When: 12 March 2021, 10h00
Where: Visioconference (Zoom link here)
Speaker: Dr. Julien Diard (website)
Institution: Laboratoire de Psychologie et NeuroCognition CNRS UMR 5105
Title: Visual attention matters during word recognition: A Bayesian modeling approach
Abstract: It is striking that visual attention, the process by which attentional resources are allocated in the visual field so as to locally enhance visual perception, is a pervasive component of models of eye movements in reading, but is seldom considered in models of isolated word recognition. We describe BRAID, a new Bayesian word Recognition model with Attention, Interference and Dynamics. As most of its predecessors, BRAID incorporates three sensory, perceptual and orthographic knowledge layers together with a lexical membership submodel. Its originality resides in also including three mechanisms that modulate letter identification within strings: an acuity gradient, lateral interference and visual attention. We show that BRAID can account not only for benchmark effects, such as word frequency, neighborhood frequency, context familiarity and transposed letter priming effects but, further, for more challenging behavioral effects, such as the optimal viewing position effect, the word length effect in lexical decision or the interaction of effects of crowding and frequency in word recognition. We show that visual attention modulates these latter effects, mimicking patterns reported in impaired readers.
Short bio: Since 2005, I hold a full time research position (Chargé de Recherche) at CNRS, and I work at the LPNC, the Psychology and NeuroCognition Laboratory, in Grenoble. In 2015, I have defended my HDR in Computer Science. In my research, I am interested in probabilistic cognitive modeling, to study cognitive functions on the one hand, such as reading and word recognition, or speech perception and production, but also to study the models themselves on the other hand, and how probabilistic information is represented and propagated in complex, hierarchical models. I am the deputy director of the Pôle Grenoble Cognition (CNRS FR3381, UGA) and a deputy director of the “Cognitive Science” Master’s degree (Phelma, G-INP).
Keywords: Bayesian modeling; word recognition; lexical decision; visual attention; optimal viewing position; crowding effect
