Department: Psychology
Professor Sarah Shomstein
Project Description: How does the human visual system sort through the massive amounts of sensory input, which it samples almost continuously, to arrive at a coherent perception of a scene? This process of searching through the
environment for information is a ubiquitous component of sensory processing and it reflects a remarkable ability of the perceptual system to dynamically select information that is relevant for the current goal of the organism. Such perceptual selectivity, referred to as attention, is central to cognition. Since what we consciously perceive will, ultimately, depend on where we direct our attention, understanding attentional mechanism is an important first step toward revealing the neural mechanisms that support conscious awareness. Over the past four decades, researchers acquired enough explanatory power to predict behavior, in simple visual environments, based on selection of spatial locations, features, and objects. Unfortunately, predictive ability of models based exclusively on physical attributes of the scene (space, objects, features) fail with increased scene complexity (e.g., scenes with multiple simple objects, or real-world scenes). Such failure in predicting how attention is allocated in a more complex scene, can partially be attributed to a lack of understanding of how higher-level properties of objects, and scenes, constrain attentional selection.
The proposed research program aims to test a set of novel predictions regarding the influence of high-level properties of the scene to attentional selection. Our goal is to rigorously test the hypothesis that task-irrelevant semantic information constrains attentional selection by directly acting on space- and object-based representations: task-irrelevant semantically related objects are more likely to be attended. We ask a fundamental question whether semantic information intrudes itself on attentional control even when it is not directly relevant to the current goals of the observer. Predictions will be tested in real world scenes, and by utilizing real world objects. Both, behavioral profile (with the use of various psychophysics and eye-movement techniques) as well as the neural underpinnings of this mechanism (by employing neuroimaging techniques), will be examined. The proposed research program not only enhances discovery but also dovetails with numerous activities that promote teaching, training, and learning within academia, as well as the general public.
Duties: Meet and greet research participants.
Help with day-to-day running of a research laboratory.
Looking for enthusiastic students who are interested in elucidating links between brain and behavior. Enthusiasm, curiosity, independence, and willingness to work hard are required.
Time Commitment/Credits: 7-9 hours per week (average); 3 credits
To Apply: Submit Cover Letter/Resume to shom@gwu.edu
Skip to content
University Honors Program Blog
Intellectual Omnivores Unite
