In order to determine the effects of disentrainment environment on the human sleep/wake cycles, somonographies were recorded during 72 hrs of social isolation on 7 subjects. During disentrainment subject were confined individually in an isolation chamber under a condition of no time cues, constant darkness and silence. They could sleep, move, eat, drink and use the toilet facilities at any time. Singing, humming or any other vocal activity was also permitted. Subjects were asked to press the micro switch, bandaged on the left palm, whenever they have an experiance of hallucination.
Two of 7 subjects gave up the experiment. One switched on the break out singnal at 7 hr 35 min and another one at 35 hr 24 min after the start of disentrainment condition. The data of those 2 subjects were omitted from present analysis.
The patterns of sleep and wakefulness were clearly modified during the disentrainment period. Polyphasic sleep patterns and hypersomnia tendencies were observed. Eightytwo percent of waking periods and 72 percent of sleep period were less than 4 hrs duration (Table 1). The distribution of sleep onset times did not differ across the 24 hr-day and the over all experimental period (Table 3, Fig. 1&2). Average total sleep time increased to 41.6 hrs, accounting 57.8 % of the 72 hrs. This hypersomnia tendency was remarkable in the first 24 hr-day and decreased gradually in following two 24 hr-day (Table 2, Fig. 2). An usual circadian pattern in the duration of sleep periods could be seen in the Day 3 (Fig.3). Even in the third 24 hr-day, the total sleep time still showed, however, a clear prolongation 48 % over the normal standard value (Williams et al 1974). These data showed that an ontogenetic regression of sleep-wake pattern was observed under the constant dark and silent environment.
To demonstrate the changes of hemispheric dynamics, EEG laterality index score (LIS) was culculated every 30 min from C3 and C4 power via FFT. The trend component of alpha or theta band LIS tended to increase (positive shift) after 30-40 hrs from the start of disentrainment condition (Fig-5). These positive shifts of LIS were observed in 3 of 5 Ss. Corresponding to this, the occurence of hallucination signals increased from the first 24 hr-day to the third 24 hr-day. Most of signals appeared in the later half of the experiment (Fig.6). These behavioral tendency suggest that the right hemispheric activation increases in the later half of disentrainment period. It might be corresponded to the positive shifts of LIS initiated when a subjects past a half way point. Psychophysiological aspects of LIS shifts were discussed.