LESION SPECIFIC DEFICITS IN VISUO MOTOR BEHAVIOR

University Of Rochester, Rochester NY

Investigators

Linked publications & trials

Abstract

Our previous work has using the block-copying task demonstrated unexpected trade-offs between observer eye and head movements and the use of visual memory in ongoing tasks. Understanding the operation of visual memory is crucial in making the transition from an understanding of low level perceptual machinery to understanding of how perception mediates ongoing sensori-motor bahavior. Recently we developed a new paradigm for exploring simple sensori-motor behaviors, in which observers construct wooden models of objects. Initial results confirm the crucial role of fixation that we observed with the simpler copying task, indicating that these memory/eye movement tradeoffs are a ubiquitous feature of normal sensori-motor behavior. During the past year we studied the way tradeoffs between eye fixations and memory change in the context of familiar visual objects, and how complex tasks become automatised". A crucial component of behavioral flexibility is our ability to learn sub-components of complex tasks (automati-sation) in order to perform the task more efficiently with minimal high level control. Initial observations suggest that automatization is revealed in the fixation patterns to the model. As learning progresses, each fixation gives more visual information. This corresponds to the process of "chunking," or consolidation of referents. Although this process has long been hypothesised and is a crucial component of cognitive models, it has proven elusive to study with standard psychological techniques. These experiments now allow us to look at the development of the chunking process. This project dovetails with a major research project at the University of Bielefeld, Germany, that uses the same wooden construction pieces in the development of machine recognition systems for language and vision.