Comprehension of reversible sentences in "agrammatism": a meta-analysis

Abstract

The functional source(s) of agrammatic aphasic patients' difficulty comprehending semantically reversible active and passive sentences was investigated in a meta-analysis of published sentence/picture matching data from patients with agrammatic production. The analysis revealed approximately equal distributions of three distinct patterns of performance on active and passive voice sentences relative to what would be expected by chance: both structures comprehended better than chance; both structures comprehended no better than chance; active voice sentences comprehended better than chance, while passive voice sentences were comprehended at levels no better than (or worse than) chance. These results are in conflict with explanations of aphasic sentence comprehension failure in which a single pattern of relative performance on active and passive voice structures is asserted to be characteristic of the comprehension of all agrammatic speakers. They also highlight the difficulty of identifying any single causal factor to account for sentence comprehension failure in patients with "agrammatic" sentence production. Results are interpreted with regard to the role of data from aphasic patients in the testing of hypotheses about the organization of normal language processes.