Abstract
Two fundamentally different motion-processing mechanisms have been proposed, one "passive" and the other "active" (Cavanagh, 1991, 1992; Lu and Sperling, 1995, 2001). The passive system is a purely feed-forward process that utilizes a motion energy computation. The active system has been shown to be rely on attentive or feature-tracking mechanisms, and can account for motion percepts that are present in specialized stimuli where energy computations do not produce a specific direction. The extent to which active and passive mechanisms contribute to the processing of an "ordinary" stimulus is difficult to asses on the basis of perceptual judgments alone because one does not know which signal the subject uses to make the report and because it is difficult to determine what type of processing occurs in the absence of attention. Evoked Potential measures are ideal for studying this question because measurements of motion processing can be made equally well when attention is directed to motion or diverted from it. Arrays of small grating patches were presented to 12 adults at different spatial displacements and temporal offsets to produce either a vivid percept of apparent motion or a percept of flashing without motion. Two spatial separations of paired-flash targets were used (0.25 wavelength of the grating or 3 wavelengths). Patch onsets were sequential (apparent motion) or simultaneous (control). Observers either fixated a cross in the center of the screen and attended to the patches or performed a difficult letter discrimination task at fixation to divert attention. There was a small, but measurable effect of attention in the motion conditions, but a substantially larger one in the control conditions. In both cases attention effects were largest in the interval of 200 to 400 msec indicating that classic apparent motion displays engage processes that are largely independent of attention.
Meeting abstract presented at VSS 2014