Auditory-Visual Speech Processing (AVSP) 2010

Hakone, Kanagawa, Japan
September 30-October 3, 2010

Disentangling Unisensory from Fusion Effects in the Attentional Modulation of McGurk Effects: A Bayesian Modeling Study Suggests that Fusion is Attention-Dependent

Jean-Luc Schwartz (1), Kaisa Tiippana (2), Tobias S. Andersen (3)

(1) Gipsa-Lab, Speech & Cognition Department, CNRS-Grenoble University, France
(2) Institute of Behavioural Sciences, University of Helsinki, Finland
(3) Technical University of Denmark, Denmark

The McGurk effect has been shown to be modulated by attention. However, it remains unclear whether attentional effects are due to changes in unisensory processing or in the fusion mechanism. In this paper, we used published experimental data showing that distraction of visual attention weakens the McGurk effect, to fit either the Fuzzy Logical Model of Perception (FLMP) in which the fusion mechanism is fixed, or a variant of it in which the fusion mechanism could be varied depending on attention. The latter model was associated with a larger likelihood when assessed with a Bayesian Model Selection criterion. Our findings suggest that distraction of visual attention affects fusion by decreasing the weight of the visual input.

Index Terms: McGurk effect, attention, FLMP, modeling

Full Paper

Bibliographic reference.  Schwartz, Jean-Luc / Tiippana, Kaisa / Andersen, Tobias S. (2010): "Disentangling unisensory from fusion effects in the attentional modulation of Mcgurk effects: a Bayesian modeling study suggests that fusion is attention-dependent", In AVSP-2010, paper S2-1.