Electronic Thesis and Dissertation Repository


Doctor of Philosophy




Everling, Stefan


Cognitive control is crucial to voluntary behaviour. It is required to select appropriate goals and guide behaviour to achieve the desired outcomes. Cognitive control is particularly important for the ability to adapt behaviour to changes in the external environment and internal goals, and to quickly switch between different tasks. Successful task switching involves a network of brain areas to select, maintain, implement, and execute the appropriate task. Uncovering the neural mechanisms of this goal-directed behaviour using lesions, functional neuroimaging, and neurophysiology studies is central to cognitive neuroscience.

The oculomotor system provides a valuable framework for understanding the neural mechanisms of cognitive control, as it is anatomically and functionally well characterized. In this project, pro-saccade and anti-saccade tasks were used to investigate the contributions of oculomotor and cognitive brain areas to different stages of task processing. In Chapter 2, non-human primates performed cued and randomly interleaved pro-saccade and anti-saccade tasks while neural activity was recorded in the superior colliculus (SC). In Chapter 3, non-human primates performed cued and randomly interleaved pro-saccade and anti-saccade tasks while local field potential activity was recorded in the SC and reversible cryogenic deactivation was applied to the dorsolateral prefrontal cortex (DLPFC). In Chapter 4, non-human primates performed uncued and cued pro-saccade and anti-saccade switch tasks while reversible cryogenic deactivation was applied to the dorsal anterior cingulate cortex (dACC).

The first study clarifies that macaque monkeys demonstrate similar error rate and reaction time switch costs to humans performing cued and randomly interleaved pro-saccade and anti-saccade tasks. These switch costs were associated with switch-related differences in stimulus-related activity in the SC that were resolved by the time of saccade onset. The second study shows that bilateral DLPFC deactivation decreases preparatory beta and gamma power in the superior colliculus. In addition, the correlation of gamma power with spike rate in the SC was attenuated by DLPFC deactivation. Lastly, bilateral dACC deactivation in the third study impairs anti-saccade performance and increases saccadic reaction times for pro-saccades and anti-saccades. Deactivation of the dACC also impairs the ability to integrate feedback from the previous trial.

Overall, these findings suggest unique roles for the dACC, DLPFC, and SC in cognitive control and task switching. The dACC may monitor feedback to select the appropriate task and implement cognitive control, the DLPFC may maintain the current task-set and modulate the activity of other brain areas, and the SC may be modulated by task switching processes and contribute to the production of switch costs.