Conscious perception of illusory colour

Visual perception can be defined as the ability to interpret the pattern of light entering the eyes to form a reliable, useful representation of the world. A well-accepted perspective suggests that these interpretations are influenced by prior knowledge about the statistics of natural scenes and are...

Full description

Bibliographic Details
Main Author: Powell, Georgina
Published: Cardiff University 2012
Subjects:
Online Access:http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.590318
Description
Summary:Visual perception can be defined as the ability to interpret the pattern of light entering the eyes to form a reliable, useful representation of the world. A well-accepted perspective suggests that these interpretations are influenced by prior knowledge about the statistics of natural scenes and are generated by combining information from different cues. This thesis investigates how these processes influence our perception of two phenomena: afterimages and colour distortions across the visual field. Both are generated on the retina, do not represent meaningful properties of the physical world, and are rarely perceived during natural viewing. We suggested that afterimage signals are inherently ambiguous and thus are highly influenced by cues that increase or decrease the likelihood that they represent a real object. Consistent with this idea, we found that afterimages are enhanced by contextual edges more so than real stimuli of similar appearance. Moreover, afterimage duration was reduced by saccadic eye-movements relative to fixation, pursuit, and blinking, perhaps because saccades cause an afterimage to move differently to real object and thus provide a cue that the afterimage is illusory. Contextual edges and saccades were found to influence afterimage duration additively, although contextual edges dominated the probability of perceiving an afterimage more than saccades. The final part of the thesis explored the hypothesis that colour distortions across the retina, produced mainly by spectral filtering differences between the periphery and fovea, are compensated in natural viewing conditions. However, we did not find evidence of compensatory mechanisms in the two natural conditions tested, namely eye-movements (as opposed to surface movements) and natural spectra (as opposed to screen-based spectra). Taken together, the experiments in this thesis demonstrate that these ‘illusory’ phenomena perceived strongly in laboratory conditions but rarely during natural viewing, are useful tools to probe how perceptual decisions are made under different conditions.