Skip to content
PaperFren

How does the striatum process reward and punishment during learning?

Different subregions of the striatum handle unexpected feedback differently, with the front-bottom area responding to both reward and punishment, while the back-top area reacts only to unexpected reward.

Source

Dissociable responses to punishment in distinct striatal regions during reversal learning

Robinson OJ, Frank MJ, Sahakian BJ, et al. · NeuroImage · 2010

doi.org/10.1016/j.neuroimage.2010.03.036Read the full paper ↗63 citationscc by

What they did

The researchers scanned 16 healthy adults using fMRI while they performed a decision task where they had to predict whether a face or scene stimulus would lead to a reward or a punishment. The rules unexpectedly reversed multiple times throughout the experiment, forcing participants to update their predictions. By analyzing brain activity during these unexpected shifts, the team mapped how different parts of the striatum responded to surprising rewards versus surprising punishments.

What they found

They discovered that the anterior ventral striatum became active during both unexpected rewards and unexpected punishments, showing a non-specific response to surprise. In contrast, the posterior dorsal striatum only showed increased activity during unexpected rewards. Additionally, the amygdala showed a drop in activity specifically during unexpected punishments.

The limits

What it doesn't show

Because participants could not control whether they received a reward or punishment, this study cannot show how these striatal regions behave when actions directly prevent or cause outcomes. Additionally, the small sample of 16 participants limits how well these findings generalize to broader populations. Finally, as an fMRI study, it can only show correlation between brain activity and learning, not whether these specific brain regions are causally required for reversal learning.

Key terms

striatum
A critical subcortical brain structure involved in motor planning, decision-making, and reward processing.
reversal learning
A paradigm where subjects must adapt their behavior when previously established rules or reward contingencies suddenly change.
Pavlovian conditioning
A learning process in which a neutral stimulus becomes associated with a biologically significant event, regardless of the subject's actions.
instrumental learning
A learning process where an organism associates its active behavior with subsequent outcomes or consequences.
prediction error
The difference between the expected outcome of an action or stimulus and the actual outcome received.
functional magnetic resonance imaging (fMRI)
A non-invasive neuroimaging technique that measures brain activity by detecting changes in blood flow.

Flashcards

1 / 34

Quiz yourself

1 / 10

How did the study's task design decouple the subjects' choices from the received outcomes?

Common questions

Why is it important that the task did not let subjects control the outcomes?

By making the outcomes independent of the subjects' choices, the researchers could isolate the pure prediction of rewards and punishments without the confounding factor of planning physical actions to actively avoid or obtain those outcomes.

What is the difference between the ventral and dorsal striatum found in this study?

The ventral striatum showed a general, valence-nonspecific activation to any unexpected outcome, whereas the dorsal striatum was highly selective, activating only for unexpected rewards.

What did the researchers find in the amygdala?

Unlike the striatum, the amygdala showed a specific decrease in activity below baseline levels when participants received an unexpected punishment.

How do these findings help resolve debates in neuroscience?

They reconcile conflicting past studies by showing that different subregions of the striatum handle prediction and action-selection separately, explaining why some studies found punishment responses in the striatum while others did not.

More on Reward