Circuit mechanisms of working memory maintenance and distractor interference

Abstract

The main goal of this thesis is to study the circuit for working memory maintenance from a mechanistic perspective. To do so, I combine behavioral experiments with neuroimaging techniques and neuronal recordings under the framework of the bump attractor model for visuospatial working memory. The work performed during this thesis is encapsulated in two main chapters, one focusing on describing the topography of the working memory circuit, and the other focusing on how distractors interfere with the working memory content.

In the Chapter Topography of the working memory circuit, I test the Sensory Recruitment Theory by evaluating whether encoding and maintenance have the same topographical signatures, as expected if they share the same neural circuit. I provide behavioral, modeling, and electrophysiological data supporting the idea that prefrontal working memory maintenance is separated from encoding processes and mediated by attractor dynamics. Furthermore, I will extend the bump attractor model to cover the radial dimension and provide a mechanistic explanation for the compression of the visual field (foveal bias) with delay.

In the Chapter Distractor filtering in the working memory circuit, I evaluate how distractors interfere spatially and temporally with working memory maintenance at the behavioral and fMRI levels. I evaluate the results in the framework of the bump attractor model, and I explore different control strategies to deal with distracting information. Furthermore, in this chapter I re-analyze two electrophysiological datasets (Suzuki & Gottlieb, Nat. Neurosc., 2013 and Qi et al., Cell Reports, 2021) to test some predictions of the model and to mechanistically explain cholinergic improvement of working memory in a distractor-filtering task when stimulating the Nucleus Basalis of Meynert.