The motion correspondence problem is encountered when the visual system generates the illusion of apparent motion. In order to see motion, for some element in Frame I of an apparent motion display, the visual system must track its identity into Frame II. That is, it has to assert that some other element in Frame II is the same entity as the Frame I element. In short, it has to determine "what went where" before motion can be filled in.
This is a problem because if there are N elements in Frame I, and N elements in Frame II, then there are N! possible solutions to the motion correspondence problem. That is, this problem is a classic problem of underdetermination. Ullman (1979) argued that one natural constraint, the nearest neighbour principle, could be used to solve this problem. Dawson (1991) developed a connectionist network that permitted multiple constraints (nearest neighbour and relative velocity) to be used simultaneously to solve the problem.
- Dawson, M. R. W. (1991). The how and why of what went where in apparent motion: Modeling solutions to the motion correspondence process. Psychological Review, 98, 569-603.
- Ullman, S. (1979). The interpretation of visual motion. Cambridge, MA: MIT Press..
(Added November 2009)