, even from a young age.
• Identifying 'bad' agents is faster and more accurate than 'good' ones, a crucial survival mechanism.
• Inaccurate moral judgments can harm social relationships, highlighting the need for precise moral inference.

Discussion:

• Moral inference follows an asymmetric Bayesian updating mechanism.
• Beliefs about immoral agents exhibit greater uncertainty and volatility compared to moral agents.
• This uncertainty extends to non-moral traits of 'bad' agents, suggesting a general learning asymmetry.

Key Insights:

• A cognitive mechanism allows flexible updating of beliefs about potentially threatening individuals.
• This asymmetry in moral learning may facilitate forgiveness of those with initially negative impressions.
• Negative moral impressions destabilize beliefs, promoting cautious and cooperative social behavior.

Outlook:

• Understanding the computational processes of moral inference is key to social functioning.
• The findings offer insights into the cognitive basis of forgiveness and social adaptation.
• Future research could explore the developmental trajectory and neural underpinnings of this asymmetric updating.