Three experiments are reported in this thesis. The first two experiments were concerned with counteracting the performance decrement observed in vigilance tasks. Previous studies have shown that the criterion parameter β ,of signal detection theory, increases during a vigilance session. The experiments reported here manipulated variables affecting β in order to keep it at a constant level throughout a session. In experiment I, signal probability was increased within sessions in an auditory vigilance task. This manipulation reduced the decrement in performance below that shown by control groups. The second experiment involved a visual task. Signal probability was held constant within sessions and artificial signal probability was increased. This also had the effect of reducing the performance decrement. The third experiment investigated the relationship between discrimination threshold and the vigilance decrement. Subjects' thresholds for discriminating between two visual stimuli were obtained by means of the PEST (Parameter Estimation by Sequential Testing) technique. A decrement in performance was found when subjects performed at a 75 or 60% correct level on the threshold task, but there was no decrement when subjects obtained 90% correct on the threshold task.

Master of Arts (MA)