New Insights into Visuospatial Attention: The Spotlight Metaphor meets fMRI

Covert spatial orienting enhances perceptual capabilities in the part of the visual field one attends. This classical observation was made by Michael Posner and colleagues a quarter of a century ago and made them coin the classical metaphor of a 'spotlight' of attention. Beyond intuitive appeal, the virtue of any powerful metaphor comes from its simplicity, and the single spotlight metaphor has inspired a vast amount of psychological research centered on the question of how spatial attention modulates perceptual performance. While confirming Posner's basic observation the models emerging from these subsequent studies have produced several variations on the theme and raised the following questions: Does the spotlight have a zoom lens, can there be multiple separate spotlights, and is the spotlight's tuning enhanced by an inhibitory surround? And finally going beyond mere spatial mechanisms: Do the objects scattered across the visual field affect the spotlight's shape? Functional neuroimaging has permitted us to submit the various hypotheses surrounding the spotlight model to testing at the level of the associated neural activity changes. Our studies are based on the general finding that covert attending enhances activity in the retinotopic visual cortex that represents the selected visual field region. In a first study, we could demonstrate that the number of activated visual subareas increased when subjects had to monitor a larger visual field region. At the same time, the level of activity in a given area decreased compared to conditions in which attention could be narrowly focused and this level correlated with discrimination performance. We took these results as evidence for a zoom lens function implemented in the spotlight metaphor. In another study, by asking subjects to attend to a spatially defined part of an object we found that activity in the retinotopic visual cortex was modulated with respect to whether a subarea coded a location on the attended object or not. Moreover, during search when the relevant information had not turned up at the cued location, preferably those areas became activated that coded locations occupied by the already attended object. In yet another study, we demonstrated that the attentional focus encompasses an inhibitory surround where activity in the respective early visual cortex representations is suppressed relative to passive viewing. Together, these studies show that the spotlight metaphor is now well supported not only at the behavioral but also the neural level. The studies illustrate how spatial attention is optimized and can flexibly adapt to the ecological needs imposed both by the real visual world and the sensory mechanisms on which we rely to perceive it.