Abstract
One hundred and thirty-five soldiers attending a 10-week training course in the Brecon Beacons were divided into three groups, each of which sat a different kit of five tests taken from the Kit of Factor-Referenced Cognitive Tests (Ekstrom et al., 1976). The tests were given three times, with the first administration serving as a baseline, and the second and third administrations being completed after exercises involving either two nights'' loss of sleep, or a physically fatiguing day. The experimental design allows the effects of practice and physical fatigue to be taken into account, and measures of the cognitive effects of sleep loss are obtained for each test. The results show that tests adversely affected by sleep loss are those that involve visual encoding and scanning, production of novel responses, as well as those in which lapses of attention critically affect performance. Performance on tests which involved logical thought and reasoning did not suffer, and in some cases improved.