The link between behaviour and neurophysiology in lexical processing : An information-theoretical approach

old_uid2916
titleThe link between behaviour and neurophysiology in lexical processing : An information-theoretical approach
start_date2007/05/25
schedule11h-12h
onlineno
location_infobt. 9, rdj/ouest-08, salle de cours
summaryI will present a collection of results from the visual lexical decision task that indicate that information theoretical measures are a useful descriptive tool to investigate the effects of semantic and morphological neighborhoods in word recognition. I will continue by introducing the Bayesian Information Theoretical (BIT) model of lexical processing. This model is intended to provide a bridge between the above mentioned effects on lexical decision, and the underlying properties of neural assemblies representing words and morphemes. The model demonstrates how the information theoretical measures of lexical neighborhoods can be directly related to the properties of the neural assemblies that represent those words. This link is addressed by a combination of tools taken from information theory, Bayesian inference, and computational linguistics. Finally, I will outline the basis of my future research, exploring the implications of the BIT model for EEG and fMRI experiments, detailing the research that I will be developing at the LPC.
responsiblesPélissier, Grainger