COLORED AFTEREFFECTS CONTINGENT UPON GLOBAL TRANSFORMATIONS

Dodwell and Humphrey (Psychol. Rev. 97, 78-79, 1990) have argued that colored after-effects (CAEs) are contingent on the global geometrics of local orientation and edge components in induction patterns. The spatiotopic dimensions underlying these geometries are derived from previous applications of the mathematics of continuous transformations (Lie groups) to the stimulus dynamics of perceptual invariance. In these terms, complementary CAEs induced by inspecting alternating colored bullseyes and spokes are attributed to the adaptation of orthogonal spatiotopic channels; the former characterized by vectorfields dealing with object invariance under rotation, and the latter by vectorfields dealing with object invariance under dilation. Importantly, the pattern-contingent CAEs induced in this pair of pattern channels are independent of CAEs induced in channels characterized by vectorfields dealing with object invariance under vertical and horizontal translations (e.g. vertical and horizontal gratings). The present investigation examined whether CAEs contingent on the stimulus dynamics originally used to define these spatiotopic channels exhibit comparable properties. It is confirmed that pairs of complementary CAEs can be induced simultaneously with radially expanding and contracting bullseyes, and clockwise- and counter-clockwise-rotating spokes (Experiment 1); and it is shown that these CAEs combine linearly in generalizing to and from the radial and rotational components of spiral motion (Experiments 2 and 3 respectively). In direct contrast to findings for CAEs induced with stationary versions of these patterns, however. it is demonstrated that complementary pairs of CAEs induced by expanding and contracting bullseyes interact with CAEs induced by left- and right-moving vertical gratings, suggesting that the components of motion underlying these contingent effects are locally and not globally determined (Experiment 4).