ARTIFACT · ATTENTION & HABIT
Reward Prediction
The brain learns most from the gap between expected and actual reward, which makes unpredictable rewards especially powerful—and risky to exploit.
↔
Genes:
- Prediction Error: Learning comes from the mismatch between expectation and outcome, not from reward alone.
- Dopamine Surprise: Dopamine is framed as tracking surprise, signaling when outcomes differ from what was predicted.
- Reward Fading: When rewards become expected, their impact weakens because the surprise signal drops.
- Variable Hook: Unpredictable rewards retain attention by repeatedly creating new surprise gaps.
- Design Choice: Systems should be built to support learning signals instead of manufacturing craving or compulsion.
Relationships:
- Prediction Error drives Dopamine Surprise
- Prediction Error explains Reward Fading
- Reward Fading creates demand for Variable Hook
- Dopamine Surprise amplifies Variable Hook
- Prediction Error implies Design Choice
- Variable Hook raises risk for Design Choice
Fingerprint 52cfae27d996a90f.