查看更多>>摘要：Visual texture, defined by local image statistics, provides important information to the human visual system for perceptual segmentation. Second-order or spectral statistics (equivalent to the Fourier power spectrum) are a well-studied segmentation cue. However, the role of higher-order statistics (HOS) in segmentation remains unclear, particularly for natural images. Recent experiments indicate that, in peripheral vision, the HOS of the widely adopted Portilla-Simoncelli texture model are a weak segmentation cue compared to spectral statistics, despite the fact that both are necessary to explain other perceptual phenomena and to support high-quality texture synthesis. Here we test whether this discrepancy reflects a property of natural image statistics. First, we observe that differences in spectral statistics across segments of natural images are redundant with differences in HOS. Second, using linear and nonlinear classifiers, we show that each set of statistics individually affords high performance in natural scenes and texture segmentation tasks, but combining spectral statistics and HOS produces relatively small improvements. Third, we find that HOS improve segmentation for a subset of images, although these images are difficult to identify. We also find that different subsets of HOS improve segmentation to a different extent, in agreement with previous physiological and perceptual work. These results show that the HOS add modestly to spectral statistics for natural image segmentation. We speculate that tuning to natural image statistics under resource constraints could explain the weak contribution of HOS to perceptual segmentation in human peripheral vision.
查看更多>>摘要：Visual texture, defined by local image statistics, provides important information to the human visual system for perceptual segmentation. Second-order or spectral statistics (equivalent to the Fourier power spectrum) are a well-studied segmentation cue. However, the role of higher-order statistics (HOS) in segmentation remains unclear, particularly for natural images. Recent experiments indicate that, in peripheral vision, the HOS of the widely adopted Portilla-Simoncelli texture model are a weak segmentation cue compared to spectral statistics, despite the fact that both are necessary to explain other perceptual phenomena and to support high-quality texture synthesis. Here we test whether this discrepancy reflects a property of natural image statistics. First, we observe that differences in spectral statistics across segments of natural images are redundant with differences in HOS. Second, using linear and nonlinear classifiers, we show that each set of statistics individually affords high performance in natural scenes and texture segmentation tasks, but combining spectral statistics and HOS produces relatively small improvements. Third, we find that HOS improve segmentation for a subset of images, although these images are difficult to identify. We also find that different subsets of HOS improve segmentation to a different extent, in agreement with previous physiological and perceptual work. These results show that the HOS add modestly to spectral statistics for natural image segmentation. We speculate that tuning to natural image statistics under resource constraints could explain the weak contribution of HOS to perceptual segmentation in human peripheral vision.
查看更多>>摘要：Spatial and temporal levels of information processing interfere with each other. The Kappa effect is a well-known spatiotemporal interference in which the estimated time between two lights increases as the distance between them increases, showing a deceleration tendency. A classical model attributes this interference to constant speeds and predicts a linear relation, whereas a slowness model attributes the interference to slow speeds and proposes that the tendency is due to the uncertainty of stimuli locations. This study integrated a unifying Bayesian framework with the classical model and argued that this tendency is the result of the Weber-Fechner law. This hypothesis was tested via two time discrimination tasks that manipulated the uncertainty of stimuli locations and the distance between stimuli. Experiment 1 showed that the estimated time was not modulated by the uncertainty of the stimuli locations. Experiment 2 revealed that the behavioral predictions made by the Bayesian model on logarithmic scales were more accurate than those made by the linear model. Our results suggest that the deceleration tendency in the Kappa effect is the result of the Weber-Fechner law.
