Further evidence of failure of reversal of ambiguous figures by uninformed subjects.

In studying the reversal of ambiguous figures investigators understandably have always informed subjects of the reversibility of the figures and of how each version appears. However, such knowledge may be a cause of reversal and is therefore an undesirable aspect of the method used to study it. An experiment is described that is an improvement in some respects on a previously reported one in which subjects are not informed about reversal. The result was that only about a third of the subjects ever reversed. In a control condition the same subjects were informed in the traditional way and then not only did they always reverse but did so very frequently in the time period tested. Theoretical implications of these findings are discussed.