Causal Capture
 
 
Here are some demos of the basic causal capture phenomena, discussed in the following paper:
Scholl, B. J., & Nakayama, K. (2002). Causal capture: Contextual effects on the perception of collision events. Psychological Science, 13(6), 493-498.
These demonstrations are provided as Quicktime movies, which can be downloaded or viewed directly in most web-browsers. (To download a free Quicktime player, go here.) These movies are a bit large and choppy, and the spatiotemporal parameters are slightly different than those used in the experiments, but they should be sufficient to illustrate the basic effect. In each case, you should judge the causal status of the upper event: does it look like a causal launch (wherein one item collides with another, causing its motion) or a non-causal pass (wherein one moving item simply passes across the entire display, passing over a stationary item)?  
 
Basic causality phenomena
This movie shows the basic 'launch' display, a la Michotte. This generates a reliable and nearly irresistible percept of causality. Contrast that with this movie, in which the added temporal gap destroys the perception of causality.
The critical 'pass' stimulus
This movie is identical to the canonical launch display, except that the two discs overlap for 100% of their area before the second motion. This display is actually bistable, but many observers are easily able to see this as a completely non-causal 'pass', wherein (1) a single object sits motionless in the center, changing color; while (2) another object moves across the entire screen, changing color. If you don't see this initially, you can see it by fixating below it somewhere, to move it into the periphery. Despite the color changes, you should at least be able to force yourself to see the noncausal pass (which is seen even when looking directly at the display by most naive observers).
The causal capture effect
In this movie you should still be judging the causal status of the top event -- which is in fact identical to that in the previous movie (with 100% overlap). Now, though, in the presence of the lower 'context' event which is perceived as an unambiguous launch, this display is 'captured' so that it too is seen as robustly causal (in contrast the noncausal 'pass' percept which dominates without the context). (Observers who tend to see the 'pass' stimulus as causal even when presented alone typically notice that they can at least force themselves to see it as a non-causal pass. In contrast, doing so is nearly impossible here, even with the identical stimulus.)
Effects of duration
This movie demonstrates that very little of the context event is necessary, so long as the actual 'impact' is visible. Here we still observe robust causal capture, even when only a brief moment of the context event is displayed.
Effects of temporal asynchrony
Finally, this movie demonstrates that although the duration of the context event isn't too important, the temporal synchrony between the two events is critical. Here we still have a completely unambiguous causal launch in the lower context position, but we observe no causal capture of the top display.
Conclusions
Collectively, these effects help to define a new class of rules which the visual system uses to construct percepts of causality.