Perception, 1984, volume 13, pages 461-468 Diverting subjects' concentration slows figural reversals Daniel Reisbergf, Mark O'Shaughnessy Department of Psychology, New School for Social Research, 65 Fifth Avenue, New York, NY 10003, USA Received 12 March 1984 Abstract. Earlier research has shown that subjects are slower to discover the alternative construal of ambiguous geometric figures if they are simultaneously engaged in distractor tasks. Results are reported which show that this effect slows reversals subsequent to the first reversal, and is therefore not merely slowing the discovery of the alternative construal. In addition, the magnitude of the distraction effect seems unchanged through the successive reversals. The results are discussed in terms of their implications for figural reversal and of their implications for the role of mental resources in perceiving. 1 Introduction What causes ambiguous figures, like the Necker cube or the Schroder staircase, to change their phenomenal appearance? While the classical account of these reversals has appealed to the satiation over time of some neural process (Kohler and Wallach 1944), recent evidence suggests that figural reversal may be more of a 'cognitive' achievement: Girgus et al (1977) report that figural reversals are considerably delayed if the observer does not know that he/she is looking at an ambiguous figure. Reisberg (1983) reports that latencies for first reversal are appreciably slowed if the subject is concurrently engaged in some other task, such as mental arithmetic or carrying a digit sequence in memory. One might well wonder whether these 'cognitive' effects apply to reversals as such, or only to the first reversal. The first reversal requires the discovery of an alternative construal of the ambiguous figure, while subsequent reversals require only the fitting of a construal already discovered. The finding of Girgus et al (1977), that knowledge of the ambiguity of a figure has considerable effect on reversal latency, is by its very nature relevant only to the first reversal. In the present experiment we ask whether Reisberg's (1983) finding that diverting attention slows figural reversal is true only for the initial reversal of an ambiguous figure. There is a plausible basis for making each of three predictions about the effects of attention manipulations on reversals after the first. As already suggested, it may be the discovery of an alternative construal which requires the subjects' attention, so that once this discovery is made attention becomes irrelevant; hence we would predict no effect of attention manipulations on reversals after the first. Alternatively, one could argue that diverting attention, having delayed the first reversal (Reisberg 1983), will then facilitate subsequent reversals. Hochberg (1970, page 120) suggests that, in viewing an ambiguous stimulus, attention may be needed to 'maintain' a pattern of anticipations, ie to maintain a percept. Hochberg suggests, therefore, that diverting attention, which will disrupt this maintenance, will speed up the rate of reversal. Finally, one might predict that attention manipulations will slow all figural reversals, and not just the first. This would follow from the claim that attention is necessary for (or synonymous with) the processes underlying the recognition of f Author to whom requests for reprints should be addressed.