Normal and deviant lexical processing: Reply to Dell and O'Seaghdha (1991).

Abstract

In their comment, Dell and O'Seaghdha (1991) adduced any effect on phonological probes for semantic alternatives to the activation of these probes in the lexical network. We argue that that interpretation is false and, in addition, that the model still cannot account for our data. Further­ more, and different from Dell and O'Seaghda, we adduce semantic rebound to the lemma level, where it is so substantial that it should have shown up in our data. Finally, we question the function of feedback in a lexical network (other than eliciting speech errors) and discuss Dell's (1988) notion of a unified production-comprehension system. Until recently, models of lexical access in speech production were almost exclusively based on speech error data. This is true both for the modular two-stage models and for the interactive connectionist models of lexical access. Both kinds of models were initially designed to account for the distributions of natu­ rally observed or experimentall y elicited speech errors. From the start, however, they were conceived as process models of normal speech production. Therefore, the ultimate test of such models cannot lie in their account of infrequent derailments of the process. Rather, the proof of their efficacy should be sought in the account of the normal process itself. An exclusively errorbased approach to lexical access in speech production is as ill-conceived as an exclusively illusion-based approach in vision research. One should, of course, hope that an ultimate theory of the normal process also has the potential of explaining ob­ served error distributions (or visual illusions, for that matter), but it should not be one's main concern.