[Derivational morphology and lexical access: prefixed and pseudo-prefixed verbs]

The subjects of this research is the role played by the morphological structure in the process of access to the lexicon. Although a lot of research has been conducted on this topic, experimental data on this role remains inconsistent. To appreciate the issue, the task of lexical decision is used with verbal material constituted by prefixed or pseudo-prefixed verbs as well as visually introduced pseudo-verbs. The nature of the radical (verb, noun or verb and noun) constitutes a specific factor yielding three types of prefixed structure (prefix-radical verb, prefix-radical noun, prefix-radical potentially noun or verb). Two types of pseudo-prefixed structures are studied in function of the nature of the pseudo-radical (existing or imaginary word). Subjects are submitted to two tests: a lexical decision test for the complete set of verbal material; a second test (to be taken subsequently) in which they are asked to check from a list the words they believe they have seen on the screen in the first test. Results indicate an effective morphological structure and that the processing of prefixed verbs is done according to the nature of the radical. Verbs with radicals that could be verbs and nouns are processed faster than those for which the radical is a verb. The second test also tends to indicate that the morphological structure is taken into account, and especially that the radical is activated at a certain point in the processing (certain radicals, if presented separately, are checked as being present in the first test where only the prefixed items containing them were present). Results obtained with pseudo-verbs show an important role played by the prefix in the identification of morphologically complex items. These results only match partially the Taft & Forster Model (1975). Illustrating the role played by the nature of the radical allows, to a certain degree, reconciling contradictory results in the literature. In fact, it is demonstrated that the different results from various experiences can be explained if this factor is taken into account. The number of prefixed items of each type (defined by the nature of the radical) can thus explain why certain studies reflect differences between morphologically simple and complex words and other discard them.