Why is the Benary effect so small?

Benary [1924 Psychologische Forschung 5 131 - 142; translated into English in A Source Book of Gestalt Psychology Ed. W D Ellis (1939, London: Routledge & Kegan Paul)] was the first to systematically show that simultaneous lightness contrast (SLC) can be observed also when the local conditions at the target edges are identical. In his famous configuration, two identical gray triangles were placed one between two arms of a black cross and the other on the top of a big black triangle. The overall background was white. Therefore, they were locally bordering with the same amount of black and white area. Nevertheless, the gray triangle between the arms of the cross was perceived darker than the other triangle. Benary explained the effect according to the belongingness principle. It must be noted, however, that the effect is smaller than that observed in the classical SLC display. Recently, Agostini and Galmonte (1999 Perception & Psychophysics 61 1345 - 1355) showed that contrast induced by belongingness is not an all-or-nothing process. To investigate whether the lightness of the two targets is influenced by the colour of the non-belonging bordering regions, in the present work we measured the Benary effect separately. A Munsell scale was placed on a white background. The two halves of the Benary configuration were measured separately under Gelb lighting. Observers were asked to choose from the scale the patch matching the target in lightness. Since the scale was on a white background, the target belonging to the white background should be matched with its veridical value, while the target belonging to the black area should be matched with a lighter value. Surprisingly, both targets were matched with a lighter value, even though the amount of the induction was larger for the target included in the black area. Our experiments show that the colour of the non-belonging bordering regions in the Benary configuration has a detectable effect on the lightness of the target patch.