Brain and Language 72, 343–374 (2000) doi:10.1006/brln.2000.2297, available online at http://www.idealibrary.com on Interhemispheric Effects of Simulated Lesions in a Neural Model of Single-Word Reading Yuri Shkuro, Mark Glezer, and James A. Reggia Departments of Computer Science and Neurology and Institute for Advanced Computer Studies, University of Maryland A neural model consisting of paired cerebral hemispheric regions interacting via homotopic callosal connections was trained to generate pronunciations for 50 mono- syllabic words. Lateralization of this task occurred readily when different underlying cortical asymmetries were present. Following simulated focal cortical lesions of systematically varied sizes, acute changes in the distribution of cortical activation were found to be most consistent with experimental data when interhemispheric interactions were assumed to be excitatory. During subsequent recovery, the contri- bution of the unlesioned hemispheric region to performance improvement was a function of both the amount of preexisting lateralization and the side and size of the lesion. These results are discussed in the context of unresolved issues concerning the mechanisms underlying language lateralization, the nature of interhemispheric interactions, and the role of the nondominant hemisphere in recovery from adult aphasia.  2000 Academic Press Key Words: artificial neural networks; computer model; hemispheric interactions; corpus callosum; lateralization; cortical lesions. INTRODUCTION Current understanding of the basic physiological mechanisms that underlie lateralization is quite limited. For example, it is unclear which of many known hemispheric asymmetries might be causally contributing to language lateralization and what role interhemispheric communication might play in the lateralization process (Hellige, 1993; Springer & Deutsch, 1993). While future experimental studies can be expected to clarify these issues, computa- tional models may also prove useful by complementing and even guiding such future empirical work. Artificial neural networks (neural models) have been widely used during the last decade to examine many other issues in This work was supported by NINDS Award NS35460. Address correspondence and reprint requests to James Reggia at Department of Computer Science, University of Maryland, College Park, Maryland 20742. E-mail: {merlin,glezer, reggia}@cs.umd.edu. 343 0093-934X/00 $35.00 Copyright  2000 by Academic Press All rights of reproduction in any form reserved.