Neuronal population activity related to inhibition-of-return saccadic phenomena

Poster Session F - Tuesday, April 16, 2024, 8:00 – 10:00 am EDT, Sheraton Hall ABC

Ivan Skelin^1,2 (ivan.skelin@uhn.ca), Marco Istasy³, Chaim Katz⁴, Qian Chu^1,5,6,7, Taufik Valiante^{1,2,8,9,10,11,12}; ¹Krembil Brain Institute, Toronto Western Hospital (TWH), ²Center for Advancing Neurotechnological Innovation to Application (CRANIA), Toronto, ³Faculty of Medicine, University of Toronto, ⁴Faculty of Medicine, University of Calgary, ⁵Max Planck - University of Toronto Center for Neural Science and Technology, ⁶University of Toronto, ⁷Max Planck Institute for Empirical Aesthetics, ⁸Electrical and Computer Engineering, University of Toronto, ⁹The KITE Research Institute, University Health Network, Toronto, ¹⁰Institute of Biomedical Engineering, University of Toronto, ¹¹Institute of Medical Sciences, University of Toronto, ¹²Division of Neurosurgery, Department of Surgery, University of Toronto

Inhibition of return (IOR) is a phenomenon whereby the saccadic eye movements show a lower probability of returning to previously explored parts of a scene. IOR likely depends on the interactions between the neuronal populations involved in eye movement planning and short term memory. However, the brain circuitry level understanding of mechanisms underlying IOR in the human brain is still incomplete. We took advantage of a wide electrode coverage in epileptic patients undergoing pre-surgical evaluation of epileptic foci to measure population level activity (high frequency activity, HFA; 70-200 Hz) distinguishing saccades that return to previously explored areas (return saccades), relative to saccades directed towards non-explored areas (non-return saccades). On individual trials, participants explored naturalistic scenes, searching for hidden targets. The analysis was done on 10602 saccades (16 datasets, 15 subjects), and a total of 536 electrodes. Overall, 12.3 ± 0.6% of saccades were classified as ‘return’, defined as ending up within the arbitrary radius of 150 pixels (15%/11% of the scene height and width, respectively) from any of the previous saccade end points during the same scene exploration. A distributed network of local populations shows significantly different pre-saccadic HFA, depending on the saccade category (return or no return), with the higher proportion in the occipital lobe (23%), relative to temporal (10%) and frontal lobe (8%). These findings suggest the IOR might result from a balance between the activity of distributed networks promoting return to previously explored or exploration of new parts of a scene, respectively.

Topic Area: PERCEPTION & ACTION: Vision