Attractor dynamics gate cortical information flow during decision-making.

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Computational Neuroscience

Background:

• Decisions are stored in memory but can be disrupted by distracting sensory information.
• The neural mechanisms for gating sensory information to protect memory during decision-making are not fully understood.

Purpose of the Study:

• To investigate how the brain protects decisions from distracting stimuli during the delay between sensory input and action.
• To elucidate the network mechanisms underlying sensory gating in the context of decision-making.

Main Methods:

• Mice were trained to detect optogenetic stimulation of the somatosensory cortex with a delay between sensation and action.
• Neural activity was recorded, and behavioral responses to distracting stimuli were analyzed.
• Computational models, including reverse-engineered neural networks, were used to analyze motor cortex activity.

Main Results:

• Distracting stimuli progressively lost behavioral influence over time during the delay period.
• Neural activity evoked by distractors propagated through the cortex without attenuation.
• Motor cortex activity showed reduced sensitivity to distractors, explained by increasing separation between decision-related attractors.

Conclusions:

• Attractor dynamics in neural networks can gate communication between brain regions, controlling information flow.
• This gating mechanism protects decisions from interference by modulating sensitivity to distracting sensory input.
• The findings reveal a novel mechanism for maintaining decision commitment through attractor state separation.