Compound naming in aphasia: effects of complexity, part of speech, and semantic transparency

Abstract

This study examines spoken naming of German nominal compounds in four individuals with chronic aphasia in order to test current theories about the production of morphologically complex nouns. We investigated the impact of morphological complexity, grammatical class of the modifier, and of the constituents' semantic transparency on spoken naming accuracies, error types, and position of preserved constituents in erroneous responses. The results show that most participants were specifically impaired with compounds compared to long matched simple nouns. Furthermore, the grammatical class of the modifier affected the qualitative error pattern for three participants with agrammatic aphasia who were particularly impaired in retrieving the verbal modifier constituent as compared to the nominal head. In contrast, for noun–noun compounds no head superiority effect was obtained. Finally, semantic transparency did not affect overall naming accuracies, but the occurrence of constituent errors, i.e. erroneous responses overlapping with the target in one constituent. More constituent errors were observed for transparent than opaque targets. The results are compatible with decompositional and dual route accounts.