Human subjects and monkeys intercepted real (RM) and apparent (AM) moving targets that traveled through a low contrast circular path. The subjects intercepted the targets at 6 o'clock by applying a net force pulse on a semi-isometric joystick which controlled a cursor on the screen. Eight target speeds (180-560°/s) were used. The starting points of the moving target were systematically placed around the circle in order to determine the effect of the target travel time and velocity on the decision to initiate the interception movement and on the interception accuracy. It was found that the probability of interception in the first revolution varied as a function of the target travel time, which followed an S-shaped psychometric curve. The minimum processing time (MPT) was defined as the target travel that corresponded to a 75% probability of interception in the first revolution on the psychometric curve. The MPT decreased slightly as a function of target speed and was larger in AM than RM. In addition, the interception accuracy increased when the target travel time was above the MPT, and the angular error was smaller in RM than in AM. Finally, the interception movement was initiated at different target locations and time-to-contacts, depending on the target speed and the motion condition. Interestingly, similar findings were observed in human subjects and monkeys. These results suggest that the neural mechanisms engaged in extracting the visual motion information and in the implementation of the response are more efficient during RM than AM, and that such mechanisms need less processing time when the target is moving faster.