Diederen · The Journal of neuroscience : the official journal of the Society for Neuroscience 2017 · randomized placebo-controlled between-subjects pharmacological fMRI study · n=?

Dopamine Modulates Adaptive Prediction Error Coding in the Human Midbrain and Striatum.

Level 2 - randomized trial

Randomized, placebo-controlled pharmacological trial

PubMed 28202786 · doi:10.1523/JNEUROSCI.1979-16.2016 · record verified 2026-08-26

What was done

Healthy participants completed a reward prediction task during fMRI in a between-subject, placebo-controlled pharmacological design. Participants received a dopamine antagonist (sulpiride), a dopamine agonist (bromocriptine), or placebo. During the task, participants predicted the magnitude of upcoming rewards drawn from distributions with varying standard deviations (reward variability) to elicit trial-by-trial prediction errors. Neural responses were measured using fMRI, and behavioral adaptation was evaluated via computational modeling.

What was found

The abstract reports directional findings without numerical values, effect sizes, or test statistics. Placebo participants demonstrated adaptive prediction error coding in the midbrain and ventral striatum. Sulpiride attenuated this adaptive coding in both the midbrain and ventral striatum, reduced overall performance, and decreased modeled behavioral adaptation. Bromocriptine had no significant impact on adaptive coding or performance. There was no differential effect of reward standard deviation on task performance between the groups.

Why it matters

The study directly links dopamine signaling to adaptive reward prediction error coding in the human midbrain and striatum, identifying a neural mechanism for how the brain optimizes learning in variable environments.

Limits

The abstract does not report sample size, participant demographics, drug dosages, or quantitative statistics (such as effect sizes or p-values). A between-subjects design introduces potential baseline variance across groups compared to a crossover design, and the study was conducted in healthy individuals rather than clinical populations.

Cited by