Function of Fixational Instability During Natural Viewing

University Of Rochester, Rochester NY

Investigators

Linked publications & trials

Abstract

PROJECT SUMMARY Humans are not aware that their eyes are always in motion. Even when attending to a single point, fixational eye movements (FEM) continually shift the stimulus on the retina in ways that would be immediately visible had the motion originated from objects in the scene rather than oculomotor activity. It is now clear that FEM are vital for visual sensitivity, fine pattern vision, and acuity. Furthermore, a considerable body of evidence, in part from our NIH-funded research, indicates that this behavior embodies a sensorimotor strategy by which the visual system processes spatial information in the temporal domain. While much has been learned about the monoc- ular functions of FEM, little is known about their consequences for binocular vision. Ocular drifts—the incessant inter-saccadic movements—differ considerably in the two eyes. How does the visual system combine continually changing inputs from independently jittering eyes? Here we focus on FEM consequences for 3D spatial repre- sentations, specifically their role in stereopsis (Aim 1), their decoding mechanisms (Aim 2), and their binocular control (Aim 3). The research strategy consists of assessing FEM effects on the binocular visual input and ex- amining the resulting implications for neural coding, perception, and control. Our driving hypothesis is that the active space-time encoding strategy that emerged in the monocular processing of luminance also applies to pro- cessing binocularly-derived features at later stages of the visual stream. Since this theory yields counter-intuitive hypotheses, each aim builds on a supporting preparatory study that sets the stage for the proposed experiments. Aim 1 builds on the surprising observation that stereopsis is impaired when fixational disparity modulations are selectively eliminated from the visual flow, even in the presence of otherwise normal luminance modulations on the retina. We will explore the causes for this impairment and elucidate FEM contributions. Aim 2 focuses on the mechanisms by which the fixational visual flow is interpreted. Contrary to traditional assumptions, our prelimi- nary evidence indicates that the visual system has access to extraretinal knowledge of ocular drift and uses it to infer spatial relations at high spatiotemporal resolution. Aim 3 examines these ideas in the context of oculomotor control. We provide the first comprehensive high-resolution measurements of head-free binocular FEM in natural real-world tasks and test the hypothesis that eye drifts are actively controlled to encode task-relevant features (e.g., disparity, spatial contrast, etc.). The experiments rely on the combination of (a) binocular measurements of human eye movements with unprecedented accuracy; and (b) highly flexible, binocularly synchronized, gaze- contingent control of retinal stimulation, an approach made possible by our recent instrumentation developments. All experiments are theoretically grounded and all hypotheses supported by new preliminary data. They are, to our knowledge, entirely novel, and confirmation of any of them will have broad implications for understanding the functional principles of the visual system, the computational mechanisms of perception, possible oculomotor contributions to neuro-ophthalmologic disorders, and the development of rehabilitative strategies and prostheses.