The Institute for Mathematical Behavioral Sciences Colloquium Series presents

“Filter Processes in Human Visual Attention: Methods, Measurements, Theory”
with Charlie Chubb, Peng Sun, George Sperling, and Ted Wright, Professors, Department of Cognitive Sciences, UCI

Thursday, October 17
4:00 – 5:00 p.m.
Social Science Plaza A, Room 2112

There is enormously more visual information in the environment than humans (or any other creature) can possibly process. Information is spatially selected by body and eye movements, and further by the distribution of attention to different retinal areas within a single eye fixation. The computational description of the limitations of spatial attention within a fixation is a relatively recent development, previously carried out in part here at UCI (e.g., Gobell et al, VisRes 2004). The selection of information according to its content is called feature attention and also has been very extensively studied. So far, the computational description of feature-attention has been primarily in terms of the priority of different features for eliciting eye movements or directing implicit search processes to spatial areas that contain high-priority features (e.g., Wolfe, Itti, Geisler).

New work: Chubb, Peng, Sperling and Wright are working within a framework that conceptualizes feature-attention as a collection of filters that selectively attenuate the retinal input thereby making it differentially available to subsequent brain processes. Such filters undoubtedly occur at different levels of visual processing. So far, the group has concentrated on the earliest visual processing levels that are controllable by attention. They have developed new, highly efficient measurement procedures that make it possible to precisely measure attention filters, and will demonstrate these measurements in several domains.

For further information, please contact Joanna Kerner, kernerj@uci.edu or 949-824-8651.

 

© UC Irvine School of Social Sciences - 3151 Social Sciences Plaza, Irvine, CA 92697-5100 - 949.824.2766