Reward prediction errors can update the learned value of actions and influence subsequent reward-seeking choices.
Emerging or mechanistic evidence
Pattern System · Output System · Adaptation System
Last reviewed: August 11, 2026
EVIDENCE BY TOPIC
Reward prediction error
How differences between expected and experienced outcomes can update future action values.
The entries below are connected by subject, not arranged as a protocol or a complete literature review.
Open an individual claim to see what the evidence supports, what it does not establish, which sources are carrying the interpretation, and where the claim appears in The Human OS Manual.