Hebb, 1955;Malmo, 1959) have suggested that the relationship between performance and arousal or "activation" is a curvilinear one. These authors propose that an optimal level of arousal exists for any given task and deviations from this level in either direction will result in a decrement in performance.A study by Belanger (1959), using heart rate as the measure of activation, has yielded results favorable to this inverted U hypothesis. In this study, heart rate and performance in a Skinner box were measured at five levels of water deprivation. Results showed that, while heart rate increased steadily from 24 through 72 hr. of deprivation, performance increased, then reached a peak at 48 hr., and decreased thereafter.A number of objections to Malmo's interpretation (1959) of Belanger's results can be raised, (a) If, during training, heart rate became adapted to the 24-hr, deprivation