The motion/disparity aftereffect: A preliminary study

The stimulus patterns used in this study were generated on the face of a cathode-ray oscilloscope (Tektronix 535). Most of these patterns were some variant of the familiar Lissajous figure, produced by driving the cathode-ray oscilloscope in X-Y mode with two Wavetek 134 function generators. Typically. the frequency of the generators differed by 1 or 2 cycles from a simple integer relationship, producing Lissajous figures which appeared to do ambiguous rotation in the horizontal plane. When these figures were viewed with a neutral density filter (typically 1.5 ND.) over one eye, the familiar Pulfrich effect (Rogers and Anstis, 1972) introduced apparent stereoscopic depth. with motion in the direction of the filtered eye being seen as lying in front of the screen. Thus an ambiguous rotating figure (e.g. Fig. 1) loses its ambiguity and appears very convincingly as a “ring”, rotating in depth. It should be noted that this experiment is not intended to study the Pulfrich effect, per se; rather, the effect is being used as a convenient tool for generating stereoscopic moving stimuli. For the experiment using random visual noise, the signal generators were adjusted to produce a raster. The raster lines were separated by about a mm when the generators were in synchrony; for the experiment they were purposely kept out of synchrony so that the lines were continually shifting and were altogether imperceptible. The Z-axis of the cathode-ray oscilloscope was driven by a 5 pet pulse generator, which was triggered by the zero crossings of a noise source (noise band I-10 kHz). The result was a random pulse train with a mean frequency of about 2200 pulses/set. This produced a stimulus similar to television “snow”, but considerably less dense. The noise “dots” were 10’ dia, and had a duration of about 20 msec. (These parameters are not very critical. The reader can verify these observations on an ordinary out-of-tune television, provided the contrast and brightness are adjusted for a low density of “snow”. The noise dots on a television are too closely spaced for fast apparent motion, but this can be remedied somewhat by viewing from a very short distance -about 6 in.)