Gurnani, Harsha; (2022) Population dynamics of electrically-coupled inhibitory Golgi interneurons in the cerebellar cortex. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text main.pdf - Submitted Version Download (63MB) | Preview

Abstract

The cerebellum has long been implicated in motor learning and sensorimotor coordination. In the input layer of the cerebellar cortex, the transformation of mossy fibre inputs into activity of granule cells can be powerfully controlled by the main inhibitory neurons – Golgi cells (GoCs). This thesis describes experiments that combine two photon imaging during spontaneous and trained behaviours, population activity analyses, and computational modelling to investigate how populations of inhibitory Golgi neurons coordinate their activity. The experiments, which were enabled by a 3D random-access acousto-optic lens two-photon microscope, revealed multidimensional activity structure of GoC networks. A slow, network-wide modulation of GoCs correlates with behavioural engagement, while faster differential modulation of individual GoCs encodes more precise sensorimotor information. This suggests the same network could perform two computations – global inhibition modulation can be utilized for adaptively controlling the gain of granule cells, while the faster differential modulation of individual GoCs can shape spatiotemporal patterns of granule cell spiking. To examine the potential role of electrical coupling between GoCs and of the structure of feedforward inputs, I built biophysically constrained computational models of the GoC network. The model not only reproduced the experimental observations but showed that the structure of electrical coupling – spatial scales, connection probability, gap junction conductance – critically determined the dynamical regime of the network, from tight synchrony to multidimensional structure to uncorrelated activity. Finally, to examine the contribution and nature of excitatory inputs to Golgi cells, I developed experimental setups to simultaneously record mossy fibres and Golgi cells during an associative conditioning task in mice. This provides further data to constrain and validate models of cerebellar input layer, and for developing a theoretical and conceptual framework for the role of inhibition in shaping cerebellar cortical representations that contribute to learning and performing coordinated movements.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item