![]() ![]() To explain the interindividual differences in our findings, one could postulate either that the learning occurs for different subjects at different levels of analysis, or that the learning occurs at the same level but different subjects give different weight to different features of the scene. Finally, changing the hemifield to which stimuli were presented enabled bias reversal in over 50% of subjects, suggesting that the learning mechanisms utilise stimulus location cues to some extent. Secondly, for the majority of subjects, changing the surface features of the cylinders had no effect on the resistance of a trained binocular bias to reversal, indicating little selectivity for surface features in the learning mechanisms. First, we found that a monocularly trained bias was equally resistant to reversal in the trained eye and the untrained eye, suggesting that the site of learning is at, or follows, the site of binocular combination. We investigated the substrates of perceptual learning for bistable structure-from-motion stimuli by first inducing a directional bias, then examining whether changing single stimulus parameters affected the efficacy of counter-training to reverse this initial bias. ![]() Visually ambiguous stimuli are often used to study perceptual learning, but little is known of the neural mechanisms involved. We also report test–retest reliabilities based on a randomly selected subset of 100 participants. We report the distribution of directional biases observed in this sample, as well as their correlation with measures of handedness, ocular dominance and ideational type. In a follow-up study, we tested 1000 participants using the structure-from-motion stimulus. Biases were highly correlated between the two stimulus types, though we found some evidence that the initial frame of the pirouetting dancer animation could influence perceived direction on a given trial. This bias was consistent between presentation locations for most participants, but a minority reliably showed an opposite bias in the two hemifields. ![]() Nearly all participants showed a robust directional bias across two testing occasions. Twenty participants reported the perceived direction of rotation of structure-from-motion cylinders and silhouettes of a pirouetting dancer presented briefly to the left or right of fixation. We demonstrate a reliable perceptual polymorphism for a class of ambiguously rotating stimuli. The correspondence between theoretical models of temporal segmentation and motion perception in the human visual system, as well as psychophysical evidence that they have similar spatial limitations, support the notion that these processes are mediated by a common neural substrate situated in early in visual cortex. Similar units are also employed in the first stages of models for the perception of local motion. Here, we propose a computational model in which the neural signal for phantom contours originates from individual neurons with receptive fields that are space-time separable, like those mapped in cat and macaque V1. These limitations correspond closely to receptive field diameters of neurons in primary visual cortex (V1), and are also consistent with the spatial limitations of motion detection. Using psychophysical procedures, we determined the spatial limitations of this form of temporal segmentation across the visual field. ![]() Neighbouring regions flickering asynchronously at a high temporal frequency appear identical, but the visual system signals a 'phantom contour' between them. The timing of events can influence spatial segmentation. ![]()
0 Comments
Leave a Reply. |