Madeleine Heynen¹ (), Kerri Walter², Jonathan K. Doyon¹, Peter J. Bex², Lotfi B. Merabet¹; ¹Mass Eye and Ear, Harvard Medical School, ²Northeastern University

Cerebral visual impairment (CVI) is a brain-based visual disorder associated with early neurological injury and maldevelopment of visual processing pathways and areas. Functional visual deficits related to impaired object localization are commonly observed in CVI. Virtual reality (VR), paired with eye-tracking, offers an objective method to assess the nature of these higher-order visuospatial processing deficits. We characterized gaze behavior associated with target recognition abilities during desktop and immersive VR-based versions of a visual search task using a screen-mounted eye tracker (Tobii Spark 60 Hz) and Meta Quest Pro integrated eye tracking (100 Hz), respectively. Twelve CVI (19.67 years ± 6.39 SD) and 10 control (21.60 years ± 7.59 SD) subjects were instructed to search and locate a pre-identified target within a series of rooms with low and high clutter to investigate the effect of environmental complexity on the exploration of naturalistic scenes. Comparing success rates, there was a significant effect of group, with the CVI group showing an impaired ability to correctly locate the target (F(1,60)=23.67, p<0.001). Analyzing reaction times showed significant main effects of group (F(1,60)=45.07, p<0.001), display type (F(1,60)=87.401, p<0.001), and clutter level (F(1,60)=6.43, p<0.05), with a significant interaction between group and display type (F(1,60)=10.07, p<0.01). Both groups exhibited longer reaction times during the immersive VR task and in high-clutter conditions. Under all conditions, the CVI group showed longer reaction times than controls. Finally, fixation patterns for both groups exhibited a larger search space under high clutter conditions (F(1,54)=14.34, p<0.001), with the CVI group demonstrating larger search spaces than controls (F(1,54)=21.58, p<0.001). These findings highlight visuospatial deficits in CVI individuals, who exhibit slower and less accurate object localization and broader search patterns. Greater reaction time differences between groups in the VR condition suggest that increased immersion may better capture the higher-order visual perceptual challenges faced by this population.