Anthony D. D'Antona

Humans make efficient use of natural image statistics when performing spatial interpolation.

Visual systems learn through evolution and experience over the lifespan to exploit the statistical structure of natural images when performing visual tasks. Understanding which aspects of this statistical structure are incorporated into the human nervous system is a fundamental goal in vision science. To address this goal, we measured human ability to estimate the intensity of missing image pixels in natural images. Human estimation accuracy is compared with various simple heuristics (e.g., local mean) and with optimal observers that have nearly complete knowledge of the local statistical structure of natural images. Human estimates are more accurate than those of simple heuristics, and they match the performance of an optimal observer that knows the local statistical structure of relative intensities (contrasts). This optimal observer predicts the detailed pattern of human estimation errors and hence the results place strong constraints on the underlying neural mechanisms. However, humans do not reach the performance of an optimal observer that knows the local statistical structure of the absolute intensities, which reflect both local relative intensities and local mean intensity. As predicted from a statistical analysis of natural images, human estimation accuracy is negligibly improved by expanding the context from a local patch to the whole image. Our results demonstrate that the human visual system exploits efficiently the statistical structure of natural images.

Induced steady color shifts from temporally varying surrounds

The color appearance of a physically steady central region can appear to vary over time if a surrounding chromatic light varies in time. The induced temporal variation, however, is strongly attenuated at surround temporal frequencies above approximately 3 Hz. At these higher temporal frequencies, the central region appears steady (De Valois et al., 1986). The posited explanation is a cortical low-pass temporal filter. Here, we investigate whether higher temporal-frequency surrounds induce color shifts in the steady appearance of the central test. Surrounds modulated in time along the l or s chromatic direction of MacLeod-Boynton color space were symmetric around equal-energy white (EEW). The temporal frequency of the surround was varied. If observers perceived the central test to be temporally modulating between two points in time, they set two separate matches to the extreme points of this modulation. If the central test appeared steady in time, then color matches were made to this steady appearance. Corroborating previous reports, measurements showed that surround temporal frequencies below approximately 3 Hz induced temporal modulation. At higher temporal frequencies, however, the surround induced steady color shifts, compared to a steady surround at its time average (EEW). The measurements imply that a nonlinear neural process affects chromatic induction from time-varying context.

What kinds of contours bound the reach of filled-in color?

Is a retinal representation of an edge necessary to constrain the reach of color filling-in? If so, then color filling-in should not be constrained by illusory contours, because they do not exist at a retinal level. Alternatively, if color filling-in is constrained by contours at a perceptual level of neural representation, regardless of whether there is a retinal representation, then color filling-in should be constrained by illusory contours. To address this question, a variety of real luminance edges and illusory contours were presented under conditions designed to cause color filling-in. The results showed that illusory contours bounded the reach of color filling-in. A neural representation of a contour may first exist at a retinal level or a cortical level; in either case, the contour exists at a perceptual level and bounds color filling-in.

Induced temporal variation at frequencies not in the stimulus: evidence for a neural nonlinearity.

The perceived color of a light depends on surrounding light. When the surround varies slowly over time, a central, physically steady light is perceived to vary also. This perceived temporal variation of the central region, however, is strongly attenuated when the surround varies faster than approximately 3 Hz (R. L. De Valois, M. A. Webster, K. K. De Valois, & B. Lingelbach, 1986). The classical explanation is low-pass temporal filtering at a cortical stage that attenuates the neural representation of temporal frequencies capable of causing induced temporal variation. This theory assumes neural responses are linear, so only temporal frequencies in the stimulus are represented in the neural response. The current experiments revealed that temporal frequencies above 3 Hz are capable of inducing temporal variation. Specifically, with two temporal frequencies superimposed in the surround, the induced temporal variation in the uniform region is at the difference frequency of these two frequencies, even though this frequency is not physically present in the stimulus. The results are accounted for by a nonlinear neural process, which causes temporal variation at the difference frequency, and a following linear temporal filter.

