Multiple excitatory cell types in primary and secondary visual cortex are recruited during accumulation-of-evidence decisions

Poster Presentation: Friday, May 16, 2025, 3:00 – 5:00 pm, Banyan Breezeway
Session: Decision Making: Perception

Lillian Wilkins1, Joao Couto1, Anup Khanal1, Anne K Churchland1; 1Department of Neurobiology, University of California, Los Angeles

The mammalian cerebral cortex plays a central role in decision-making and different visual cortical regions have unique functional roles in processing visual information. Further, cortical excitatory pyramidal neurons (PyNs) have rich diversity, but the unique roles of different PyN types across primary and secondary visual areas in decision-making remains unclear. Here, we address this gap by characterizing the activity of two classes of excitatory neurons in mouse anteromedial (AM) and primary (V1) visual cortex. We retinotopically mapped the visual cortex and used 2-photon calcium imaging to record neural activity while mice passively view stimuli or make perceptual decisions in a novel accumulation of evidence task. In the task, visual patches appeared on the left and/or right sides of the screen. Head-fixed mice judged the side with higher rate, held their choice for a brief delay period, and reported by licking one of two spouts. We recorded from ~6,000 cells in 7 mice (~3000 in AM and V1), out of which approximately 60% were responsive during the task. We measured responses in layer 2/3 (L2/3) and layer 5b (L5b) of AM and V1 and, in each population, we found both sharp orientation tuning in the passive condition and stimulus selectivity during decision-making. Interestingly, L5b neurons (that mostly target structures outside of cortex) are highly stimulus selective: individual neurons respond to stimuli corresponding to one choice option but not the other. Overall, we conclude that L2/3 and L5b neurons in both AM and V1 are highly selective to stimulus features during task performance, which invites questions about the roles of these responses in accumulation of evidence decision-making.

Acknowledgements: This work was supported by the National Institute of Health, award R01EY022979.