Production and processing of time reference in Moroccan Arabic agrammatism: evidence from a semitic language

Abstract

ABSTRACT Background Several studies on agrammatic aphasia have suggested a selective deficit in the production and processing of temporal reference to the past. To account for this pattern of performance, the PAst DIscourse LIinking Hypothesis was formulated (PADILIH). The PADILIH assumes an impairment of discourse-linking operations in agrammatic aphasia, which manifests in greater difficulty with reference to the past than to the present and future, which do not rely on D-linking. Aims The first goal of the present study was to examine the validity of the PADILIH, drawing evidence from a morphologically rich Semitic language; Moroccan Arabic (MA). The second goal was to examine whether different experimental manipulations will yield different patterns of performance; hence, suggesting an interaction between task demands and accuracy. Methods & Procedures Five MA agrammatic participants and five non-brain-damaged participants (NBDs) took part in the present study. Production, processing, and comprehension of temporal reference were tested using a number of experimental tasks. For production, a picture description, and a transformational sentence completion were used. For processing, a grammaticality-judgment task was used. For comprehension, a sentence-picture matching task was used. Outcomes & Results NBDs performed at ceiling across all the tasks used. For agrammatic participants, different tasks revealed different patterns of performance. Both findings of the transformational sentence completion and the grammaticality-judgment tasks support the predictions of the PADILIH, in the sense that, at the group level, past reference was more compromised than present and future reference. Picture description data, however, suggested a tendency to supply more verbs in the past form. Comprehension data suggested asymmetries between the present and the future. Conclusion Sentence completion and grammaticality-judgment data provide cross-linguistic support for the PADILIH. To explain the findings of the picture description task, it was proposed that the high number of verbs supplied in the past reflects an economy-of-effort strategy whereby patients omitted present and future markers to avoid cognitive overload. To account for the findings of the comprehension task, it was suggested that the fact that the action denoted by the verb is not overtly displayed in pictures referring to past and future contexts, but rather inferred from the contextual clues, might have given rise to a future deficit as well. The theoretical implications of these findings are discussed.