Representational Momentum -- JAVA version
Under appropriate conditions an observer's memory for the final position of an abruptly halted object is distorted in the direction of the represented motion.
In a classic representational momentum experiment, subjects are presented with a series of static inducing displays sampled from a path of motion (e.g. 3 rectangles at successive 17° rotations about the rectangle's center), and instructed to remember the last position. The subjects are then tested by asking them to decide if an additional stimulus (e.g. a fourth rectangle), presented at either the same orientation as the last inducing stimulus or slightly further or back along the implied path, was in fact at the same orientation as the last inducing display. Subjects are much less likely to reject forward than backward distractors.
In the example simulation on this page, the test position is in fact further along the path of motion than the last stimulus position!
The simulation above is implemented using Java. It will probably not display effectively with other world wide web browsers. An alternative version using animated GIF is also available, and is the recommended version. Note that the constraints of raster displays and a network environment imply that timing and positioning for the demonstration are very imprecise, particularly for the non-Java version. To maximize the momentum effect given these inducing stimuli, we would use stimulus durations of perhaps 250ms, interstimulus intervals after first and second stimuli of perhaps 250ms, and a retention interval before the test stimulus of about 300ms.
By the way, an inducing stimulus of 3 rotated rectangles is
not the most powerful way to produce RM. Try
continuous linear motion of a directional (arrow-like) cartoon
creature, e.g. a fish with a big tail.