The Colavita Visual Dominance Effect

Abstract

In this article, we will focus mainly (although not exclusively) on the Colavita effect present in the error data (in line with the majority of published research on this phenomenon). We start by summarizing the basic findings to have emerged from studies of the Colavita visual dominance effect conducted over the past 35 years or so. By now, many different factors have been investigated in order to determine whether they influence the Colavita effect: Here, they are grouped into stimulus-related factors (such as stimulus intensity, stimulus modality, stimulus type, stimulus position, and bimodal stimulus probability) and task/participant-related factors (such as attention, arousal, task/ response demands, and practice). A range of potential explanations for the Colavita effect are evaluated, and all are shown to be lacking. A new account of the Colavita visual dominance effect is therefore proposed, one that is based on the “biased competition” model put forward by Desimone and Duncan (1995; see also Duncan 1996; Peers et al. 2005). Although this model was initially developed in order to provide an explanation for the intramodal competition taking place between multiple visual object representations in both normal participants and clinical patients (suffering from extinction), here we propose that it can be extended to provide a helpful framework in which to understand what may be going on the Colavita visual dominance effect. In particular, we argue that a form of cross-modal biased competition can help to explain why participants respond to the visual stimulus while sometimes failing to respond to the nonvisual stimulus on the bimodal target trials in the Colavita paradigm. More generally, it is our hope that explaining the Colavita visual dominance effect may provide an important step toward understanding the mechanisms underlying multisensory interactions. First, though, we review the various factors that have been hypothesized to influence the Colavita visual dominance effect.