Galmonte and Agostini (1998 Investigative Ophthalmology & Visual Science, 39(4), S158), Agostini and Galmonte (2002 Psychological Science, 13, 89-93), Bressan (2001 Perception, 30, 1031-1046), and Gilchrist and Annan (2002 Perception, 31, 141-150) reported three different lightness contrast configurations in which grouping factors make a gray target totally surrounded by black appear darker than an equal gray target surrounded by white, reversing the classical contrast effect. In this paper we demonstrate that the three configurations known as 'reversed contrast' are based on different mechanisms. Sixteen participants judged the lightness of the gray targets of the original and modified versions of the three configurations. Our results highlight that the Agostini and Galmonte effect is reversed when the global grouping factors are removed, while in a number of variations of Bressan's and Gilchrist and Annan's displays the direction of the effect does not change, even in absence of global grouping factors. Our results indicate that the factors determining the Agostini and Galmonte effect are different from those acting on the other two configurations, in which the lightness change is also due to factors other than belongingness.