Abstract
In a time-to-contact task 18 volunteers were asked to tilt a lever right or left at the exact moment a moving spot overlapped a stationary spot on a computer screen. At different times before this critical moment, a third spot moved, indicating the side the lever should be tilted. The interval between the instant the direction was indicated and the instant the lever should be tilted varied randomly from 51 to 1071 ms. Clearly, for very short intervals (50–150 ms) responses in the direction indicated were 50% of the total (random performance), since it is not possible to react to the third spot that fast. When the time available to react was >350–400 ms, on the other hand, performance saturated at 100% correct. The half way mark, between random and perfect performance occurred when the time available to respond was around 220–260 ms. Two interesting phenomena were observed: for short latencies (50–200 ms) there was a tendency towards late (5–7 ms) responses; for much longer latencies (450–550 ms), minimum condition for perfect responses, there was a strong tendency to anticipate responses (10–12 ms, p < 0.05). It is easy to understand the delays with short latencies as a natural, even if involuntary, tendency to wait just that instant longer when unsure which way to go. Much more difficult to understand is the anticipatory tendency when the time available to respond is much longer. Perhaps in the range of 500 ms prior to the actual movement, preparatory internal processes occur, involved with synchronizing the response to the visual stimulus. When the indication of the side to which the movement should occur coincides with these processes, the resulting interference might be the cause of this phenomenon. Indirect evidence for this was obtained from eye-movement recordings, but other explanations cannot be excluded at this time.