Invited Speaker

Massimo Stella

CogNosco Lab, Department of Computer Science, University of Exeter, UK

http://www.compscicons.net, or http://cognoscoexeter.compscicons.net

Multiplex networks and AI unveil the influence of the mental lexicon

on picture naming and its failures by people struck with aphasia.

Word production from the mental lexicon is a complex phenomenon, entailing multiple layers of conceptual representations. I will introduce in my talk the notion of multiplex networks. A multiplex lexical network is a model that merges semantic, phonological, and taxonomic layers of associations, yielding a coherent mathematical framework with great potential for tasks like word-finding.

I will show in my talk via several studies, hence via quantitative data, that multiplex representation, outperform other models for well-studied problems in psycholinguistics, like picture naming or word access failures. Particular attention will be given to the exploration of the mental lexicon through a picture-naming task performed by people struck by aphasia, or hit by some other form of language impairment.

I will present novel results concerning the resilience of multiplex lexical networks with regard to progressive word-access failure. Emphasis will be given to the added value of multiplex networks for picture naming, or word access by and large. I will finish my talk by summarizing key insights, and by discussing limitations, and directions for the study of the mental lexicon (map, structure), and its possible usage to enhance the electronic dictionaries of tomorrow.