Circadian rhythms determined by cosine curve fitting: Analysis of continuous work and sleep-loss data

Abstract

In this study, we report the effects of sleep loss upon circadian rhythm parameters analyzed by the cosine curve fitting (cosinor) method. Rhythm alterations are described as reductions in rhythm strength, increases in individual variations producing an increase in the 95% confidence limits, and reductions in rhythm amplitude. Subjects worked continuously at tasks for 45 h with time-of-day cues. Circadian cycles in physiological and mood variables remained intact, but rhythms in some task performance measures no longer showed significant 24-h/cycle activities. The relationship between oral temperature, mood, and pulse rhythms continued undisturbed during the continuous work period; however, the performance linkage to oral temperature was lost. These findings direct attention to individual difference in susceptibility to continuous work periods and suggest that 24-h rhythms in some performance and physiological measures perhaps are more readily responsive to an altered wake/sleep cycle than other circadian rhythms.