THE CONTRIBUTION OF ONE-DIMENSIONAL MOTION MECHANISMS TO THE PERCEIVED DIRECTION OF DRIFTING PLAIDS AND THEIR AFTEREFFECTS

When motion aftereffects (MAEs) are measured by adapting to a drifting plaid (simultaneous adaptation) or by adapting to the plaid's component gratings in alternation (alternating adaptation), it has been shown that the velocity and duration of the MAE are smaller in the latter case [Wenderoth, P., Bray, R. and Johnstone, S. (1988) Perception, 17, 81-91; Burke, D. and Wenderoth, P. (1993) Vision Research, 33,351-359]. However, Burke and Wenderoth additionally reported that the directions of MAEs induced by simultaneous and alternating adaptation were identical, an apparent inconsistency if the differences in duration and velocity were due to the presence of ''blobs'' at the component grating intersects in the simultaneous case. Presumably, the direction of the ''blobs'' should also affect perceived plaid direction during adaptation and, hence, the MAE direction. In five experiments, we have measured both perceived adapting plaid and MAE direction, tested with both alternating and simultaneous adaptation, measured interocular transfer of plaid-induced MAEs and obtained MAE and plaid direction judgments under monocular and binocular viewing conditions. All of the data indicate that there is a blob tracking mechanism which is preferentially stimulated by plaids whose component gratings have high spatial frequency, low temporal frequency and high contrast. Differences between simultaneous and alternating adaptation emerge only when more optimal blobs are used, thus accounting for Burke and Wenderoth's failure to find a difference. The data also support Burke and Wenderoth's claim that the blob tracking mechanism is monocular: alternating and simultaneous adaptation produce identical MAEs under interocular transfer conditions, even using plaids with more optimal blobs, We also report the unexpected finding that plaids with more- and less-optimal blobs appear to drift in directions 20 degrees apart yet their aftereffects differ in direction by only 3-5 degrees. That is, more optimal blob plaids-compared with less optimal blob plaids-change both perceived plaid direction during adaptation and subsequent perceived MAE direction but the latter change is much more modest. Possible explanations of this dissociation are considered.