Thesis defense Bahadir Kasap (Donders series386)

9 July 2019

Promotor: prof. dr. A. van Opstal

Neuro-computational modeling of the gaze control system

How long does it take to check the rear mirrors when the car behind is honking? Such effortless rapid movements require nontrivial information processing steps for the brain to handle. Different sensory systems, learned beliefs on how the world work and decision mechanisms are involved in this task, yet once a target location is available gaze control system orchestrates the muscles for the goal directed movement.
In this work, we studied the computational principles of the gaze control system, specifically by focusing on the role of the midbrain superior colliculus (SC) in gaze control. The gaze control system orchestrates coordinated eye and head movements for fast and accurate gaze shifts (‘saccades’) towards a target location. SC signals encode the kinematics of saccades dynamically, with a small contribution of each neural activation, to drive downstream control plant with information on the direction and size of the movement. We studied optimal control signal generation in SC by recreating firing patterns of SC populations in a spiking neural network model. Positioning SC in the gaze control system as an instantaneous nonlinear vectorial pulse generator, we proposed a new perspective on gaze control system.