Eyes on the Road: Eye Movements and the Visual Control of Locomotion in Curve Driving

The thesis presents three studies on the visual control of locomotion. Each investigates the visual behaviour of human subjects driving a car. The focus is on the basic visual behaviours involved in the task of negotiating bends in a road car, at normal road speeds. The aim was to investigate the properties of visual behaviour relevant to the two predominant classes of visual steering models in the lite¬rature: tangent point models, and future path models.

Study I is the first on-road study where the visual projection of the future path in the field of view is modeled explicitly, enabling a re-evaluation of the generality of the tangent point hypothesis and its methodological grounds based on simple area of interest (AOI) measures incapable of producing decisive evidence, prompting the need for complementary data. Empirically and theoretically, the future path can be considered a potential gaze target that is as equally valid as the tangent point. Study II is the first on-road study to demonstrate optokinetic nystagmus (periodic, slow eye movements) in curve-driving. Combining eye-movement data with vehicle telemetry, we were able to show that the fixations lose their pursuit-like appearance, and are stable relative to real allocentric locations. Study III used these optokinetic pursuit movements as a means to empirically contrast the tangent point and future path models. We found that 1) the gaze position was typically above the tangent point, and displays a large horizontal variation in relation to it, being spread into the far zone along the future path, 2) the optokinetic pursuit had a horizontal component in the direction opposite to the bend and 3) the magnitude of the horizontal component of the pursuit movements was approximately equal to one half of the vehicle yaw rate. All three observations are what one would predict given the hypothesis that the drivers were targeting fixed target points on the future path beyond the tangent point and tracking them with pursuit eye movements, but difficult to reconcile with a tangent point strategy.

This pattern of results in the three studies is consistent with the drivers targeting points on the future path instead of, or in addition to, the tangent point. This challenges the generality of the tangent point hypothesis as an account of where we look when we steer.