Differences in sun time within the same time zone affect sleep–wake and social rhythms,but not morningness preference: Findings from a Polish

Abstract

Morningness–eveningness describes individual preferences for activity at specified times of the day. The present research aimed to test whether sun time entrains humans and whether this effect is observable in sleep–wake timing, in the timing of social rhythm and in morningness preference. Furthermore, we tested whether different reference points (activity expressed in standard time or in sun time, morningness preference scores) provide concordant results about differences in chronotype according to longitude. University students were tested in two locations (Warsaw, Poland; Heidelberg, Germany) positioned within the same time zone but differing according to longitude, thus daylight appeared earlier in the east (Warsaw) than in the west (Heidelberg). Sampling was scheduled to obtain similar photoperiods and other environmental factors in two locations. Measures consisted of times of day when various activities occurred (e.g. going to bed, waking up, going outdoors) in the seven days prior to data collection, morningness preference, and depressiveness. Varsovians and Heidelbergers did not differ in morningness preference and depressiveness, but Varsovians, compared to Heidelbergers, undertook a number of activities (e.g. wake up, get up, having breakfast, first contact with another person and going outdoors) at an earlier clock time (21–38 min earlier, depending on activity), did not differ in starting and finishing classes/work, and were later in eating lunch and dinner. However, all the activities of Varsovians (except for going outdoors) were positioned later according to sun time reference. Thus, residents from east as compared to those from the west had similar morningness preference, were more morning positioned in many aspects according to standard time, but were later according to sun time. Results indicated university students entrained to sun time to some extent, and morning activities more coupled to sun time.

Keywords

Solar time standard time social time chronotype eveningness sleep timing entrainment Zeitgeber cross-culture depressiveness

Introduction

Chronotype, also termed morningness, describes preferences for activity at specified times of the day and is determined by a variety of factors, both endogenous and exogenous (Adan et al., 2012). This preference for morning or evening hours is partly determined by genes (Barclay et al., 2011; Osland et al., 2011) and early/late endogenous circadian rhythms are reflected in different diurnal profiles of hormones secretion and other physiological variables. For example, morningness is related to a phase advance in circadian rhythms of melatonin and body temperature (Duffy et al., 1999), and in the morning, evening-oriented individuals have a later and smaller peak of cortisol (Bailey and Heitkemper, 2001; Randler and Schaal, 2010) and higher levels of testosterone (Randler et al., 2012a) when compared with morning types. Moreover, diurnal profiles of evening types revealed higher heart rate, systolic blood pressure (Roeser et al., 2012) and pain sensitivity (Jankowski, 2013a) when compared with morning types. Furthermore, evening and morning oriented individuals differ in a number of psychological characteristics, for example, eveningness has been related to greater aggression-hostility and sensation seeking (Muro et al., 2011), impulsivity (Selvi et al., 2011), lower endurance (Jankowski 2012a, 2014a), higher scores in inductive reasoning (Díaz-Morales and Escribano, 2013), less restricted sociosexuality (Jankowski et al., 2014), and lower life satisfaction (Díaz-Morales et al., 2013; Jankowski, 2012b).

There are recognised environmental factors determining human chronotype, but their impact is rather complex. A number of possible determinants have been revealed in between-country comparisons which suggest temperature, climate and photoperiod (its length and position within a day) play a role. For example, Indian university students had higher morningness scores than Italians, what led Tonetti et al. (2012) to the conclusion that higher temperatures enforce morningness. The relationship between morningness and warmer temperatures has also been suggested by Lehmann et al. (2012) in an animal model and by Smith et al. (2002), who compared university students from six countries and showed that those from warmer areas (i.e., Colombia, Spain and India) were more morning oriented than those from colder climates (USA, Great Britain and the Netherlands). Furthermore, climate zone (temperate, subtropics and tropics) was found to significantly influence adolescents’ morningness, with adolescents in the subtropics being more evening oriented and those in the tropics more morning (Borchers and Randler, 2012; Randler, 2008b).

