This is a course on sentence processing in Hungarian, involving experimental data and processing theories, intended for students of language, cognition, psycholinguistics, as well as students of Hungarian. One essential issue in sentence processing research during the last 20 years has been the attempts to see how models developed for Indo-European languages, and especially English do fit processing in other languages. There were several attempts to clarify the universality issue here from the competition model of Bates and MacWhinney through the Canonical Forms model of Slobin and Bever to parameter setting. The course will present data on Hungarian, a highly inflected agglutinative language that support the notion that some languages use more analytic while others more holistic strategies. Data on simple sentence interpretation, reading of complex sentences, lexical decision, cross-modal priming as a function of morphology all support a mixed model for Hungarian where primary decisions would be made on the basis of localistic cues, but a second pass would still consider more general sentence frames. The course will discuss some of the implications of these data regarding how universal resources are differently combined together to deal with different languages.
Days and Time: Wednesday, 2:30-5:00.