White's Effect

Perception of colour and contrast in complex configurations

The appearance of a surface as white, grey, or green is determined not only by the light reflected from that surface but also by the light reflected from surrounding regions. The change in the appearance of a surface caused by its juxtaposition with other surfaces is referred to as an induction. Although much studied, many colour induction phenomena still await a satisfactory explanation. My initial results implied that early sensory processing must be accompanied by a more elaborate stage of perceptual interpretation and surface analysis in order to achieve reliable perception of surface colours. I have been investigating the perception of colour and contrast in complex scenes with special emphasis on the possible mechanisms by which cues that are used to perceive three-dimensional organisation influence perceived contrast and colour. Such cues have to be factored into any adequate explanation of colour and contrast induction. Understanding the mechanisms by which these perceptual cues modify and affect the output of low-level mechanisms as well as their involvement in processes such as surface segmentation and figure ground segregation will eventually lead to mechanistic models that can explain surface perception in natural settings.

Perceptual grouping of contour fragments

The visual system parses the visual scene into objects and their spatial locations by integrating local signals into global object representations. It routinely and successfully decomposes images into distinct surfaces along extracted edges, and/or segregates image regions into multiple layers when viewing objects through transparent or semi-transparent media. We have been investigating the grouping of segments in configurations containing illusory, partially closed and occluded contours, with special emphasis on the role of the contrast polarity of the elements to be grouped. The effect of contrast polarity has featured prominently in investigations of perceptual grouping and, in this research, we revisit and challenge some widely held assumptions on the sensitivity of a particular grouping process to variations in the contrast polarity of the elements to be grouped. Our results indicate that the effect of contrast polarity reversals on the perception of occluded, occluding, and partially closed surfaces depends crucially on the spatial distribution of such reversals within the inducing configurations. These results suggest the previously unsuspected importance of image properties related to ecologically relevant constraints in the perception of illusory contours and occlusion.

Attentional Capture

The determinants of attentional capture

What controls the allocation of attention to particular locations in visual space is a question of fundamental theoretical and practical interest in cognitive psychology. It has been well documented that salient aspects of external stimuli can capture visual attention, but a considerable controversy exists over whether attention is captured at the level of object-representation or at the level of sensory change detection. In our research we examine the principles that determine which sources of information are attended during involuntary attention shifts. The differential effects of low-level versus object-based considerations in attentional capture are dissociated and directly compared.

White's Effect

Aesthetic preference in natural and synthetic imgages

Experimental aesthetic is the second oldest area of experimental psychology but the question of whether aesthetic preferences are primarily driven by the context in which the art is created and viewed, or whether they are automatic, universal and integral to our experiences remains a crucial and unanswered question. The view of ‘beauty as in the eye of the beholder’, attributes aesthetic experience to the ever-changing individual, intellectual, and cultural contexts and considers efforts to systematically (and scientifically) study aesthetics as futile. The opposite view of aesthetic experience as primarily determined by objective stimulus features such as symmetry, proportion or complexity, falls short of being able to specify the exact functional relationship between aesthetic experience and any of these stimulus qualities. This work is motivated by the initial work of Richard Taylor (University of Oregon) showing that paintings by the artist Jackson Pollock have statistical properties similar to those of natural scenes. Taylor's hypothesis is that the aesthetic appeal of Pollock's work derives from its similarity with the scenes in which our visual systems have evolved to function. Indeed, our initial data suggest an aesthetic preference for natural images such as clouds and waves, which share same fractal dimension as the most popular of Pollock's paintings. These data are consistent with a biologically based theory of aesthetics that focuses on the role of visual processing in the rewarding experiences of visual images.