However, a climate is defined not only by temperature but also by latitude, and consequently the length of photoperiod. Namely, sunlight exposure changes from the equator to the poles – at the equator, day length is nearly identical throughout the year but towards the poles day length increases during the summer and decreases during the winter. Consequently, adolescents residing in the north within a given time zone (central Europe) were more morning oriented than those residing in the south, suggesting an effect of latitude which might be ascribed to a longer day length in the north during the study period (Randler, 2008b). Thus, day length is considered as a factor influencing chronotype (White and Terman, 2003). Calendar seasons, which is also related to day length (at least at locations distant from the equator), have been shown to influence sleep timing. Namely, individuals tend to shift towards earlier sleep in seasons with greater day length (Friborg et al., 2012; Honma et al., 1992), though there is also contrary evidence, suggesting no relation between morningness and season (Park et al., 2007).

Still, greater day length is also related to earlier sunrise and later sunset, thus observation of greater morningness at locations with longer photoperiods could be ascribed to earlier exposure to sun light in the morning. Such supposition seems corroborated by research comparing morningness levels between countries located differently within an east–west dimension, but within the same time zone. Namely, Spanish students were more evening oriented than Italian (Adan and Natale, 2002) and German (Randler and Díaz-Morales, 2007). The latter comparison was replicated using adolescent samples (Díaz-Morales and Randler, 2008) and also implies dispute over the role of temperature/latitude, since Spain and Germany differ in this domain. However, comparisons within the same country and the same time zone revealed individuals residing in the east scoring higher on morningness than those in the west (Randler, 2008a), and Roenneberg et al. (2007b) showed that earlier circadian phase position, as measured by mid sleep time on free days, is linearly related to eastern location, strengthening the hypothesis that the human circadian clock entrains to sun time (sunlight occurs earlier in the east than in the west).

However, it is not quite clear how strongly people entrain to sun time. Firstly, it is not known whether or not differences in midpoint of sleep are concordant with differences in timing of social rhythm and circadian preference in residents from different longitudes. Secondly, if humans entrain entirely to sun time, then residents from different longitudes should not differ in their circadian phase position when it is expressed in sun time (as opposed to typically used standard time reference). Thirdly, if humans entrain to sun time, then it seems suitable to express circadian phase position in sun time reference to determine which residents are actually earlier/later.

Thus, we aimed to clarify the above issues presumably in the most controlled research setting possible to obtain for morningness field studies, and to test whether the ‘entrainment to sun time’ hypothesis holds true in comparisons between countries. In our opinion, previous studies have not provided clear answers to the stated questions, because they used diverse data from different sampling periods, e.g. data stemming from different seasons and different climate zones, resulting in various photoperiods, and often from different time zones. Here, we used a strict sampling to reduce the influence of different day length and to minimise environmental effects (climate and temperature), and to focus on differences in sun time, within the same time zone. Moreover, we measured simultaneously morningness preference and clock times of various activities, whereas previous studies typically used a single marker (morningness preference or mid sleep time). Further, we defined formal circumstances that have to occur to conclude that the human circadian clock entrains predominantly to sun time:

If people entrain predominantly to sun time, then in an eastern location, with a 1 h earlier phase of the sun, time of activities expressed in standard time should occur 1 h earlier than in a western location (but there should be no differences between the two locations in time of activities expressed in sun time).

If people do not entrain sun time, but follow standard time (‘synchronization’ to social time), then in an eastern location, with a 1 h earlier phase of the sun, time of activities expressed in standard time should be the same as in a western location (but time of activities expressed in sun time should occur 1 h later than in a western location).

A remark should be made here, that standard time is simply a clock time (base point of 00:00 refers to calendar midnight). Sun time is amount of time (in hh:mm) from a base point of 00:00 referring to ‘true midnight’ (mid dark phase; halfway between sunset and sunrise).

We also aimed to test whether one of the psychological characteristics most consistently ascribed to differences between morning and evening individuals – depressive mood (Jankowski 2014a; Levandovski et al., 2011; Merikanto et al., 2013; Randler et al., 2012b) – is related to the differences in circadian phase position according to sun time. Namely, an epidemiological study suggests that shorter time interval between sunrise and arising times is linked to lower depression prevalence (Olders, 2003), whereas other studies show that possibly immediate exposure after awakening to possibly bright light increases mood, particularly during autumn–winter seasons (Terman and McMahan, 2012), what implies that better mood occurs if time interval between sunrise and arising times is longer (the brightest sunlight occurs around noon).

