One of my favorite perceptual illusions isn't actually visual. It's often called the "cutaneous rabbit" illusion1, for reasons that will be apparent in a moment. I stumbled across it when reading a paper by Dennett and Kinsbourne2. Here's their description of the illusion (p. 188):
The subject's arm rests cushioned on a table, and mechanical square-wave tappers are placed at two or three locations along the arm, up to a foot apart. A series of taps in rhythm are delivered by the tappers, e.g., 5 at the wrist followed by 2 near the elbow and then 3 more on the upper arm. The taps are delivered with interstimulus
intervals between 50 and 200msec. So a train of taps might last less than a second, or as much as two or three
seconds. The astonishing effect is that the taps seem to the subjects to travel in regular sequence over equidistant
points up the arm--as if a little animal were hopping along the arm. Now that after the 5 taps on the wrist, there were going to be some taps near the elbow? The experienced "departure" of the taps from the wrist begins with the second tap, yet in catch trials in which the later elbow taps are never delivered, all five wrist taps are felt at the wrist in the expected manner.
That's not really a visual illusion, but fortunately for us, the future can affect the past in vision, too. There are several examples of this, the most famous of which is probably the "color-phi phenomenon," in which two dots of different colors presented near each other in rapid succession produces the illusion that the dot is moving from one the position of the first dot to that of the second, and changes color in between3 (Dennett and Kinsbourne discuss this one, too). But the illusion I like the most is the flash-lag illusion.
The flash-lag illusion is pretty subtle, so instead of having you look at an example first, and then explaining it, I'm going to do it the other way around. There are several ways of doing the flash lag illusion, but the basic illusion is this: when you present a flash aligned with an object, and immediately after the flash, that object moves, the flash is displaced from its actual position, and seems to lag behind the moving object. In the demonstration linked below (which is taken from an experiment by Eagleman and Sejnowski4, you'll see a green ring moving around a red square. The ring starts on the right, goes all the way around the square, then starts to go around again, but reverses direction when it gets to the left of the square, reverses direction again when it gets to the right, and finally settles to the left of the square. During all of this, three flashes will occur inside of the ring. The first flash occurs as the ring moves downward, on the left of the square, he second flash occurs just before the ring changes direction, and the final flash occurs when the ring finally stops for good. If you're experiencing the illusion, you should see the flash "lag" (i.e., it appears behind the ring) on the first and second flashes. On the first, it should appear slightly behind (and thus above) the ring, and in the second, slightly below the ring. OK, now go try it and then come back:
Did you see it? Remember, it's very subtle, so if you didn't notice it on the first attempt, go try again, or try this slightly different version (follow the instructions on the page). If you're still not seeing it, well then, I guess your visual system just isn't psychic like everyone else's.
What causes the flash-lag illusion? There's been a pretty heated debate about that over the last decade or so, and I'm not really qualified to say whether it's been resolved, but I can summarize the main positions and the evidence for them. The first researchers to study the illusion extensively believed it to be a result of the visual system predicting future motion based on cues at the time of the flash, in order to accomodate for delays in processing visual input5. In other words, the visual system is guessing. This hypothesis was, as far as I can tell, ruled out by the findings of Whitney and Murakami6, as well as Whitney, Murakami, and Cavanagh7. In their studies (which used lines instead of rings), the motion of the object after the flash was unpredictable based on the motion before the flash. The first demonstration above is similar to their stimuli, in that, while the motion of the ring is predictable when it is continuous (when the flash occurs as the ring goes around the first time), it is not predictable when the ring suddenly changes direction, yet the illusion still occurs. Thus, unless the visual system really is psychic, it can't be predicting. Instead, Whitney and Murakami argue that the illusion is due to differences in the amount of time it takes to process a flash vs. motion (the latency difference hypothesis). Motion, they argue, gets processed faster, and therefore the perception of the motion biases the later processing of the flash, even though the flash actually occurred first (in other words, the visual system ain't psychic, it's just slow).
A third theory argues that the illusion is the result of postdiction (as opposed to prediction)8. Under this view, the visual system uses both pre- and post-flash motion cues from the object (in the demonstrations above, the rings) to estimate the position of the object, and this results in the object appearing to be ahead of the flash. Evidence for the latency difference hypothesis (which involves postdiction as well) is also consistent with the posdiction hypothesis. However, Eagleman and Sejnowski9 argue that the latency difference hypothesis would also predict an illusory temporal ordering of the stimuli. If the motion information is processed faster than the flash, then the movement of the ring should be perceived as occurring before the flash. However, Eagleman and Sejnowski were unable to find evidence for this illusion, and conclude that the latency hypothesis cannot account for this.
So that's the debate in a nutshell. As far as I can tell, it's clear that the illusion is the result of the integration of the flash and motion information, which means that the visual system is postdicting, rather than predicting, the position of the moving object and its relationship to the flash. Whether this is due to response latencies or something else, I'm not qualified to say. It looks like the debate is still raging, and perhaps they'll discover new and interesting variants of the illusion in the process of resolving it.
1Geldard, F.A., & Sherrick, C.E., 1972, The Cutaneous "rabbit": a perceptual illusion. Science, 178, 178-9.
2Dennett, D.C. & Kinsbourne, M. (1992) Time and the observer. Behavioral and Brain Sciences, 15(2), 183-247.
3Kolers, P. & von Grunau, M. (1976). Shape and color in apparent motion. Vision Research, 16, 329-335.
4Eagleman, D.M., & Sejnowski, T.J. (2000). Motion integration and postdiction in visual awareness. Science, 287, 2036-2038.
5Nijhawan, R. (1994). Motion extrapolation in catching. Nature, 370, 256-257.
6Whitney, D. & Murakami, I. (1998). Latency difference, not spatial extrapolation. Nature Neuroscience, 1, 656-657.
7Whitney, D. V., Murakami, I., & Cavanagh, P. (2000). Temporal facilitation for moving stimuli is independent of changes in direction. Vision Research, 40, 3829-39.
8Eagleman & Sejnowski (2000).
9Eagleman, D.M., Sejnowski, T.J. (2001). The flash-lag illusion: distinguishing a spatial from a temporal effect, and why that matters for interpreting visual physiology. Journal of Vision, 1(3), 16a.
There are some recent papers out on the neural correlates of these illusions-- e.g. for the cutaneous rabbit illusion:
Blankenburg F, Ruff CC, Deichmann R, Rees G, Driver J (2006). The Cutaneous Rabbit Illusion Affects Human Primary Sensory Cortex Somatotopically, PLoS Biology, 4(3), doi:10.1371/journal.pbio.0040069.
Of course, we could discuss what exactly we gain from localization but it nevertheless interestingly extends our knowledge...