Psychological Review 2000, Vol. 107, No. 3, 556-577 Copyright 2000 by the American Psychological Association, Inc. 0033-295X/00/S5.0T ~" * 5,00 DOT: 10.1037//0033-295X.107.3.556 Gamma-Range Oscillations in Backward-Masking Functions and Their Putative Neural Correlates Gopathy Purushothaman, Haluk Ogmen, and Harold E. Bedell University of Houston Backward-masking functions have been hitherto categorized into two types, commonly named Type A and Type B. The analysis of a model of Retino-Cortical Dynamics produces the prediction that spatially localized stimuli should reveal an oscillatory metacontrast function. The predicted new type of meta- contrast masking function was investigated in a psychophysical experiment. The results show oscillatory metacontrast functions with significant power in the gamma range (30—70 Hz). A marked decrease in the oscillations is observed when the spatial extent of the stimuli is increased. The theoretical basis of the study relates the oscillations found in the metacontrast function to gamma-range oscillations observed in scalp and intracerebral recordings. The qualitative agreement between the model and data provides support for this putative relationship. Backward masking, which is a form of visual masking, refers to a reduction in the perceived brightness (or a similar attribute) of a target stimulus when it is followed in time by a mask stimulus. The term metacontrast, introduced by Stigler (1910), refers to the special case of backward masking in which the target and mask stimuli do not overlap spatially. The perceived brightness (or other attributes such as contrast, contour completeness or detail, ftgural configuration or identity, etc.) of the target plotted as a function of the delay between the onsets of the target and the mask stimuli (stimulus onset asynchrony; SOA) is referred to as the metacon- trast function. Historically, metacontrast has been studied as a phenomenon and has been used as an experimental tool to probe spatiotemporal interactions in the visual system for more than a century (for reviews, see Alpern, 1952; Bachmann, 1994; Breitm- eyer, 1984; Breitmeyer & OJmen, in press; Kahneman, 1968; Weisstein, 1972). On the basis of these extensive studies, meta- contrast functions have been classified into two generic types: (a) A monotonic Type A function in which maximum masking occurs at SOA = 0 and (b) a nonmonotonic Type B function in which maximum masking occurs at a positive value of SOA (see Figure 1). An exception to this broad Type A/Type B classification scheme had been reported in a series of investigations in which the target and mask stimuli consisted of 1 arc min diameter dots 1 and the SOA values were sampled at 8-ms intervals (van der Wildt & Gopathy Purushothaman and Haluk Ogmen, Department of Electrical and Computer Engineering, University of Houston; Harold E. Bedell, College of Optometry, University of Houston. This study was supported in part by National Institutes of Health (NIH) Grants MH49892 and EY05068 and by NIH Core Center Grant T30- EY07551. We thank Talis Bachmann, Bruno Breitmeyer, Yuzo Chino, Gregory Francis, Laura Frishman, Dennis Levi, and Earl Smith for helpful suggestions. Correspondence concerning this article should be addressed to Haluk Ogmen, Department of Electrical and Computer Engineering, University of Houston, Houston, Texas 77204-4793. Electronic mail may be sent to ogmen@uh.edu. Vrolijk, 1981; Vrolijk & van der Wildt, 1982, 1985a, 1985b). It was shown that, under these conditions, the metacontrast function is bimodal; that is, peak masking occurs at two distinct SOA values (see Figure 1). Moreover, when the stimulus sizes were increased to values used in other studies, the shape of the metacontrast function changed from bimodal to Type B. Vrolijk and van der Wildt (1985a) attributed the existence of two peaks in the masking function to a propagating inhibition generated by the edges of the mask produced at its onset and offset. However, this explanation is inconsistent with their data showing that the SOA values at which maximum masking occurs increase as the target-to-mask separation is increased (see Figure 4 in Vrolijk & van der Wildt, 1985a). If masking were due to a propagating inhibition, the SOA for both peaks (i.e., the maxima) should decrease because the mask has to be presented earlier to compensate for the additional delay caused by the increased target- to-mask separation. Another explanation of multimodal masking functions can be based on Bridgeman's (1971) neural-network model, hi a compar- ative analysis of metacontrast models, it was shown that Bridge- 1 This stimulus configuration deviates somewhat from the more widely used stimuli in metacontrast. Typically, the contours of the mask stimulus surround at least partially and symmetrically the contours of the target stimulus (e.g., a disk [target] surrounded by a spatially nonoverlapping ring [mask] or a line [target] flanked by two adjacent lines [mask]). However, Stigler's (1910) definition of metacontrast does not necessitate these stimulus properties (e.g., Alpern, 1952). More important, previous studies using two spatially adjacent disks (a spatially extended version of the dot stimuli used in Vrolijk's & van der Wildt's [1985a] and our experiments) showed nonmonotonic, Type B masking functions (Breitmeyer, Battaglia, & Weber, 1976; Breitmeyer, Love, & Wepman, 1974). A comparison of data obtained from the disk-disk and disk-ring stimulus configurations shows virtually identical masking functions (Breitmeyer et al, 1974). For these reasons, we use the term metacontrast for the backward masking obtained with spatially and temporally adjacent dot stimuli. A previous model of metacontrast was based on the concept of anomalous apparent motion (Kahneman, 1967), which would arise in the symmetric stimulus configurations. Metacontrast was proposed to result from the inability of This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.