查看更多>>摘要：Spatial and temporal levels of information processing interfere with each other. The Kappa effect is a well-known spatiotemporal interference in which the estimated time between two lights increases as the distance between them increases, showing a deceleration tendency. A classical model attributes this interference to constant speeds and predicts a linear relation, whereas a slowness model attributes the interference to slow speeds and proposes that the tendency is due to the uncertainty of stimuli locations. This study integrated a unifying Bayesian framework with the classical model and argued that this tendency is the result of the Weber-Fechner law. This hypothesis was tested via two time discrimination tasks that manipulated the uncertainty of stimuli locations and the distance between stimuli. Experiment 1 showed that the estimated time was not modulated by the uncertainty of the stimuli locations. Experiment 2 revealed that the behavioral predictions made by the Bayesian model on logarithmic scales were more accurate than those made by the linear model. Our results suggest that the deceleration tendency in the Kappa effect is the result of the Weber-Fechner law.
查看更多>>摘要：Face recognition skills are distributed on a continuum, with developmental prosopagnosics and super-recognisers at the bottom and top ends, respectively. Holistic processing propensity is associated with face recognition ability and may be impaired in some developmental prosopagnosics and enhanced in some super-recognisers. Across two experiments we compared holistic processing of 75 super-recognisers and 89 typical-range ability controls using The Part-Whole Effect (PWE) paradigm. A subgroup of super-recognisers demonstrated enhanced PWEs in the nose region, suggesting they integrate the nose into the holistic face percept more effectively than controls. Focussed processing of the nose region, an optimal viewing position to extract the holistic properties of faces, has previously been associated with superior face recognition, and this may partly explain the superiority of some super-recognisers. However, a few super-recognisers generated extreme nose region performance patterns in an opposite direction across both experiments, suggesting their superiority is driven by alternative mechanisms. These results support proposals that super-recognition is associated with heterogeneous underlying processes.
查看更多>>摘要：Face recognition skills are distributed on a continuum, with developmental prosopagnosics and super-recognisers at the bottom and top ends, respectively. Holistic processing propensity is associated with face recognition ability and may be impaired in some developmental prosopagnosics and enhanced in some super-recognisers. Across two experiments we compared holistic processing of 75 super-recognisers and 89 typical-range ability controls using The Part-Whole Effect (PWE) paradigm. A subgroup of super-recognisers demonstrated enhanced PWEs in the nose region, suggesting they integrate the nose into the holistic face percept more effectively than controls. Focussed processing of the nose region, an optimal viewing position to extract the holistic properties of faces, has previously been associated with superior face recognition, and this may partly explain the superiority of some super-recognisers. However, a few super-recognisers generated extreme nose region performance patterns in an opposite direction across both experiments, suggesting their superiority is driven by alternative mechanisms. These results support proposals that super-recognition is associated with heterogeneous underlying processes.
Durai, C. Vijay ReenaRajendran, SiddhartWebster, Michael A.Vempati, Sandeep...
9页
查看更多>>摘要：In the Pulfrich illusion, the depth of a moving object is misperceived due to induced retinal disparity and/or interocular velocity differences arising from differences in luminance, contrast, or spatial frequency between the two eyes. These effects have been shown to occur both for visual deficits and for optical corrections that introduce significant binocular differences between the retinal images. However, it remains unknown to what extent the illusion might arise given normal variation between the eyes, such as natural interocular variation in pupil diameter (anisocoria). To assess this, we examined the threshold interocular retinal illuminance difference required to experience illusory depth in two random-dot fields moving in opposite directions in 24 normallysighted observers with dilated pupils. Interocular difference in retinal illuminance was induced by placing neutral density filters of different intensities before the left eye. A minority of subjects (n = 8) did not provide meaningful data on changes in the experience of illusory depth with interocular difference in retinal illuminance and four subjects showed biases >+/- 10% from the 50% point of subjective equality in the psychometric function. For the remaining 12 participants, the retinal illuminance had to differ by approximately 40% for the depth between the planes to become visible at threshold levels. This difference was approximately constant over a range of absolute luminance levels from 10 to 80 cd/m2. Our results suggest that while motion-in-depth illusions due to interocular differences in retinal illuminance may be pronounced in certain ophthalmic diseases or following certain optical interventions, it is unlikely to be manifest as a result of normal interocular variations in retinal illuminance. Further, our results also point towards the existence of substantial individual differences in the experience of what is otherwise thought of as a readily appreciable motion-in-depth illusion.
