Dopamine reward prediction error coding.
Level 5 - mechanism / opinion, no new human data
Narrative review synthesizing mechanistic animal and human neuroscience without original data or systematic methodology
PubMed 27069377 · doi:10.31887/DCNS.2016.18.1/wschultz
What was done
This review synthesizes neurophysiological and neuroimaging findings from rodents, monkeys, and humans on how midbrain dopamine neurons and interconnected structures (striatum, amygdala, and frontal cortex) encode reward prediction errors.
What was found
The abstract reports no quantitative values or effect sizes. It describes neurobiological signaling characteristics: midbrain dopamine neurons increase firing during positive prediction errors (reward greater than predicted), maintain baseline activity for fully predicted rewards, and decrease firing during negative prediction errors (reward less than predicted), nonlinearly coding formal economic utility.
Why it matters
It outlines the neuronal implementation of reward prediction errors, providing a mechanistic link between reinforcement learning models, economic utility, and addiction pathology.
Limits
The abstract lacks systematic search criteria, quality appraisal of included studies, and quantitative data. Findings rely largely on mechanistic animal neurophysiology alongside human neuroimaging.
Cited by
- context Dopamine neural circuitry operates such that following a major success, subsequent achievements will not feel rewarding unless they surpass the magnitude of the prior event.