Neural mechanisms of resource allocation in working memory.

Area of Science:

• Cognitive Neuroscience
• Neuroimaging
• Computational Neuroscience

Background:

• Working memory has limited capacity, necessitating resource allocation based on item relevance.
• The neural basis for prioritizing information within working memory is not well understood.

Purpose of the Study:

• To investigate the neural mechanisms underlying working memory prioritization.
• To decode and differentiate neural responses for items with varying priorities in working memory.

Main Methods:

• Development of computational neuroimaging techniques to analyze neural activity.
• Decoding and demixing neural signals associated with multiple working memory items.
• Correlating neural activity patterns with behavioral measures of memory prioritization.

Main Results:

• Neural response gain in visual cortex (striate and extrastriate) correlated with item priority.
• Higher-priority items were decoded with greater accuracy and less uncertainty.
• Frontal cortex activity predicted neural differences in visual cortex, linking top-down control to representation fidelity.

Conclusions:

• Feedback signals from frontal cortex modulate neural representation gain in visual cortex based on behavioral relevance.
• This mechanism facilitates efficient working memory resource allocation.
• Identified a key neural pathway for prioritizing information in working memory.