Measurement of visual channels by contrast adaptation ROBERT J. SNOWDEN School o f Psychology, University o f Wales College o f \ PO Box 901, Cardiff CF1 3YG, U.K. SUMMARY Inspection of a high-contrast grating pattern affects our ability to detect patterns that are similar. This technique can be used to infer the underlying mechanisms of the visual system. By using this technique, measurements of the bandwidth of orientation channels are taken for different levels of adapting contrast and adapting duration. If the threshold elevation is plotted as the difference between the unadapted and adapted threshold in decibels, then the orientation bandwidth is invariant if taken at some fraction of the maximum elevation. This results from the fact that, as the orientation difference between the adapting and test patterns increases, the function relating threshold elevation to adapting contrast reduces in slope. These data contradict the often-used ‘ equivalent contrast transformation ’ (in which the fall off in the adaptation effect with respect to orientation is expressed in terms of an equivalent reduction in adapting contrast) as this would produce quite different bandwidths at different adapting contrasts. The data also address the issue of the neuronal mechanisms of adaptation. 1. I N T R O D U C T IO N After viewing a high-contrast stimulus for some time, our threshold to see that stimulus, as defined by the minimum contrast, is increased. If the test pattern differs in some respect (such as in its orientation) then the effect is often reduced. By measuring the amount of threshold elevation as a function of the difference in orientation between the adapting and test patterns we get an indication over the range of orientations that the two patterns interact. Such a technique has been used many times to reveal visual channels for spatial frequency (Pantle & Sekuler 1968; Blakemore & Campbell 1969), orientation (Blakemore & Nachmias 1971), direction of motion (Levinson & Sekuler 1980), and temporal frequency (Moulden et al. 1978) etc. Despite its widespread use as a psychophysical tool, the nature of the underlying changes that produce these effects are still unclear. This is most unsatisfactory as it severely limits inferences that can be drawn from data obtained in this way. For instance, adaptation is often used to measure the bandwidths of underlying channels. However, the estimates obtained by this technique are quite often at odds with those obtained by other techniques, such as subthreshold summation (for a review see Braddick et al. (1980)). One possible reason for the discrepancy between estimates is that certain stimulus variables may affect the orientation bandwidth. For example, orientation bandwidths may be affected by the spatial or temporal frequency of the patterns (Sharp & Tolhurst 1973; Phillips & Wilson 1984; Snowden 1992). In this paper I examine the role of two other variables that have varied from study to study, namely the contrast and duration of the adapting pattern, upon the orientation bandwidth. Clearly it would be desirable if these two factors did not have an important role in determining bandwidth as then they could be ignored when comparing data across studies. One important development in the use of contrast adaptation to measure channels was the introduction of the equivalent contrast transformation (ect) by Blakemore & Nachmias (1971). They had the in- genious idea that, as threshold elevation is reduced by decreasing the contrast of the adapting pattern, or by making the adapting pattern more dissimilar than the test pattern in terms of orientation (or spatial frequency etc.), then one effect could be described in terms of the other. Hence one could calculate the orientation difference that was equivalent to reducing the adapting contrast by a factor of two —the ect. This technique appears to have had great success in tying together data gathered under many different conditions (Blake- more & Nachmias 1971; Movshon & Blakemore 1973; Sharp & Tolhurst 1973). Given the success of the ect, can it also help in unifying data produced at different adapting contrasts and durations? The answer depends upon how increasing the contrast of the adapting pattern affects threshold elevation when the test and adapting patterns have the same orientation and when they have different orientations. To answer this question it is first necessary to have some notion of the way in which adapting contrast affects threshold elevation, and how increasing the difference in orien- tation between the adapting and test patterns affects threshold elevation. Thus the development of a quantitative model is delayed until this data has been presented. It should, however, be noted that this paper builds upon similar ideas and experiments performed in the domain of spatial frequency where the results have been contradictory (Stetcher et al. 1973; Dealy & Tolhurst 1974; Stromeyer et al. 1977; Swift & Smith 1982; Georgeson & Harris 1984). Proc. R. Soc. Lond. B (1991) 246, 53-59 Printed in Great Britain 53 Downloaded from https://royalsocietypublishing.org/ on 24 January 2023