A distributed and efficient population code of mixed selectivity neurons for flexible navigation decisions

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Systems Neuroscience

Background:

• Decision-making requires cognitive flexibility to adapt actions based on sensory input and memory.
• Rapidly switching navigation strategies based on remembered information is a key aspect of flexible decision-making.

Purpose of the Study:

• To identify cortical areas and neural activity patterns enabling behavioral flexibility in decision-making during virtual navigation.
• To investigate the role of specific neural populations in integrating current sensory cues with stored memory for adaptive choices.

Main Methods:

• Utilized virtual navigation tasks in mice to study decision-making flexibility.
• Employed optogenetics screening to identify brain regions critical for accurate choices.
• Applied calcium imaging to record neural activity and identify neurons encoding mixed sensory and memory information.

Main Results:

• Optogenetics identified V1, posterior parietal cortex (PPC), and retrosplenial cortex (RSC) as essential for decision accuracy.
• Calcium imaging revealed mixed-selectivity neurons integrating current and remembered visual cues, mediating rapid navigation switches.
• These neurons emerged with task learning, predicted choices via efficient population codes, and were distributed across the posterior cortex, notably in RSC and PPC.

Conclusions:

• Flexible navigation decisions depend on neurons that combine visual and memory information.
• A visual-parietal-retrosplenial network, featuring mixed-selectivity neurons, underlies cognitive flexibility in decision-making.