Metacontrast masking and the cortical representation of surface color: dynamical aspects of edge integration and contrast gain control

Abstract

This paper reviews recent theoretical and experimental work supporting the idea that brightness is computed in a series of neural stages involving edge integration and contrast gain control. It is proposed here that metacontrast and paracontrast masking occur as byproducts of the dynamical properties of these neural mechanisms. The brightness computation model assumes, more specifically, that early visual neurons in the retina, and cortical areas V1 and V2, encode local edge signals whose magnitudes are proportional to the logarithms of the luminance ratios at luminance edges within the retinal image. These local edge signals give rise to secondary neural lightness and darkness spatial induction signals, which are summed at a later stage of cortical processing to produce a neural representation of surface color, or achromatic color, in the case of the chromatically neutral stimuli considered here. Prior to the spatial summation of these edge-based induction signals, the weights assigned to local edge contrast are adjusted by cortical gain mechanisms involving both lateral interactions between neural edge detectors and top-down attentional control. We have previously constructed and computer-simulated a neural model of achromatic color perception based on these principles and have shown that our model gives a good quantitative account of the results of several brightness matching experiments. Adding to this model the realistic dynamical assumptions that 1) the neurons that encode local contrast exhibit transient firing rate enhancement at the onset of an edge, and 2) that the effects of contrast gain control take time to spread between edges, results in a dynamic model of brightness computation that predicts the existence Broca-Sulzer transient brightness enhancement of the target, Type B metacontrast masking, and a form of paracontrast masking in which the target brightness is enhanced when the mask precedes the target in time.

Keywords: achromatic color; brightness; brightness induction; edge integration; lightness; masking; metacontrast; paracontrast; type B masking.

Figure 1.

Demonstration of edge integration in lightness perception. The disks and rings on the two sides of the display have identical luminances, but appear lighter when viewed against a dark background. The effect of contrast effect induced by the background affects not only to the ring, which shares a border with the background, but also to the disk, which does not. The disk lightness is also affected by its luminance contrast with respect to the ring (simultaneous contrast). Quantitative studies of lightness matching have shown that the lightness of a target disk is determined by a weighted sum of the local log luminance ratios evaluated at the disk/ring and ring/background borders.

Figure 2.

Schematic diagram illustrating the stages involved in computing the brightness of a light target surrounded by a dark ring viewed against a light background, according to the edge integration model with contrast gain control. The graph at the top of the figure, labeled “luminance” shows a one-dimensional cross-section of the stimulus profile. This stimulus comprises the input to the edge integration computation. The graph below that, labeled “neural edge code,” shows the locations in which edge detector neurons encode the presence and the log luminance ratios of luminance borders in the input image. Separate neurons are assumed to encode edges having different contrast polarities. The third graph in the figure illustrates the fact that the responses of the edge encoding units that are nearer to the target disk are weighted more heavily in the computation of target brightness than are the response of remote edge encoding units. Contrast gain control acting between the inner and outer edges of the surround ring also contributes to the steady state values of the weights applied to the two edges. The bottom graph shows the profile of the target brightness, which is computed from the weighted sum of the disk/ring and ring/background edges. The inner edge, which has a light-inside contrast polarity, lightens the target to a degree that depends on the weighted log luminance ratio of the inner edge. The outer edge, which has a dark-inside contrast polarity, darkens the target to a degree that depends on the weighted log luminance ratio of the outer edge. Since the absolute magnitude of the weighted log luminance ratio at the inner edge is larger than the absolute magnitude of the weighted log luminance ratio at the outer edge, the target will appear light, rather than dark, relative to the background.

Figure 3.

Broca-Sulzer brightness enhancement occurs at stimulus onset for high intensity incremental targets. Here flash brightness is plotted as a function of duration for flashes of different luminances. Data from Hart (1987).

Figure 4.

A metacontrast masking paradigm modeled after the experiment of Weisstein (1971). A target consisting of an incremental disk displayed against a dark background is shown to one eye. Following a dark interstimulus interval of variable duration, a masking stimulus consisting of a decremental disk, larger in size than the target disk, is displayed to the contralateral eye. This stimulus differs from Weisstein's in that here the masking disk is dark, whereas in Weisstein's paradigm the target and mask both consisted of bright disks displayed against dark backgrounds. In both experimental paradigms, the target and mask each have only one edge.

Figure 5.

Proposed explanation of metacontrast based on edge integration and contrast gain control. The top graph in the figure shows the luminance profile of the stimulus. The target outline is indicated by a dotted line to signify that the target appears in an earlier frame than the mask (solid line). The presentation of the mask activates neural edge encoding units having the appropriate contrast polarity sensitivities and receptive fields at the locations of the mask edges (solid lines in the second graph). During the period in which the mask is presented, there may also be persisting activations in the edge encoding neurons that were activated by the target edges (dotted lines in the second graph). Both types of neural activations will potentially contribute to the target brightness, to a degree that depends on the edge weights. The third graph illustrates a case in which the weighted values of the neural activations corresponding to the target and mask edges happen to be identical. The edge weights are affected by two different processes. First, the target brightness computation algorithm tends to weight the target edge more heavily than it weights the more distant mask edge, all other things being equal. Second, a time-delayed contrast gain modulation acting from the target edge onto the mask edge will tend to boost the weight applied to the mask edge, with a particularly strong transient boost occurring at the optimal delay for metacontrast. In the hypothetical case illustrated, the darkness-inducing effect of the mask edge exactly cancels that lightness-inducing effect of the target edge, which results in the target brightness being neither higher nor lower than that of its immediate surround; thus, the target is made invisible. More generally, the target brightness may be modulated to a variable degree by the contrast gain control mechanism, with the largest target suppression effect occurring at the optimal SOA for metacontrast masking. If the contrast polarity of the mask edge is reversed, as in Weisstein's 1971 masking study, the transient gain modulation is attenuating, rather than amplifying (Rudd & Popa, in press). Since the presence of the mask edge in that case tends to lighten, rather than darken, the target, the transient attenuation of the lightness induction signal generated by the mask edge will also result in metacontrast masking.