The proposed research tests an alternative account of derivational morphology from a crosslinguistic perspective, using both behavioral and modeling techniques. The three main lines of inquiry involve: 1) connectionist modeling of English derivational morphology: 2) cross-linguistic behavioral comparison of morphological processing in English and in Hebrew, a language with a typologically different morphological system; and 3) modeling acquisition of complex words in English and in Hebrew. The implemented model will test the hypothesis that morphological structure reflects the interaction of semantic and phonological representations, and that a connectionist system can simulate specific properties of morphological processing. Two cross-modal priming experiments will test the hypothesis that priming effects in a nonconcatenative language can be explained by the same principles proposed for English. Finally, connectionist simulations will explore the development of morphological structure, allowing explicit testing of the roles of several factors (e.g. type and token frequency, productivity, semantic transparency, and formal simplicity) in two typologically distinct language systems. Results from the models will be compared to results from behavioral studies of children. This research constitutes a significant step towards a larger goal: a theory of lexical representation and processing that combines research in acquisition, normal processing, and impairment, using both behavioral and computational modeling techniques. The computational models are not only valuable in understanding the precise mechanisms involved in certain aspects of lexical processing, but also generate testable predictions which can be investigated in further behavioral studies. The studies also provide valuable insight into the adequacy of the connectionist framework for investigating the development and skilled processing of language.

Agency
National Institute of Health (NIH)
Institute
National Institute on Deafness and Other Communication Disorders (NIDCD)
Type
Postdoctoral Individual National Research Service Award (F32)
Project #
1F32DC000374-01
Application #
2772554
Study Section
Special Emphasis Panel (ZRG1-CMS (01))
Project Start
1999-09-30
Project End
Budget Start
1999-09-30
Budget End
2000-09-29
Support Year
1
Fiscal Year
1999
Total Cost
Indirect Cost
Name
Carnegie-Mellon University
Department
Psychology
Type
Schools of Arts and Sciences
DUNS #
052184116
City
Pittsburgh
State
PA
Country
United States
Zip Code
15213
Gonnerman, Laura M; Seidenberg, Mark S; Andersen, Elaine S (2007) Graded semantic and phonological similarity effects in priming: evidence for a distributed connectionist approach to morphology. J Exp Psychol Gen 136:323-45
Devlin, Joseph T; Jamison, Helen L; Gonnerman, Laura M et al. (2006) The role of the posterior fusiform gyrus in reading. J Cogn Neurosci 18:911-22