| |||||||||||||||
![]() | ||||
| PI Name: | STEVEN YANTIS | |||
| DEPARTMENT OF PSYCHOLOGY JOHNS HOPKINS UNIVERSITY BALTIMORE, MD | ||||
| Project Title: | COGNITIVE NEUROSCIENCE OF VISION | |||
Abstract: Research in my laboratory addresses the mechanisms of intermediate-level vision. We are concerned with the perceptual issues surrounding the creation, selection, and maintenence of visual object representations. These mechanisms include perceptual organization, visual attention, and certain aspects of object recognition. The methods we use to investigate these mechanisms include the measurement of response time and accuracy using the techniques of psychophysics and experimental psychology, and measurements of human brain activity patterns using functional magnetic resonance imaging (fMRI) in the F. M. Kirby Research Center for Functional Brain Imaging.
Several current projects are concerned with the mechanisms of
object-based visual selection. Although it is clear that visual attention
can be directed overtly via eye movements or covertly via deployments of
attention to locations in space, much evidence has accumulated
that visual attention can also operate on representations of perceptual
objects as distinct from their locations. Our commerce with the world
involves reaching for objects, not spatial locations, so it is hardly
suprising that the visual system should be optimized for selecting
relevant or salient objects. Our experiments have examined the extent to
which the deployment of attention to part of an object causes other parts
of that object to also be selected in preference to other locations that
are equally distant from the attended location.
Links:
Classic Articles in Visual Perception
5 subvolumes are available in two formats: ANALYZE (7.5) and MATLAB formats.
Analyze should be able to read in the data irrespective of computer platform. The MATLAB data was generated on Linux but can be read in by other *IX systems.
Explanation of MATLAB format
Hand segmentation labels
This data is available with registered usernames and passwords. If you have already registered, and your registration has been confirmed, you may proceed to the download page.
Supported by