DIFFERENTIAL INTERHEMISPHERIC TRANSFER
FOR ABSTRACT AND SPECIFIC VISUAL-FORM INFORMATION

Christopher D. Nicholas & Chad J. Marsolek
University of Arizona

(Presented at the 3rd Annual Meeting of the Cognitive Neuroscience Society, San Fransisco, CA, April 1996)

Abstract

When subjects compare two visually presented letters, interhemispheric transfer of visual information must take place when the letters are displayed briefly in different visual fields (i.e., across-hemispheres [AH]), but not when they are displayed briefly in the same visual field (i.e., within-hemisphere [WH]). In an abstract comparison task, subjects decide whether pairs correspond to the same letter (e.g., s/S, a/A) or not (e.g., s/A, a/S) and perform more accurately in AH than in WH trials. In a specific comparison task, subjects decide whether pairs are physically the same (e.g., s/s, a/a) or not (e.g., s/S, a/A) and perform more accurately in WH than in AH trials. One interpretation of this pattern of results from other laboratories is that AH advantages are found for complex tasks because the benefits of distributing component processes (e.g., visual analysis and name comparison) across hemispheres outweigh the costs of transfer, whereas WH advantages are found for simple tasks because the costs of transfer outweigh any benefits of distributing a simple process (e.g., visual comparison). However, we suggest that abstract visual-form (AVF) and specific visual-form (SVF) subsystems underlie the two tasks, respectively, and are differentially affected by interhemispheric transfer. Tasks requiring an AVF subsystem should produce AH advantages, because this subsystem processes relatively invariant features of letters which may transfer effectively. If so, the AH advantage in the abstract task should be greater for similar- (e.g., s/S) than for dissimilar-case letters (e.g., a/A), because greater amounts of relatively invariant information per letter are found in the former than in the latter. Tasks requiring an SVF subsystem should produce WH advantages, because this subsystem processes fine-grained distinctive information which may not transfer effectively. If so, the WH advantage in the specific task should be greater for similar- than for dissimilar-case letters, because finer-grained distinctive information is needed to distinguish the former than the latter. Recent results support these predictions.

The entire poster is available in HTML format.