Causal Role of Cortico-Cerebellar Networks in Working Memory

点击翻译按钮获取中文摘要

PROJECT SUMMARY/ABSTRACT Working memory (WM) enables the short-term maintenance of mental representations in support of ongoing cognitive operations and is thought to underlie optimal cognitive performance across a wide range of contexts. Achieving a better understanding of the psychological and neural mechanisms of WM can provide valuable insight into one of the primary determinants of both cognitive success and failure. A substantial body of literature has sought to reveal the specific brain structures and mechanisms that support WM. To date, this literature has predominantly focused on the contributions of a limited set of fronto-parietal cortical areas. Recent work, however, additionally implicates portions of the cerebellum in WM processes. Yet, there remain substantial gaps in our understanding of the precise role that the cerebellum plays in WM. As prior work has primarily used correlational measurement techniques to argue for cerebellar contributions to WM, the question of whether the cerebellum is actually necessary for the maintenance of information in WM remains unanswered. The goal of the proposed research is to investigate the causal role of the cerebellum in WM using a combination of transcranial magnetic stimulation (TMS), functional magnetic resonance imaging (fMRI), and computational modeling of behavior. In Aim 1, participants will receive TMS to disrupt activity in regions of interest in cerebellum, parietal cortex, and frontal cortex in independent sessions just prior to the performance of a task that requires precise recall of presented spatial locations. We will then examine the effect of this disruption on parameters derived from a computational model that index distinct aspects of WM, such as the fidelity of memory or the probability that an item is successfully encoded. In Aim 2, we will combine TMS with fMRI to examine the causal influence of cerebellar and cortical areas on neural markers of WM storage throughout the brain. Specifically, we will use a Bayesian generative model to decode the location stored in memory, as well as the uncertainty with which that location is encoded. The proposed experiments will enable the dissociation of the relative roles of cerebellar and cortical areas in WM task performance, thus providing a more nuanced understanding of the neural underpinnings of WM. A more complete characterization of the WM system has the potential to inform the development of therapies aimed at treating a broad range of clinical conditions for which WM dysfunction is a primary symptom.