Materials and methods

Questionnaires/measurement instruments

The timing of sleep–wake rhythm and social rhythm were measured with a self-report time table questionnaire (e.g., Leonhard and Randler, 2009). The timetable consisted of questions about time (hh:mm) of waking up (whether with an alarm clock or not), getting up, eating breakfast, first contact with another person, going outside for the first time, beginning classes/work, eating lunch, finishing classes/work, eating dinner, going to bed, time needed to fall asleep and whether it was a free day. The above questions referred to the most recent seven days. The questions about sleep timing were based on the results of Roenneberg et al. (2003), who showed that actual sleep–wake timing can be approximated by self-report of the most recent past, whereas questions about social activities were delivered from the Social Rhythm Metric (Monk et al., 1990). Self-reports for separate days enable precise quantification of the timing of various activities during the most recent past, and can reveal patterns of presumably irregular week schedules in university students.

Morningness preference was measured using the Composite Scale of Morningness (CSM; Smith et al., 1989) in Polish (Jankowski, 2014b) and German (Randler, 2008c). It is a 13-item questionnaire with Likert-type response format with scores ranging from 13 (extreme eveningness) to 55 (extreme morningness). Cronbach’s α for CSM in the present sample was 0.87 and 0.85 for the German and the Polish sample, respectively.

Depressiveness was measured using the Center for Epidemiologic Studies Depression inventory (CES-D; Cole et al., 2004). We used 10 items constituting a short form of CES-D, which is a depression screening tool for the general population. The CES-D asks about the frequency of occurrence of depression symptoms within the most recent week. The items score from 0 (rarely/none) to 3 (most of the time), and the scale ranges from 0 to 30 with higher values indicating severe depressiveness. Cronbach’s α for CES-D in the present sample was 0.76 and 0.83 in the German and the Polish sample, respectively.

The following demographic variables were assessed: household size (number of adults older than 18 and number of children), religion (Roman Catholic, Protestant, atheist and other) civil status (paired and living together, paired and living apart and single), weekly workload in addition to studies during semester and during semester break, monthly net earnings and number of cigarettes smoked per day.

Participants

Participants were university students studying in Warsaw or Heidelberg. Polish participants were studying at the University of Warsaw (teacher education of foreign languages, psychology) and The Maria Grzegorzewska Academy of Special Education in Warsaw (teacher education, sociology), whereas German participants were studying at the University of Education in Heidelberg (teacher education). Amongst students from Warsaw (N = 291) and Heidelberg (N = 279), females were 82.8% and 77.8%, respectively, thus gender did not differ between the two countries (χ²= 2.29, p = 0.130). Participants from Warsaw were slightly younger (M = 20.81, SD = 1.73) than those from Heidelberg (M = 21.73, SD = 2.01) (t = 5.75, p < 0.001), however, the age range of 18–28 was the same in both countries. Furthermore, Polish students were mainly Roman Catholics (73.3%), whereas Germans were equally Roman Catholics (38.5%) or Protestants (37.4%), thus, the two samples differed in religion (χ²= 124.92, p < 0.001). Amongst Polish students, when compared to German, there was a similar representation of individuals paired but not living together (35.4% vs. 32.4%, respectively) or single (55.3% vs. 48.9%, respectively), and some under-representation of individuals living together (9.2% vs. 18.3%, respectively) amongst Poles, as compared to Germans (χ²= 112.17, p < 0.05). In Polish households, as compared to German ones, the number of children was lower (M = 0.02, SD = 0.22 vs. M = 0.29, SD = 0.75, respectively; t = 5.60, p < 0.001), but the number of adults was similar (M = 3.01, SD = 1.13 vs. M = 3.10, SD = 1.87, respectively; t = 0.64, p = 0.520). Furthermore, Polish students, as compared to Germans, were working a similar number of hours per week during the semester break (M = 10.16, SD = 18.51 vs. M = 12.85, SD = 12.67, respectively; t = 1.90, p = 0.058) and during the semester (M = 7.03, SD = 10.52 vs. M = 7.50, SD = 7.75, respectively; t = 0.59, p = 0.555), and had less money to spend per month (Me = €150–299 vs. Me = €300–499, respectively; U = 17352.5, p < 0.001). Polish and German students did not differ in the number of cigarettes they smoked daily (M = 1.38, SD = 3.93 vs. M = 1.62, SD = 5.49, respectively; t = 0.60, p = 0.552).

