Change blindness as a tool for investigating compensation in hemianopia

Abstract

Hemianopic visual field loss is blindness or reduction in one half of the visual field caused by damage to the visual cortex or visual pathways. Despite these significant impairments, there is evidence that individuals with hemianopia appear to compensate to varying degrees in everyday functional activities such as driving. Using a driving-related task based on the change blindness paradigm, this research investigated compensation in hemianopic field loss. Thirty-one cases with hemianopic and quadrantanopic visual field loss and thirty-one matched controls participated in the study. Of interest was whether individuals with hemianopia could accurately respond to targets across the blind and seeing regions of visual space. The focus here was to describe the characteristics of scanpaths used by individuals with hemianopic field loss, and it was expected that successful performance on the change blindness task would be associated with altered scanpaths, suggesting compensation. In addition, the study investigated the relationship between performance on the change blindness task and cognitive and vision tests commonly used in driving research and assessment. The general approach was to compare the characteristics of individuals with hemianopia who performed well with those who did not; and ultimately to identify the variables that best predicted performance on the change blindness task. Three key findings emerged. Firstly, it was found that although the cases as a group performed more poorly than controls overall, a subset of the hemianopic group were able to accurately identify changing targets in both their blind and seeing regions. Thus, the capacity to respond to targets in areas with no visual function suggests that this subset of cases were able to compensate well for their visual field defect on this task. Consistent with previous research, the study found that as a group, the hemianopic cases searched the scenes differently to controls. In general, the search patterns of hemianopic cases were characterised by an increased number of fixations with smaller saccade amplitudes. However, further analysis revealed different patterns of scanpaths amongst the hemianopic cases, not previously reported in the literature. Those who compensated for their field loss (fewer errors) searched the scenes differently to those who did not. Specifically, those who compensated made fewer fixations and had longer saccade amplitudes, closely matching those of the controls; however the spatial scanpaths used to reach the target were markedly different to the controls. In contrast, the cases who performed poorly on the task made more fixations with smaller saccades relative to controls, but searched the scenes using similar spatial paths, suggesting that spatial scanpath is a key mechanism underpinning compensation on this task. Performance on the change blindness task was best explained by age and selected cognitive assessments, including Trails B and the Motor Free Visual Perception Test (Visual Closure Subtest), whereas measures of visual function, including the extent of field loss, did not predict performance well. These results suggest that age and cognitive function were more predictive of the ability to compensate for visual field loss on the change blindness task than measures of visual function. This research represents the first attempt to link attentional processing with eye movements and compensation for hemianopic field loss in naturalistic driving scenes. Outcomes of this research provide new evidence describing the characteristics of scanpaths associated with successful compensatory performance in hemianopia. The findings highlight the usefulness of the change blindness task for discriminating those individuals who were able to successfully compensate. Further research is recommended to explore the utility of the driving-related change blindness task as a suitable screening assessment for visual fitness to drive with hemianopic field loss.