Visual attention

For more than a century, psychologists and physiolo-gists such as William James, Wilhelm Wundt, Johannes M€ uller and Hermann von Helmholtz have proposed that attention plays an important role in perception: it is necessary for effortful visual processing, and may be the ''glue'' that binds simple visual features into an object. In more recent times, many psychologists, including Ulric Neisser, Anne Treisman and Michael Posner, have developed experimental paradigms to investigate what attention does and what perceptual processes it affects. Over the past few years, there has been a growing interest in the mechanisms of visual attention: how visual attention modulates the spatial and temporal sensitivity of early filters; how attention influences the selection of stimuli of interest; how and where the neuronal responses are modulated; how attention and eye movements interact. In June of last year, about 40 visual scientists (Fig. 1) came together to discuss these issues in a quaint Fran-ciscan monastery in the Tuscan town of San Miniato, near Vinci, home of the great Leonardo (see Fig. 2). The meeting (San Miniato Workshop on Visual Attention, 12–16 June 2003: www.percezione.org/Attention2003) brought together visual scientists from many complementary disciplines: psychophysics, primate neurophys-iology, oculomotor research, and functional imaging. This special issue on visual attention reports many of the studies presented at the workshop. The topics cover a wide range of aspects of attention, including the properties and mechanisms of the selection processes, the spatial and temporal properties of attention, and the link between eye movements and attention. Many interesting techniques were employed in the various studies, including dual-task interference, noise and masking, contrast thresholds, speed-accuracy trade off, reverse correlation, visual search, neural imaging, and eye-movement monitoring. Can the visual system select a stimulus? Eckstein et al. report classification image studies showing that humans successfully select the stimulus at relevant locations and ignore information at irrelevant locations with both a 100% valid location cue and 100% invalid location cues. Solomon shows that an uninfor-mative cue may improve sensitivity at the cued location and that simultaneous cues can improve sensitivity at several locations. Other visual search studies also illustrate that the visual system is able to select cued locations in the presence of noise. Verghese and McKee find that in the absence of a location cue, the observer monitors early detectors at all locations that are well matched to the target attributes. Baldassi and Burr show that when visual search occurs …