Many studies have examined the impact of varying levels of sleep loss or sleep disturbance upon the multiple sleep latency test (MSLT). Virtually no studies have examined the impact of level of physiologic arousal upon measured sleep tendency. In the current study, 12 normal-sleeping young adults took modified multiple sleep latency tests after either watching television for 15 minutes or after a 5-minute walk. This entire protocol was repeated on another week after subjects had been partially sleep deprived by reducing their time in bed by 50%. It was hypothesized that sleep latencies would be significantly shorter after watching television as compared to walking and after partial sleep loss as compared to normal sleep and that these effects would be independent. The results of the study supported all of these hypotheses. ANOVAs showed no significant interaction effects, but sleep latencies were 11.6 and 5.8 minutes following the walk and watching television respectively. Sleep latencies were 9.8 and 7.6 minutes following baseline and partial sleep-deprivation conditions. Heart rate, used as a measure of physiological arousal, was significantly elevated throughout naps following the walk as compared to naps following television viewing. On a theoretical level, these data imply that measured sleepiness is a combination of sleep drive and physiological arousal, and these effects appear to be independent. On a practical level, these data indicate that more care may be necessary in monitoring the activity levels of patients and subjects prior to MSLT evaluations, since physiological arousal may mask the measurement of sleep tendency. Knowledge of the role of arousal in modulating sleepiness can be important in many settings.