Visual search and target detection during simulated driving in Parkinson’s disease

Abstract

Patients with Parkinson’s Disease (PD) often exhibit difficulties with visual search that may impede their ability to recognize landmarks and cars while driving. The main objective of this study was to investigate visual search performances of both billboards and cars in patients with PD using a driving simulator. A second objective was to examine the role of cognitive functions in performing the visual search task while driving. Nineteen patients with PD (age: 68 ± 8yo, sex (Men/Women): 15/4) and 14 controls (age: 60 ± 11yo, sex: 7/7) first performed a battery of cognitive tests. They then drove in a simulator and were instructed to follow a lead vehicle while searching for billboards with the letter A (stationary target) or red cars (moving target) among other distractors. Accuracy and response times of visual search were the main outcome variables. Standard deviation of lateral position (SDLP) was the secondary outcome. During driving, patients were less accurate in identifying the targets, particularly for the stationary billboards located in the outer periphery. Within the group of patients, significant correlations were found between several measures of cognitive tests and simulator-based visual search accuracy. By contrast, only the score on the MOCA test correlated significantly with visual search accuracy in controls. Findings suggest that patients with PD have impaired visual search for more eccentric stationary targets while driving a simulator, which is likely due to cognitive deficits. Difficulties identifying objects in the outer periphery may have implications for driving safety. Decreased functional field of view under increased cognitive load may have attributed to the difficulties identifying these landmarks. This may impact the ability to identify, anticipate, and respond to important information (e.g., pedestrians, navigation signs, landmarks), especially in complex driving situations (e.g. urban driving or intersections).Future studies should be conducted in a larger sample size to determine whether a visual search task on a driving simulator may predict on-road driving performances.