Induced contrast asynchronies may be useful for luminance photometry

Shapiro et al. (2004) introduced a new visual effect (the induced contrast asynchrony) that demonstrates a perceptual separation between the response to a modulated light and the response to contrast of the light relative to background. The effect is composed of two physically identical disks, one surrounded by a dark annulus and the other by a light annulus. The luminance levels of both central disks were modulated in time, producing a stimulus with in-phase luminance modulation and antiphase contrast modulation. Observers primarily perceived the disks to be modulating asynchronously (i.e. they perceived the contrast), but at low temporal frequencies could also track the luminance level. Here we document that the induced contrast asynchrony disappears when the surrounds are achromatic and the center lights are modulated near the equiluminant axis. Observers viewed 1-deg-diameter disks embedded 2-deg-diameter achromatic surrounds. The chromaticity of the disks was modulated in time (1 Hz) along lines in an S versus Luminance cardinal color plane and an L-M versus Luminance cardinal color plane; observers responded as to whether the modulation appeared in phase. For all observers and both color planes, the lights appeared in phase most frequently at angles near the standard observers equiluminant line and out of phase at angles further away from that line. Observers differed in the range of angles that produce the appearance of in-phase modulation. The results suggest that induced contrast asynchronies may be useful as a technique for equating luminance of disparate lights.

Separating monocular and binocular neural mechanisms mediating chromatic contextual interactions.

When seen in isolation, a light that varies in chromaticity over time is perceived to oscillate in color. Perception of that same time-varying light may be altered by a surrounding light that is also temporally varying in chromaticity. The neural mechanisms that mediate these contextual interactions are the focus of this article. Observers viewed a central test stimulus that varied in chromaticity over time within a larger surround that also varied in chromaticity at the same temporal frequency. Center and surround were presented either to the same eye (monocular condition) or to opposite eyes (dichoptic condition) at the same frequency (3.125, 6.25, or 9.375 Hz). Relative phase between center and surround modulation was varied. In both the monocular and dichoptic conditions, the perceived modulation depth of the central light depended on the relative phase of the surround. A simple model implementing a linear combination of center and surround modulation fit the measurements well. At the lowest temporal frequency (3.125 Hz), the surrounds influence was virtually identical for monocular and dichoptic conditions, suggesting that at this frequency, the surrounds influence is mediated primarily by a binocular neural mechanism. At higher frequencies, the surrounds influence was greater for the monocular condition than for the dichoptic condition, and this difference increased with temporal frequency. Our findings show that two separate neural mechanisms mediate chromatic contextual interactions: one binocular and dominant at lower temporal frequencies and the other monocular and dominant at higher frequencies (6-10 Hz).

Chromatic interocular-switch rivalry

Interocular-switch rivalry (also known as stimulus rivalry) is a kind of binocular rivalry in which two rivalrous images are swapped between the eyes several times a second. The result is stable periods of one image and then the other, with stable intervals that span many eye swaps (Logothetis, Leopold, & Sheinberg, 1996). Previous work used this close kin of binocular rivalry with rivalrous forms. Experiments here test whether chromatic interocular-switch rivalry, in which the swapped stimuli differ in only chromaticity, results in slow alternation between two colors. Swapping equiluminant rivalrous chromaticities at 3.75 Hz resulted in slow perceptual color alternation, with one or the other color often continuously visible for two seconds or longer (during which there were 15+ eye swaps). A well-known theory for sustained percepts from interocular-switch rivalry with form is inhibitory competition between binocular neurons driven by monocular neurons with matched orientation tuning in each eye; such binocular neurons would produce a stable response when a given orientation is swapped between the eyes. A similar model can account for the percepts here from chromatic interocular-switch rivalry and is underpinned by the neurophysiological finding that color-preferring binocular neurons are driven by monocular neurons from each eye with well-matched chromatic selectivity (Peirce, Solomon, Forte, & Lennie, 2008). In contrast to chromatic interocular-switch rivalry, luminance interocular-switch rivalry with swapped stimuli that differ in only luminance did not result in slowly alternating percepts of different brightnesses.