Lee, Julie J;
(2020)
Task selectivity in mouse parietal cortex.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
THESIS_JJL_corrected.pdf Download (2MB) | Preview |
Abstract
A challenge for the brain is to flexibly deal with the vast variety of potential situations encountered in a lifetime. One way to deal with this challenge is to distribute multiple representations across the same neurons, another is to allocate distinct subpopulations for different contexts. Posterior parietal cortex (PPC) has been implicated in motor planning, decision making and navigation. Are the same PPC neurons involved in these diverse functions, and if so, how are these processes shared across the population? I recorded from neurons in PPC using two-photon calcium imaging and compared their activity in mice trained to perform two different visual detection tasks. In one task, the mouse turned a steering wheel to report whether a visual grating was on the left or right side. In the other, the mouse navigated through a virtual T-maze using a spherical treadmill and turned at the end to report whether a grating appeared on the left or right wall. Both tasks involved visual detection, choice, and motor execution, but many neurons were selectively active only in either task. Running could not fully account for task selectivity. Instead, selectivity was related to context more generally: activity during the task was well-predicted by activity during passive conditions on the same apparatus. Finally, I related each neuron’s task event-related activity to ask whether neurons active in both tasks shared choice and stimulus preferences. Neurons did not share choice preferences across tasks, suggesting that choice selectivity is not represented abstractly in single neurons, however, there was some evidence that stimulus representations were shared across tasks. These results indicate that PPC allocates distinct neuronal populations for different purposes, determined by the context in which a task is performed. Within each task, event-evoked activity of individual neurons may be governed by multi-dimensional characteristics specific to each context.
Archive Staff Only
 |
View Item |
