transient mechanism for the perception of apparent motion

This study investigated the viability of a transient mechanism for the detection of counter-changing luminance, argued by Hock, et al. (2002) to be the informational basis for the perception of apparent motion. A series of experiments verified assumptions of the proposed mechanism and provided additional support for counter-changing luminance as the basis for apparent motion perception. It was found that: (1) The likelihood of perceiving apparent motion is best predicted by the product of local changes in luminance. This provided the basis for the multiplicative combination of subunit responses in the proposed mechanism (i.e. there is no motion signaled without coincident activation of both subunits). (2) When a brief interval of time separates a sequence of luminance onsets and offsets, or a sequence of luminance offsets, at a single element location, subunits exhibit summation of excitatory/inhibitory and excitatory/excitatory responses, respectively. This was consistent with the output of each subunit being determined by its biphasic temporal impulse response. (3) Apparent motion is specified only when there is a luminance offset at one location accompanied by a luminance onset at another location, and the likelihood of perceiving motion decreases with increases in the interval of time (ISI) separating the luminance offset from the luminance onset. Evidence that motion is not perceived beyond a limited range of ISI values indicated that the subunits respond transiently to luminance change. Accordingly, the effects of (ISI) are attributed to a reduction in the temporal coincidence of transient responses. This was supported by evidence that motion can be perceived when a luminance onset (indicating the end of the motion path) occurs before a luminance offset (indicating the start of the motion path). Computational simulations based on the product of transient responses to luminance offsets and onsets provide good qualitative matches to the experimental findings.