The general aim of this study is to clarify the sentence-processing deficit in agrammatic Broca's aphasics, and thereby to contribute to an understanding of sentence-processing mechanisms more generally. The starting point for this research is a recent finding that despite their 'asyntactic' language comprenhension, a number of agrammatic Broca's aphasics were able to accurately judge grammaticality of English sentences embodying a wide range of syntactic deformations. Furthermore, a distinctive pattern of performance emerged among these subjects: they were able to assess certain constructions but not others. No such pattern was found in the performance of normals. The hypothesis that guides the proposed research is that this pattern reveals the limits of certain intact sentence processing mechanisms whose function appears to involve the retrieval of constituent structure but little or nothing else. The investigation will focus on further analysis of this pattern, and on the surprising disparity between the syntactic abilities that judgment tasks have revealed in these agrammatic subjects and the apparent inability on the part of these same subjects to make use of such abilities for the purposes of comprehension. These issues are explored through further development of the grammaticality judgment paradigm; exploration of the pattern of performance on this task within other aphasic populations, with particular concern for the issue of whether the pattern that has emerged is unique to agrammatism; close tracking of the relationship between sentence comprehension and judgments of grammaticality; cross-linguistic studies to determine whether agrammatic speakers of other languages demonstrate the same pattern observed in the English-speaking aphasic subjects; exploration of the possibility that the task may be able to chart quite accurately the course of recovery from aphasia; examination of the role of prosodic information in the grammaticality judgment paradigm; and, finally, a series of studies on normals which may provide crucial evidence about the source of the pattern and the nature of the agrammatic comprehension deficit. What seems particularly promising in this regard is the attempt to experimentally elicit the agrammatic pattern in normal subjects by embedding the grammaticality judgment test in a dual task paradigm.
