Determination of sleep and wakefulness with the actigraph data analysis software (ADAS).

Current evidence has shown that, overall, actigraphy is an excellent tool for unobtrusive documentation of sleep/wake activity in normal individuals. However, a number of methodological issues remain to be resolved to warrant its use in clinical research. In this paper, we report the results of a study aimed at the development of a new scoring software that can accurately identify sleep and wakefulness. Using total sleep time as an index of comparison, the software was optimized on a calibration sample and prospectively tested on a validation sample. A strong correlation coefficient (r = 0.93, p < 0.008), with an average discrepancy value of 10 minutes, was observed for the calibration sample. The application of the optimal software to the validation sample revealed an even higher correlation coefficient (r = 0.97, p < 0.0001), with an average discrepancy value of 12 minutes.

Circadian determinants of subjective alertness.

Four healthy male subjects each experienced a temporal isolation experiment lasting several months. During part of each experiment (2-5 weeks), the subject's entire imposed daily routine (including light-dark, rest-activity, and meal routines) was either stretched (two subjects: T = 25.8 hr, 26.0 hr) or reduced (two subjects: T = 22.8 hr, 23.1 hr) to beyond the range of entrainment of the endogenous circadian pacemaker (ECP), which then ran at a different period (tau). Subjective alertness was measured approximately three times per hour (during wakefulness), using a computerized visual analogue scale technique. Circadian rhythms in subjective alertness were then plotted both at tau, the period length of the ECP, and at T, the period length of the imposed sleep-wake cycle (SWC) and light-dark cycle. At tau, the alertness rhythm was closely in phase with the temperature rhythm. At T, the alertness rhythm showed an "inverted-U" function with a peak toward the middle of the subjective day, upon which was superimposed a "postlunch dip" for one subject. Thus, subjective alertness would appear to be under the control fo both ECP and SWC mechanisms, which combine to produce the composite time-of-day function normally observed in a diurnal setting.

Human growth hormone release is decreased during sleep in temporal isolation (free-running).

Growth hormone (hGH) secretion was measured during sleep in 10 healthy male subjects isolated from all time cues. HGH concentrations following sleep onset were compared between scheduled sleep episodes (entrainment) and self-selected sleep episodes (free-running). Peak sleep-related hGH values were decreased significantly during free-running. The duration of the first slow wave sleep (SWS) episode and the latency to the first REM sleep episode also decreased significantly during free-running. The latencies from sleep onset to SWS and to peak hGH secretion did not differ between entrainment and free-running. These results suggest that sleep-related hGH secretion begins 'on time' during free-running, but is terminated earlier. Thus, while sleep onset facilitates hGH release, the timing of other stages of sleep such as REM may alter the magnitude of sleep-related hGH secretion.

Diurnal variation in mood and performance in a time-isolated environment.

In order to document circadian rhythmicity in various psychological functions under the chronobiologically 'pure' condition of temporal isolation, a battery of mood and performance tests were administered about 6 times per day to a heterogeneous group of 18 subjects (ages 19-81, 5 female). Each subject spent about 5 days in temporal isolation, entrained to a routine equivalent to his/her own habitual sleep/wake cycle. Average time of day functions were obtained for the mood and performance variables, and compared to rectal temperature data subjected to exactly the same statistical analysis. Significant time of day effects were found in the mood variables of alertness, sleepiness, weariness, effort required, happiness and well-being. Times of 'best' mood were different from the time of peak temperature. Moreover, the minima of sleepiness, weariness and effort occurred earlier in the day than the maximum of alertness. Significant time of day effects were also found in the speed with which search and dexterity tasks were completed. Only the dexterity tasks showed a complete parallelism with the temperature rhythm.

Circadian rhythms in human performance efficiency under free-running conditions.

A series of four experiments is described in which two men and two women (age range 22-79 years) were allowed to 'free-run' under conditions of isolation from all time cues. The circadian rhythms of performance efficiency were studied for three tasks: serial search, verbal reasoning, and manual dexterity (left and right hand versions). Strong evidence was obtained to suggest that different tasks are under the control of different circadian oscillators. The circadian rhythm in verbal reasoning performance tends to run at a shorter circadian period than that in more simple repetitive tasks, and slight differences in circadian behavior can occur between left- and right-handed dexterity.

Task variables determine which biological clock controls circadian rhythms in human performance.

There are circadian (approximately 24 h) rhythms for a wide range of human physiological and psychological functions including mood and performance efficiency. These rhythms are self-sustaining in conditions of temporal isolation, indicating that internal oscillators (or biological clocks) control them. Recent research has proposed an endogenous two-oscillator model of the human circadian system, with one oscillator indicated by the core body temperature rhythm and a second oscillator responsible for the daily cycle of sleep and wakefulness. The present study was designed to produce a desynchronization in period between the two oscillators, to assess directly the behaviour of the rhythms of different performance tasks. The results, reported here, indicate that a simple manual dexterity task is almost entirely under the control of the temperature rhythm oscillator, whereas a more complex cognitive task demonstrates a periodicity which appears to be influenced by those oscillators controlling temperature and the sleep/wake cycle.