ABSTRACT
Experiments are described that were designed to test a theory that the disproportionate slowing of performance by older subjects in continuous (as opposed to discontinuous) tasks is due to aftereffects of neural activity which increase in duration with age. At long interstimulus or interresponse intervals these effects would have time to dissipate, and thus would not affect the speed with which responses were initiated. At very short intervals, however, they would impair the making of different responses but have less effect on, or even facilitate, the making of repeated responses. It was argued that if the theory was correct, reaction times for repeated responses should tend to be faster than those for different ones, and that this tendency should increase with age. The results showed clearly that this did not happen but that, instead, there was an increased tendency with age for times for repeated responses to be slower than for different ones. The increase could be accounted for in terms of older subjects taking longer to discriminate repeated signals, and tending to monitor their responses more than young subjects. The results appear to endorse certain aspects of the single-channel hypothesis that were proposed many years ago but have not hitherto gained. much attention, and to suggest that previously neglected factors of discriminability and monitoring of responses can influence serial effects in reaction times.
