Abstract
According to the zoom-lens model (Eriksen & Yeh, 1985), the focus of attention (FOA) is adjusted in spatial extent to match the visual angle subtended by attended object(s), with a trade-off between the size of the FOA and the concentration of attention: the larger the FOA, the greater the diffusion of attention and the weaker its effects. Consistent with this trade-off, when informative spatial cues (e.g., squares) of different sizes are presented prior to target onset, RT is faster for small vs. large cues (Castiello & Umilta, 1992; Turatto et al., 2000). It remains unclear, however, whether the size of the FOA is necessarily determined by the physical extent of the attended stimulus, and whether the relationship between the size of the FOA and the concentration of attention is fixed. We investigated these issues by presenting spatial cues that were perceptually different in size despite being physically identical. When the cues were superimposed on a simple line version of the Ponzo illusion, RT was reliably longer to targets appearing within the perceptually larger cue. This effect was replicated when cues were superimposed on a realistic, natural scene that induced the Ponzo illusion. This finding suggests either (a) that the size of the FOA is modulated by the perceived size of the attended object, or (b) that the size of the FOA is locked to the physical size of the object, but the concentration of attention is influenced by the perceived size of the object. Both possible accounts suggest specific ways in which current models of attention will need to be refined.
Support for this research was provided by an NSERC post-doctoral fellowship to LNJ, a National Institute on Aging Training Fellowship (AG027668-01) to LG, and NIH grant R01-DA13165 to SY.