Reward impacts visual statistical learning.

Area of Science:

• Cognitive Neuroscience
• Visual Perception
• Learning and Memory

Background:

• Humans possess an innate ability for visual statistical learning (VSL), detecting regularities in visual environments.
• The influence of reward learning on VSL remains underexplored, despite similarities between reward learning and VSL.

Purpose of the Study:

• To investigate how reward contingencies impact visual statistical learning.
• To examine the behavioral and neural signatures of VSL under varying reward values.

Main Methods:

• Two experiments were conducted involving participants learning image values through a risky choice task.
• Participants underwent recognition and reward memory tests for image pairs (High-High, High-Low, Low-High, Low-Low).
• Brain responses were analyzed using fMRI to compare neural activity related to value and statistical contingencies.

Main Results:

• Higher reward values were associated with stronger behavioral and neural evidence of VSL.
• Recognition accuracy was better for pairs starting with high-value images, particularly High-High pairs.
• Brain activation in regions like the inferior frontal gyrus and hippocampus was greater when the first image had high value, indicating an interaction between reward and statistical learning.

Conclusions:

• Reward information embedded in stimulus associations can alter the VSL process.
• High-value first images may facilitate enhanced memory for statistically learned pairs and associated reward information.
• These findings highlight the interplay between value-based learning and statistical learning in visual perception.