"A criticism of both the above dichotomies [i.e., short range vs. long range motion, and first vs. second order motion] is that they reflect differences in the choice of stimuli, rather than qualitative differences in the underlying motion detection mechanisms. For example, much of the evidence for the existence of first-order and second-order mechanisms comes from comparing results from two types of stimuli: first, stimuli where the motion signal is carried by a spatio-temporal luminance correlation, and second, stimuli where motion is perceived in the absence of a luminance correlation, such as motion in a dynamic kinematogram or an amplitude modulated grating. Similarly, there is a wealth of information and data thought to characterize the short-range process, where the stimuli are spatially dense and movement is across small spatial displacements, whereas the long-range process has been investigated by using movement of figural stimuli across large displacements."
J. C. Boulton and C. L. Baker, Jr. Different parameters control motion perception above and below a critical density. Vision Res, 33(13):1803-11, Sep 1993.
Pubmed (Original entry:
http://saccade.tumblr.com/post/13628688847)