LOCALIZING OBJECTS IN DYNAMIC SCENES

University Of California Berkeley, Berkeley CA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Most visual and visuomotor functions hinge on our visual system's ability to localize objects in dynamic scenes-to determine the positions of objects as we move or as objects in the world move around us. However, dynamic localization requires that the visual system must be able to both assign and update object positions, and it remains unclear which visual areas play these necessary roles and what mechanisms are responsible. Assigning and updating object locations is especially relevant considering that the eye is rarely stabilized and objects are constantly drifting across the retina as they move. Given the sluggish nature of visual processing, how do we perceive the positions of moving objects with clarity and precision? To understand how the visual system assigns and updates object positions, we must approach the task of localization not as a static process, but as a dynamic one. The goal of our proposed experiments is to characterize the neural pathways and mechanisms that determine perceived position in dynamic situations. Our preliminary results suggest that two mechanisms are required to perceive the positions of objects when motion is present on the retina: trailing edge deblurring and edge-shifting. Without these mechanisms, we would perceive the world as a blurry smear, unable to accurately predict the future locations of objects or interact with them as they move around us. We hypothesize that these two independent mechanisms integrate information between early visual areas (V1) and the motion sensitive region (MT+). Our pilot data for the proposed experiments suggest that (1) MT+ carries precise information about object position, supporting the hypothesis that it serves an integral role in object localization, and (2) feedback as well as feed-forward connections between V1 and MT+ may be necessary to assign and update the positions of objects when retinal motion is present. Using functional magnetic resonance imaging (fMRI), transcranial magnetic stimulation (TMS), and psychophysics, we will characterize and isolate the neural locus of the two mechanisms and establish their causal roles in perceiving object position. Dynamic localization is the default situation faced by the visual system, but many neurological disorders produce specific deficits in dynamic visual and visuomotor localization, including optic ataxia, dyslexia, akinetopsia, autism, and schizophrenia, among others. Until we understand how position is determined in the normal brain for images that are constantly moving across the retina, we lack the necessary insight to develop diagnostic tools, predictive markers, and therapies for these impairments.