Trajectories of the Human Binocular Fixation Point during Conjugate and Non-conjugate Gaze-shifts

This paper describes the spatial trajectories of the binocular fixation point (the intersection point of the two lines of sight) during gaze-shifts within a horizontal plane of regard. Gaze was voluntarily shifted between pairs of real, continuously visible LED targets that were either iso-vergent at 5-25 deg convergence (conjugate version saccades) or differed in vergence angle (by 5-20 deg) as well as in direction (by 5-60 deg; combined version and vergence). Orientations of both eyes were recorded by phase detection in a homogeneous magnetic field with scleral sensor coils. "Conjugate" saccades showed an outward-looping, curved trajectory as a result of transient divergence, typically associated with horizontal saccades. These outward loops were disproportionately larger for far than for near targets, due to the non-linear relation between vergence and distance. Transient divergence increased moderately in magnitude and duration when basic vergence increased from 5 to 25 deg. As a result, transient saccadic disparities increased in angular magnitude as targets got close. Increasing tonic vergence did not, however, slow down conjugate saccades, in contrast to the previously described dynamic slowing effects of vergence on version during gaze-shift involving simultaneous vergence and version changes. Convergent and divergent non-conjugate gaze-shifts each had characteristic trajectories; outward loops were much reduced in convergent and virtually absent in divergent movements. The saccadic component of non-conjugate gaze-shifts was preceded by a pre-saccadic vergence component in the direction of the imminent gaze-shift; its magnitude increased systematically with the increase in vergence demand and with the decrease in version demand. For both pre-saccadic convergence and divergence, this pre-saccadic part of the trajectory tended to follow an iso-direction line through the target of origin; directional change did not start until the saccade began. This suggests that for targets that differ in direction as well as distance, control of the vergence and version components of the gaze-shift can be dissociated to some degree. This seems to argue against models of binocular oculomotor control which assume that each eye responds primarily to its own target, and suggests rather that target vergence and target direction may be processed and responded to separately by ocular vergence and version, with a strong interaction between the two oculomotor activities whenever they occur at the same time.