Decomposing morphologically complex words in a nonlinear morphology.

Abstract

Most Hebrew words are composed of 2 intertwined morphemes: a triconsonantal root and a phonological word pattern. Previous research with conjugated verb forms has shown consistent priming from the verbal patterns, suggesting that verbal forms are automatically parsed by native speakers into their morphemic constituents. The authors investigated the decomposition process, focusing on the structural properties of verbal forms that are perceived and extracted during word recognition. The manipulations consisted of using verbal forms derived from "weak" roots that have one consonant missing in some of the forms. The results demonstrated that if 1 consonant is missing, the parsing system collapses, and there is no evidence for morphological priming. In contrast, when a random consonant is inserted into the weak form, the verbal-pattern priming re-emerges. This outcome suggests that the constraint imposed on the decomposition process is primarily structural and abstract. Moreover, the all-or-none pattern of results is characteristic of rule-based behavior and not of simple correlational systems.