abstract Kristof Strijkers

Kristof Strijkers (Université d'Aix-Marseille, LPC, CNRS, France)

Using time-sensitive neurophysiological measures to constrain theories of language production

The goal of this talk is to demonstrate how fine-grained temporal knowledge about the various mental operations necessary to produce words can be exploited to address complex cognitive questions underlying language production. First, I will give a brief overview of where the field stands in terms of the chronometry of speech production, paying special attention to applications of neurophysiological measures to overt and immediate naming paradigms. Then I will talk about two recent studies using such temporal estimates to test different theoretical predictions of language production models. In a first study, we investigated the temporal dynamics underlying lexical access in word production through MEG recordings during picture naming. More concretely, we contrasted the spatiotemporal signature of the lexical frequency effect, as a proxy of the onset of word retrieval, with that of the type of articulator movements required to utter the first phoneme of a picture's name (lip-movement versus tongue-movement; e.g., monkey vs. donkey), as an uncorrelated upper limit of phonological encoding. By comparing a lexical load manipulation with a phonological/phonemic distributional manipulation in both space and time, we sought to shed light on the cerebral dynamics between these two core operations in speech production. In a second study, we employed the time course and ERP morphology elicited by the lexical frequency effect in order to explore how conceptual knowledge triggers lexical representations in the course of speech. In particular, we compared the lexical frequency effect registered with ERPs in a task where participants had the conscious intention to engage in a speech act (picture naming) versus a task where there was no a-priori intention to verbalize the same depicted input (picture categorization). Results of both these studies cannot be accounted for by any of the formalized accounts of language production and necessary extensions or modifications of current models will be discussed. All together, this research highlights the importance of investigating overt language production with precise temporal measures.