Visually perceived motion in depth resulting from proximal changes. I

Abstract

According to a model for fonn and motion perception proposed by Johansson (1964), every two-dimensional change in the proximal stimulation is projected out as a motion in depth. The amount of perceived depth motion can then be predicted from the projective relationship between the proximal change and the projected motion. This prediction was tested in a series of experiments by using squares that continuously changed their sizes as stimuli, and measuring perceived distance of motion in depth. The obtained relationship between perceived and predicted distance of motion was curvilinear for all Ss. Furthermore, the majority of the Ss underestimated the motion systematically, the remainder overestimated it. Thus, the prediction given in the model could not be verified. However, an alternative projective relation based on the assumption that a fixed proportion of the change is not projected out as a motion but perceived as a change of size agreed quite well with the data, both with distance judgments and with judgments of perceived change of size.

Visually perceived motion in depth resulting from proximal changes. II

“…However, the effect was very weak. According to the result of the earlier study (Marmolin, 1973), the difference in PSE-Z between the a and c patterns can be estimated to correspond to an increase in perceived motion of about 10%. This must be regarded as a very weak effect in comparison with the variation in rate of change, For example, at 80% change (42%/sec mean rate of change), the rate of change of the a pattern was 8%/sec and of the c pattern, 80%/sec.…”

“…In the earlier experiments (Marmolin, 1973), large interindividual differences in amount of perceived relative motion were found As in these experiments, the Ss had to be isolated from all sources of information about depth other than the proximal change; there was no information available about the initial distance. Therefore The perceived infinity can be interpreted as corresponding to the horizon of the perceived space.…”

“…In the earlier experiments (Marmolin, 1973;Marmolin & U1fberg, 1972: Johansson, 1964, the transformation patterns shrank or grew periodically with a constant rate of change, that is~x = r • t (t = time, r is a constant).…”

“…However, an alternative equation, based on the assumption that a fixed proportion of change (a) is not projected out as a motion, agreed quite well with the data. This equation was written In a previous article (Marmolin, 1973), a series of experiments was reported concerning the relation between perceived relative distance of motion in depth and relative amount of proximal change. The aim of these experiments was to test the validity of a formula for the stimulus·percept relationship derived from a model proposed by Johansson (1964).…”

“…The eartier study (Marmolin, 1973) showed that Eq. 1 does ,not predict the perceived amount of relative motion in depth under the condition used.…”