Abstract: What kinds of information do humans use to avoid or achieve a collision? This question has driven a large number of research studies on a variety of topics, including time-to-collision (TTC), and interception. Many studies have been designed around the assumption that TTC is estimated entirely on the basis of ‘tau’, but some other researchers have disagrees with this approach, reasoning instead that observers could start intercepting using relative distance or absolute distance alone. The aim of the present study was to investigate what kind of information is used in interception by hand. In addition, the speed coupling effect, has been explained by different and inconsistent theories. This study was designed to examine whether this phenomenon was related with the startup of hand.
Fifteen undergraduate students participated in both of two experiments. All of them had normal vision or vision correction to normal. Experiment 1’s program was run on a Dell computer using Borland C++ Builder. Subjects were asked to press the key to release a ball, in order to achieve a collision with a moving target ball. The distance between the target ball and the collision point were recorded when subjects released the ball. In experiment 2, Subjects were asked to hit the moving ball using their index fingers along the fixed paths. The tracks of the subject’s hand were recorded by a movement-analysis system based on active infrared markers (Optotrak 3020; Northern Digital). The program also recorded the distance between target ball and the interception point when the hand started to move. All of the dependent variables were analyzed with repeated measure ANOVA.
The results showed that (1) Different velocities of the target ball influenced the distance and time to collision differently. Time to collision was shorter in Experiment 1 if the target ball moved more quickly, and it was longer when the target ran more slowly. The same situation occurred in the interception made by hand. But the effects of the size of the target ball were different in Experiment 1 and 2. (2) The acceleration of the hand was affected significantly by the velocity of the target ball. If the velocity of ball was beyond Medium, there were no significant effects.
The conclusions are directed by the study: These results did not favor the attempts to account interception onset solely with tau hypothesis, nor the distance hypothesis. It appears that tau and distance may both be used in interception. The speed coupling effect is related with the error of the estimate.

Key words: interception, tau, onset of movement, speed coupling effect