Durai, C. Vijay ReenaRajendran, SiddhartWebster, Michael A.Vempati, Sandeep...
9页
查看更多>>摘要：In the Pulfrich illusion, the depth of a moving object is misperceived due to induced retinal disparity and/or interocular velocity differences arising from differences in luminance, contrast, or spatial frequency between the two eyes. These effects have been shown to occur both for visual deficits and for optical corrections that introduce significant binocular differences between the retinal images. However, it remains unknown to what extent the illusion might arise given normal variation between the eyes, such as natural interocular variation in pupil diameter (anisocoria). To assess this, we examined the threshold interocular retinal illuminance difference required to experience illusory depth in two random-dot fields moving in opposite directions in 24 normallysighted observers with dilated pupils. Interocular difference in retinal illuminance was induced by placing neutral density filters of different intensities before the left eye. A minority of subjects (n = 8) did not provide meaningful data on changes in the experience of illusory depth with interocular difference in retinal illuminance and four subjects showed biases >+/- 10% from the 50% point of subjective equality in the psychometric function. For the remaining 12 participants, the retinal illuminance had to differ by approximately 40% for the depth between the planes to become visible at threshold levels. This difference was approximately constant over a range of absolute luminance levels from 10 to 80 cd/m2. Our results suggest that while motion-in-depth illusions due to interocular differences in retinal illuminance may be pronounced in certain ophthalmic diseases or following certain optical interventions, it is unlikely to be manifest as a result of normal interocular variations in retinal illuminance. Further, our results also point towards the existence of substantial individual differences in the experience of what is otherwise thought of as a readily appreciable motion-in-depth illusion.
查看更多>>摘要：Viewers can summarize redundant features in groups of objects into an ensemble percept. There appears to be separate mechanisms underlying ensemble perception of low-and high-level visual features, but it is unclear whether ensemble perception of different low-level features is supported by common mechanisms. Yo center dot ruk and Boduroglu, in Attention, Perception, & Psychophysics 82 (2020) 852-864, investigated whether length and orientation summarization tap common mechanisms by examining the correlation between errors on length-and orientation-averaging tasks and concluded that because they did not find any correlations, the two features are summarized by different, feature-specific mechanisms. However, their study was conducted with a small sample size and included sources of individual performance variance that may have diminished correlations. We report two studies that tested the correlation between performance in the length-and orientation-averaging tasks, with larger samples and modifications that sought to reduce the sources of variance. Study 1 used ensembles that varied in both feature dimensions and Study 2 used ensembles that only varied in the task-relevant dimension. Both studies showed that errors in length-and orientation-averaging are correlated, suggesting that ensemble perception of these low-level features is supported, at least to some extent, by a common ability.
查看更多>>摘要：Viewers can summarize redundant features in groups of objects into an ensemble percept. There appears to be separate mechanisms underlying ensemble perception of low-and high-level visual features, but it is unclear whether ensemble perception of different low-level features is supported by common mechanisms. Yo center dot ruk and Boduroglu, in Attention, Perception, & Psychophysics 82 (2020) 852-864, investigated whether length and orientation summarization tap common mechanisms by examining the correlation between errors on length-and orientation-averaging tasks and concluded that because they did not find any correlations, the two features are summarized by different, feature-specific mechanisms. However, their study was conducted with a small sample size and included sources of individual performance variance that may have diminished correlations. We report two studies that tested the correlation between performance in the length-and orientation-averaging tasks, with larger samples and modifications that sought to reduce the sources of variance. Study 1 used ensembles that varied in both feature dimensions and Study 2 used ensembles that only varied in the task-relevant dimension. Both studies showed that errors in length-and orientation-averaging are correlated, suggesting that ensemble perception of these low-level features is supported, at least to some extent, by a common ability.
NSTL
Elsevier