摘要：In principle, two domains of temporal control in human behavior and cognition can be distinguished, i.e., long-term mechanisms as observed in circadian or circannual rhythms reflecting evolutionary adaptations to geophysical cycles, and self-organized mechanisms as mainly exhibited for short-term intervals in the sub-second to minutes domain. It is an open question whether the mechanisms of temporal control within and also across the temporal domains are inter-dependent, thus, reflecting one dominant temporal control machinery, or whether temporal control within and across the different temporal domains underlies independent operative principles. One way to get more insight in these problems is to manipulate environmental variables, and to investigate whether such manipulations show similar or distinct effects on temporal control; one class of such environmental variables are represented by social interactions.We describe results of an exploratory study in which insight into the problem of homogeneity or inhomogeneity in temporal processing could be gained. Four adult young male participants were isolated completely for three weeks without any access to external information; the participants did not have any knowledge about the time of day. The same participants had participated before this group experiment in an experiment with individual isolation. All participants showed during individual isolation “free-running” circadian rhythms different from 24hours. The question was in the group experiment whether the participants would socially synchronize, i.e., exhibit one circadian rhythm for the group. The rest-activity cycle was measured for each participant, and body temperature and kidney functions were monitored. In addition, participants during their subjective days had to perform several tests like the production of 10-s intervals, or they had to indicate their “personal tempo”, i.e., tapping with their fingers at a comfortable and self-selected speed.The results indicate interesting synchronizations and desynchronizations of functions. In the first half of the experiment, the rest- activity cycle of the four participants was synchronized with a circadian period of 26.2hrs. One participant, however, showed an important phase difference compared to the others starting his subjective day on average 2hrs earlier. The phase difference observed in this “critical participants” indicates that his circadian clock had a tendency for a shorter subjective day compared to the others. This could be proven by the circadian modulation of the vegetative functions as represented for instance by kidney functions. Most surprisingly, the circadian modulation of the kidney functions disappeared a few days after the beginning of the experiment, but reappeared spontaneously after several days. This suppression of the circadian modulation can be interpreted as a beat phenomenon as also observed in physical phenomena; when two near frequency sounds are superimposed they suppress each other in antiphase. The same mechanism can be assumed for our observation: The rest-activity-cycle of this participant was synchronized with the other group members, but this period was apparently too long for other physiological functions to be synchronized; they followed their own circadian cycle which was shorter. Thus, the social synchronization of the participant resulted in an internal desynchronization, which became observable through the beat phenomenon. Within this participant, two circadian clocks were running at the same time, one super-imposed by the group and one dominated by the individual circadian clock. A support for this interpretation is provided by the second half of the experiment when the “critical participant” no longer lived synchronized with the group but followed his own temporal schedule close to 24hrs, whereas the other three group members extended their circadian day in synchrony to even 27.2hrs. In this second half of the experiment, the “critical participant” no longer exhibited an internal desynchronization. To the best of our knowledge such a beat phenomenon with two superimposed circadian cycles has not been observed previously.Another unique result was obtained in experiments on temporal production. Before the experiment started each participant went through a learning phase to produce as accurately as possible a 10-s interval without counting. All participants reached within a few trials a reliable set point with only little variance. During isolation each participant selected within several days of transitions very different individual set points when producing the 10-s intervals: One participant produced the target interval as 6.5 s, another one as 8.5 s, a third one as 11 s, and the “critical participant” as 16 s. Thus, no correlation was observed between the circadian rhythms of these participants and their temporal production confirming a previous observation that temporal perception in the domain of seconds is unrelated to the circadian clock. Interestingly, the “critical participant” with the shortest circadian rhythm showed the longest temporal production supporting the above statement of independence.While the experiment on temporal production showed high inter-individual variance, a rather different observation was made when measuring the “personal tempo”. During the experiments of individual isolation, each participant had selected different tempi with intervals between successive taps of 0.2 and 0.8 s; for the “critical participant” the period length was on average 0.55 s for successive days. During the group experiment an interesting social synchronization was observed: Three participants including the “critical participant” selected a personal tempo with intervals close to 0.2 s; they preferred all the same speed. The participant who had selected the rather slow tempo of 0.8 s in the individual experiment changed his personal tempo in the group within several days to a value close to 0.4 s, i.e., a harmonic value of the group. Thus, a strong social effect on temporal control could be demonstrated.Taken together, the observations on the circadian modulation of functions and the personal tempo, and the data of temporal production strengthen on the one hand the concept of an independence of temporal mechanisms in different functional domains (Pöppel & Bao, 2013), but they also demonstrate the power of environmental variables as indexed by social factors which has been demonstrated also for temporal order threshold (Bao et al., 2013).
关键词：Circadian rhythm;Rest-activity cycle;Social synchronization;Desynchronization;Beat phenomenon;Temporal production;Personal tempo
