Reconciling opposite neighborhood frequency effects in lexical decision: Evidence from a novel probabilistic model of visual word recognition

Thierry Phénix (1) , Sylviane Valdois (1) , Julien Diard (1)

1 LPNC - Laboratoire de Psychologie et NeuroCognition

A new Bayesian model of visual word recognition is used to simulate neighborhood frequency effects in lexical decision. These effects have been reported as being either facilitatory or inhibitory in behavioral experiments. Our model manages to simulate the apparently contradictory findings. Indeed, studying the dynamic time course of information accumulation in the model shows that effects are facilitatory early, and become inhibitory at later stages. The model provides new insights on the mechanisms at play and their dynamics, leading to better understand the experimental conditions that should yield a fa-cilitatory or an inhibitory neighborhood frequency effect.

Mots clés: Visual word recognition; Lexical decision; Visual attention; Bayesian algorithmic modeling

Domaines: Informatique; Psychologie

Thierry Phénix, Sylviane Valdois, Julien Diard. Reconciling opposite neighborhood frequency effects in lexical decision: Evidence from a novel probabilistic model of visual word recognition. 40th Annual Conference of the Cognitive Science Society (CogSci 2018), Jul 2018, Madison, WI, United States. pp.2238-2243. ⟨hal-01850020⟩