- INSTRUMENTATION & TECHNIQUES Suitability of the IBM XT, AT, and PS/2 keyboard, mouse, and game port as response devices in reaction time paradigms
Sidney J. Segalowitz, Roger E. Graves · 1990
The IBM PC keyboard is a convenient response panel for subjects in a reaction time task when the stimuli are presented on the same machine. However, there is a mean delay of about 10 msec and a random error of ±7.5 msec (±5 msec on the AT or PS/2). Our analyses show that for typi cal single response experimental situations, this added variance is acceptable. With mouse but tons, timing resulted in a delay of 31 ± 2 msec if the mouse ball was steady but 45 ± 15 msec if it was moving, and a 25·msec refractory period before a second response could be detected. With keys connected to the game port, timing was accurate to 1 msec. For timing the interval between two nearly simultaneous responses, only the game port method is recommended. Any research application should provide an external check on reaction timing accuracy and should correct any mean error. The IBM PC family is becoming widely used for be havioral research, including reaction time studies. Unfor tunately, published reports almost universally fail to pro vide documentation of the quality of the timing provided, which may render the results suspect or provide data that are not comparable to data obtained with conventional tim ing apparatus. In this paper, we report the results of meas urements of timing made with the three most popular response approaches, and we offer recommendations for research applications. Keyboard