TASK DEMANDS AND SENTENCE COMPREHENSION IN PATIENTS WITH DEMENTIA OF THE ALZHEIMERS TYPE

Patients with dementia of the Alzheimer's type (DAT) and matched norma l controls were given three tests of syntactic comprehension in which nonlinguistic visual and memory task demands were varied. In all tasks , subjects were presented spoken semantically reversible sentences wit h a variety of syntactic structures and required to match the sentence to a picture. In the first experiment, subjects matched the spoken se ntence to one of two pictures that appeared either before or immediate ly following the presentation of the sentence. The target picture depi cted the spoken sentence correctly and the foil depicted the reversed thematic roles to those in the sentence (i.e., it was a syntactic foil ). The second experiment employed a sentence video-verification task i n which subjects were required to determine if the spoken sentence mat ched a videotaped depiction of the action in the sentence or a syntact ic foil. In the third experiment, in different conditions, subjects we re required to determine whether the spoken sentence matched a single picture or to choose the picture that matched the sentence from an arr ay of two or three pictures. In this experiment, both lexical and synt actic foils were used. In all tasks, DAT patients were affected by the number of propositions in the presented sentence, but not by the synt actic complexity of the sentence. Control subjects also were unaffecte d by the syntactic complexity of the sentence; the number-of-propositi on effect was present in some experiments in the control population. C omparison of performance across the one-, two-, and three-picture vers ions of the task showed that the magnitude of the effect of number of propositions increased as the number of pictures in the array increase d. In addition, analysis of the data from each of the tasks separately showed that the effect of number of propositions only occurred when s ubjects were attempting to match the target to a syntactic foil (one-p icture version) or to choose between the target and a syntactic foil ( two-and three-picture versions). The results support the view that pat ients with DAT do not have disturbances affecting syntactic processing . In addition, they suggest that the effect of number of propositions arises at a stage of analysis that is partially separate from assignin g sentence meaning, such as in holding a representation of the sentenc e in memory until the pictures can be analyzed and encoded and/or in c omparing the results of the picture analysis with a stored representat ion of the sentence meaning. (C) 1998 Academic Press.