Mechanisms involved in the Münsterberg "shifted chequerboard" illusion: an empirical test

In the present work three alternative hypotheses proposed as explanations of the Münsterberg illusion are discussed: the Moulden and Renshaw 1979 hypothesis, which explains this effect using Helmholtz's 1856 "irradiation theory," the Gregory and Heard 1979 hypothesis, which proposes a mechanism of "border locking," and the McCourt 1982 hypothesis, which sustains a relation between this effect and brightness induction phenomena. To check these hypotheses, we took four measures, each one obtained by using a different pair of patterns mounted on circular displays which could be tilted by 360 degrees. The results do not support McCourt's hypothesis but, on the other hand, they seem to be partially in agreement, or at least not to be in disagreement, with the hypotheses both of Moulden and Renshaw and of Gregory and Heard.