Date Thesis Awarded

5-2019

Document Type

Honors Thesis

Degree Name

Bachelors of Science (BS)

Department

Mathematics

Advisor

Mainak Patel, M.D., Ph.D.

Committee Members

Daniel Cristol, Ph.D.

Michael Drew LaMar, Ph.D.

Abstract

The easily identifiable structure and clearly defined function of the rodent somatosensory barrel cortex has made it a model system for study in neuroscience. The barrel consists of ~3600 regular-spiking (RS) cells that receive inhibitory inputs from(FS) cells, both of which are excited by ~240 thalamocortical (TC) cells. RS cell population dynamics such as rate of spiking and region of increased spiking can encode velocity and directional information from the initial whisker deflection, however, behavior of any single RS cell (or small group of RS cells) may be ambiguously affected by both velocity and directional changes. This project set out to create an additional layer of RS-like cells, hereby referred to as “excitatory extensions,” whose individual behavior or small group dynamics can correctly classify velocity and directional information from the incoming input stimulus. The model shows that an architecture that takes advantage of the net scaling of RS cell layer activity can lead to EE cells that correctly classify velocity without dependency on directional input. The model also shows that an architecture that engages an inhibitory feedback system can lead to EE cells that correctly classify direction without dependency on velocity input. Further areas of study include putting two barrel systems in communication with one another while limiting the model to experimentally-supported cell dynamics to see if architecture similar to the ones discovered in this project arise.

Available for download on Wednesday, May 05, 2021

Share

COinS