Abstract
Visual attention is considered to be a selection process that favors some aspects of the visual input over others. A multitude of attentional effects were found for both neural and behavioral responses, however, the nature of the attentional mechanism is an unsettled issue. Models of attention typically attempt to explain attentional effects by some top-down mechanism. There are indeed findings showing that higher visual areas in the visual hierarchy are affected by attentional modulations prior to lower ones, however, these modulations were observed while participants (whether primates or human) practiced sustained (volitional, endogenous) attention. Yet, attention may also be attracted involuntarily by events or stimuli in the visual field (i.e., reflexive, exogenous attention). Here we propose a feed-forward model that is based on the conception of attention as an attraction field: The allocation of attention to a location attracts (shifts) the centers of receptive fields towards this location. We show that the influence of this attraction mechanism propagates up the visual hierarchy, and can serve as a simple unifying framework to explain a diverse range of spatial as well as temporal attentional effects, including gain enhancement, enhanced contrast sensitivity, enhanced spatial resolution, prolonged temporal integration, prolonged perceived duration, prior onset and degraded temporal resolution. Additionally, the model successfully explains the Mexican-hat profile of attention and distortions at the focus of attention. Thus, this model offers a simple mechanism that can explain reflexive attention linking physiological measurements at the unit level with psychophysical observations of both the spatial and temporal domains of perception.
Meeting abstract presented at VSS 2013