DYNAMIC PROPERTIES OF ORIENTATION DISCRIMINATION

Columbia University Health Sciences, New York NY

Investigators

Linked publications & trials

Abstract

The overall objective of this research proposal is to provide a more comprehensive understanding of fine orientation discrimination--the ability to reliably distinguish subtle angular differences. Two specific aims are proposed to accomplish this objective. The first aim entails developing a physiologically plausible computational model for orientation discrimination and orientation learning. The model posits a single neural principle, namely, the reduction in neural firing, as the cause of both long-term orientation learning and short- term tilt aftereffects--two seemingly unrelated perceptual phenomena. The model's prediction that tilt aftereffects should be significantly stronger after orientation learning will be psychophysically tested. The second aim is to establish whether a single neural algorithm, such as the one in the proposed model, is responsible for simultaneously detecting an orientational difference (i.e., same vs different) and identifying the sign of that difference (i.e., clockwise vs counter-clockwise). Accordingly, the short-term dynamics of orientation discrimination will be investigated in an experiment designed to determine whether an observer's ability to detect and identify orientational changes is differentially affected by stimulus duration. Together, these two aims will serve as the foundation for future attempts to understand the dynamic properties of other visual discriminations, such as motion and depth discrimination.