The movement aftereffect and a distribution-shift model for coding the direction of visual movement.

At present the only widely accepted explanation for the movement aftereffect is Sutherland's so-called ratio model, which states that motion is coded by taking the ratio between the outputs of detectors tuned to opposite directions. However, as yet there have been few attempts to derive predictions from the model in the context of movement aftereffects and test them experimentally. This paper reportws experiments which attempt to determine whether such a simple model is sufficient, or requires additional assumptions which recast it in a form more akin to the distribution-shift models used in other domains (which assume comparisons between outputs in the whole population of direction detectors, rather than just those tuned to opposite directions). These experiemtns examined the interactive effects of two simultaneous directions on subsequent aftereffect durations and directions. The results obtained are difficult to explain in terms of a simple ratio model but can be incorporated into a more complex distribution-shift type model.