Noise, sex and time of day effects in a mathematics task

Abstract

A pilot experiment explored the effects of list length (four, six or eight digits) and presentation rate (1-6.5 per second) in a task of computer-paced addition. As the experimental variables changed, men tended to vary their speed of response while women tended to vary their level of accuracy. Conditions selected from the first experiment (six digits at two and four per second) were employed in a second, with the object of examining interactions among noise, time of day and sex. Practice effects over three sessions of testing confirmed the sex difference in speed-accuracy strategies. Male and female performance tended to change between the a.m. and p.m. test periods, such that women in quiet behaved similarly to men in noise, showing poorer p.m. performance in terms of response times and non-responses, but fewer commissive errors. Although the effects of circadian arousal were varied, noise consistently produced an effect different from that of time of day. A complex model, supported by more factual information, would be required to account fully for the results.