Studies have shown occasional evidence of increased physiological activity in patients with primary insomnia. We hypothesized that metabolic rate, as measured by overall oxygen use (VO2), might be a more general index of increased physiological activity. An initial experiment found elevated VO2 both at night and during the day in patients with primary insomnia as compared with matched normal sleepers. A second experiment found significant but more modest increases in VO2 in patients with Sleep State Misperception Insomnia [who complain of poor sleep but who had normal sleep by electroencephalographic (EEG) criteria]. In a third experiment, normal young adults were given caffeine 400 mg three times per day (TID) for 1 week as a means of increasing VO2 and possibly producing other symptoms of insomnia. Participants developed many symptoms consistent with those seen in patients with primary insomnia (poor sleep, increased latency on the Multiple Sleep Latency Test, increasing fatigue despite physiological activation, and increased anxiety on the Minnesota Multiphasic Personality Inventory (MMPI)). In a final experiment, physiological arousal was again produced by caffeine to determine if sleep with elevated arousal would be less restorative. All subjects (Ss) slept for 3.5 h after being given 400 mg of caffeine. During 41 h of sleep deprivation that followed, there was no significant condition difference for the Multiple Sleep Latency Test or mood measures. The results provided only weak support for the idea that sleep is less restorative after physiological arousal.