Abstract
The moment just before an eye movement is a turbulent time for the visual system, and for visual attention in particular. A number of competing theories have been proposed that aim to provide a unifying account of pre-saccadic processes. According to the “remapping” hypothesis, visual information is transferred within retinotopic maps to compensate for eye movements. Remapping is believed to be “predictive” (i.e., starting before the onset of a saccade; Mathôt & Theeuwes, 2010, Exp. Brain Res.) and to last for some time after an eye movement has ended (Mathôt & Theeuwes, in press, Psychol. Sci.). However, there are at least two alternative frameworks, which account for much of the same data in a very different way. Hamker, Zirnsak and Lappe (2008, PloS Comp. Biol.) have proposed a model based on receptive field shifts towards the saccade target. Cavanagh, Hunt, Afraz and Rolfs (2010, Trends. Cogn. Sci.) have suggested that the visual system anticipates which retinal locations will be relevant after a saccade, and that “attention pointers” are shifted towards those locations. This differs from the remapping hypothesis in that visual features, such as color, are not remapped. To resolve the current debate on pre-saccadic processes, we have investigated pre-saccadic attentional effects in more detail and found a number of surprising results. For example, just before the execution of a saccade, we found a small attentional effect in the direction opposite from the saccade target. Although none of the frameworks would a-priori predict this result, a tentative explanation can be offered in terms of shifting attention pointers, or, analogously, predictive remapping of attention. We discuss the implications of this result for existing theories of pre-saccadic processes.
NWO (Netherlands Organization for Scientific Research), grant 463-06-014 to Jan Theeuwes.