The goal of the project is to study the build up of visual memory for objects in scenes when saccadic eye movements are required to scan the scene. A controversial issue has been how much visual information can be remembered across separate fixations. Prior studies may have underestimated memory either because the tasks imposing competing cognitive requirements or because the strong semantic content of the scenes may have overridden memory of components. The proposed tasks will reduce extra cognitive load and use displays of realistic scenes containing large numbers of objects with weak semantic links to the background (e.g. objects on shelves).
Specific aims are: (1) to evaluate memory capacity for objects in structured 3-D scenes, relative to objects on homogenous backgrounds while the displays are inspected for various durations either with or without use of saccades; (2) to determine which, if any, aspects of the saccadic pattern were most predictive of memory performance; (3) to study long-term build-up of visual memory by testing the same scenes in a different task (visual search); (4) to extend previous work on saccadic localization of target shapes by finding out whether saccades land at the center of target objects, or whether other landing positions are adopted in the context of this more realistic tasks. Overall, this project should improve understanding of the visual and motor contributions to scene memory.
| Melcher, D; Kowler, E (2001) Visual scene memory and the guidance of saccadic eye movements. Vision Res 41:3597-611 |
