Cyclopean Motion Processing Does Not Depend Exclusively Upon Selective Attention

Abstract

This study investigated the degree to which the visual processing of cyclopean motion (motion of binocular disparity information) depends upon selective attention by employing a motion aftereffect paradigm. Cyclopean motion aftereffects were induced under conditions of low, high, and no attentional load. For comparison, luminance motion aftereffects were induced under the same conditions. Specifically, observers adapted to a moving cyclopean or luminance grating (spatial frequency = 0.5 c/deg; temporal frequency = 2 Hz) while performing a rapid serial visual presentation (RSVP) task that was superimposed over the central portion of the motion display. Observers attended to the RSVP task and ignored the motion display. For the low attentional-load RSVP task, observers determined whether words presented were upper or lower case; for the high attentional-load RSVP task, observers determined whether the words contained one or two syllables. For the no attentional-load condition, observers adapted to the moving grating without performing any RSVP task. Results showed that the duration of the motion aftereffect declined with increasing attentional load, but there was no reliable difference in the rate of decline between cyclopean and luminance motion (under high attentional load, the duration of the cyclopean motion aftereffect was 61% of that obtained under no attentional load). This equivalence between the two types of motion suggests that selective attention is mechanism invariant. These findings challenge the validity of the three-systems model of motion perception by Lu and Sperling (1995, 2001), which postulates that the processing of cyclopean motion, not luminance motion, depends exclusively upon selective attention.