Purchase this article with an account.
Sarah Tower-Richardi, Julie Golomb, Nancy Kanwisher; Binding of location and color in retinotopic, not spatiotopic, coordinates. Journal of Vision 2011;11(11):521. doi: https://doi.org/10.1167/11.11.521.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Our eyes are in constant motion, and therefore, the visual input projected onto the retina is continually changing. This presents a major challenge to our visual system: with each eye movement, objects of behavioral importance stay in the same world-centered (spatiotopic) locations, while the locations relative to our eyes (retinotopic) change. Past research has shown that when attending to a spatiotopic location, attention briefly remains in retinotopic coordinates after an eye movement before transferring to spatiotopic coordinates (Golomb, Chun, & Mazer, 2008 J.Neurosci.). Object location is thought to be so important that it is automatically encoded even when trying to remember a non-related aspect of a stimulus (e.g., its identity or color: Tsal & Lavie, 1993, JEP:HPP). Our study addresses the question: Do these task-irrelevant location benefits also persist in retinotopic and/or spatiotopic coordinates? Participants fixated on a fixation dot while a colored square was presented in the periphery; the task was to remember the exact color of the square over a short (∼1 sec) delay. During the delay, participants made a guided saccade to a different location. A second colored square was then presented, and participants had to report whether or not it was the same exact color. The second square appeared in one of three critical locations: the same spatiotopic location as the sample, the retinotopic location, or a control location. Despite the fact that location was irrelevant to the task, participants were faster and more accurate at judging the color when the test square appeared in the same retinotopic location as the sample square. On the other hand, participants were no more accurate for the spatiotopic compared to the control location. This demonstrates that task-irrelevant location benefits are encoded primarily in retinotopic coordinates, suggesting that feature-location binding also occurs in native retinotopic space.
This PDF is available to Subscribers Only