PSYREFLECT
RESEARCHJuly 20, 20262 min read

When the brain stops rewarding its own effort

Key Findings
  • In 95 young people at clinical risk for psychosis (subthreshold positive symptoms) plus 31 healthy controls, a purpose-built fMRI "fractal memory" task separated intrinsic motivation (mastery, curiosity) from extrinsic reward, and the ventral striatum tracked reward prediction error during performance feedback.
  • The ventral striatum and connected motivation circuitry activated more strongly to higher-confidence choices and to reward prediction error – that is, to the brain's own internal evaluation of performance, not only to external payoffs.
  • Ventral-striatum activation during task choices was tied selectively to intrinsic motivation and not to extrinsic motivation, dissociating the two systems at the level of neural signal.
  • Across both the psychosis-risk and control groups, weaker ventral-striatum activation tracked lower self-reported intrinsic motivation dimensionally, cutting across the diagnostic boundary rather than respecting it.

Predictive processing usually reaches psychosis through its loud symptoms: a prediction-error system with mis-set precision assigns significance to noise, and aberrant salience hardens into delusion. This study from the University of Pennsylvania looks at the quiet end of the same machinery – amotivation, the negative symptom that predicts disability and psychotic transition – and asks whether it is a failure of the brain to reward its own effort.

What the task was built to isolate

The problem with studying motivation in psychosis has always been that intrinsic and extrinsic drives are entangled in ordinary reward tasks. Curiosity and mastery arrive on the same trial as points and money. The authors built an fMRI "fractal memory" paradigm with feedback conditions designed to pull those apart, ran it in 95 adolescents and young adults at psychosis risk and 31 controls, and paired the imaging with self-reported intrinsic and extrinsic motivation.

The ventral striatum – the dopaminergic hub already implicated in aberrant salience – responded exactly where a predictive-processing account expects a self-generated reinforcement signal to live. It activated to reward prediction error during feedback, but also to higher-confidence choices: the moments when the person is evaluating their own performance against an internally set expectation. Crucially, that choice-related striatal signal tracked intrinsic motivation selectively, not extrinsic motivation. And its magnitude fell with self-reported intrinsic motivation across both groups, dimensionally – a transdiagnostic gradient, not a categorical deficit.

What changes at the bedside

The reframe is the useful part. Amotivation in a high-risk young person is easy to read as low effort, poor engagement, or "not trying." This puts a mechanism under it: the striatal signal that should turn one's own mastery and curiosity into reinforcement is blunted, so effort genuinely stops paying itself back. Extrinsic scaffolding – tokens, external rewards, contingency management – leans on the system that is comparatively spared here, which is why it can produce compliance without restoring drive. Interventions that rebuild a sense of competence and internally-set goals are aiming at the signal that is actually impaired. It also argues against waiting for a categorical threshold: because the striatum-to-intrinsic-motivation link is dimensional, the amotivated but not-yet-converted patient is on the same continuum, and their disengagement is a clinical target now, not a prodrome to observe.

In psychosis risk the problem may not be that the world stops offering rewards, but that the brain stops paying itself for its own effort.

Limitations

This is a cross-sectional imaging association in a risk sample defined by subthreshold symptoms, not transition, and the abstract reports directional effects rather than effect sizes or outcomes; the control group (n=31) is small relative to the risk group (n=95).

Source
Schizophrenia Research
Impaired intrinsic motivation in psychosis risk relates to ventral striatum activation to self-generated reinforcement signals
2026-04-16·View original
Tags
psychosis riskventral striatumprediction errorintrinsic motivationnegative symptoms
Related
Industry
Before Prediction Error Becomes a Biomarker: A Moscow Lab Reads the Fine Print
Brain SciencesRead →
Research
One symptom, two broken computations: why effort fails differently in psychosis and mood disorders
Molecular PsychiatryRead →
Tool
Mismatch Negativity: A Prediction-Error Index That Flags iTBS Responders in Depression
Journal of Affective DisordersRead →
PsyReflect · Free · Mon & Thu
Get analyses like this every Monday and Thursday.
Only what matters for practice. Curated by a clinical psychologist. 5 minutes instead of 4 hours of monitoring.
← Previous
Eleven Items to Flag Borderline Features in Early Adolescence