Procedure and variables

Sampling in Warsaw (52.2°N, 21.0°E) and in Heidelberg (49.4°N, 8.7°E) was scheduled with a one-week interval to achieve a nearly similar photoperiod, to exclude effects of season and effects of transition from daylight saving time into standard time (Schneider and Randler, 2009). Thus, students in Warsaw were tested between 16 and 22 of November 2012, whereas sampling in Heidelberg took place between 26 and 28 of November 2012. Average time of sunrise, sunset, mid-dark and duration of photoperiod in Warsaw and Heidelberg for the days of the self-report time table questionnaire were respectively 6 : 54 vs. 7 : 47, 15 : 35 vs. 16 : 34, 23 : 14 vs. 00 : 10 and 8 : 41 vs. 8 : 47, thus there was a 6 min difference in photoperiod and 56 min difference in sun time (mid-dark) between the two sampling sites. Participants were tested in groups before starting classes.

Data were processed using analysis of covariance (ANCOVA) with the country as a factor, and age, religion (1 = Roman Catholic vs. 0 = other religion), cohabitation (1 = living with vs. 0 = without a partner) and number of children in the household as covariates. The above covariates were entered to the analyses as they are factors possibly affecting circadian rhythms (e.g. age: Jankowski, 2014b; social contacts: Duffy et al., 1996), and they differed between the two samples. The variable ‘Money to spend’ was not entered as a covariate since the difference in this domain between the two countries seemed to be proportional to the differences in cost of living (NUMBEO, 2013). This is further backed up by the similar number of working hours in addition to studying at the university.

Dependent variables were sleep outcomes, time of activities, morningness preference and depressiveness. Sleep outcomes were as follows: sleep duration on free days, work days, and averaged for the whole week; time of sleep onset (bedtime + sleep latency) on free days, work days and averaged for the whole week; wake up time on free days, work days, and averaged for the whole week; mid sleep (half way between sleep onset time and wake up time) on working days (MSW), on free days (MSF), and on free days corrected for accumulated sleep debt during workweek (MSFsc; Roenneberg et al., 2004). MSF and particularly MSFsc (Roenneberg et al., 2004) are considered as an indicator of internal timing, whereas MSW is considered an indicator of circadian rhythm restricted by social obligations – later times indicate later circadian phase position. Consequently, social jetlag (Wittmann et al., 2006), an indicator of misalignment between biological and social time, was calculated as |MSF–MSW|, with greater values showing greater misalignment. The remaining sleep (latency, bedtime) and social rhythm variables were averaged across a week for comparisons between countries. In this study, free days were considered as those for which sleep was not restricted by an alarm clock, thus free days meant free to follow internal timing. Time of day was expressed in two ways: in standard time; and in sun time, which is distance in h:min from mid-dark phase (‘true midnight’). For example, when one got up at 7:00, then 7:00 is the standard time, but is 7:46 in sun time for Warsaw and 6:50 in sun time for Heidelberg.

Considered (and reported as statistically significant) are results performing two conditions: p < 0.05 and effect size (η_p² – partial eta squared) at least 0.01 (according to the guidelines of Cohen, 1992). Partial eta squared indicates proportion of variability in a dependent variable ascribed to a given factor, and if multiplied by 100, one can get a percentage of a common variability.

Results

ANCOVA analyses revealed a number of differences in studied variables (Table 1). Referring to standard time, Varsovians compared to Heidelbergers woke up earlier on work days by 21 min, with similar differences on free days. Moreover, Varsovians got up 30 min earlier, ate breakfast 28 min earlier, had first contact with another person 32 min earlier, and went outside 38 min earlier than Heidelbergers. But then, Varsovians started and finished classes/work at a similar time as Heidelbergers, but were later in eating lunch by 1 h 53 min and dinner by 1 h 44 min. Students from both cities generally went to bed at a similar time and needed a similar timespan to fall asleep. However, considering free and work days separately, Poles compared to Germans fell asleep 17 min later on work days and similarly on free days. Furthermore, in Varsovians, sleep duration during work days was shorter by 57 min than with their German counterparts, but there was no difference on free days. Taking into account sleep duration, sleep phase on free days (MSFsc) in Varsovians occurred 27 min earlier than in students from Heidelberg, but no difference was observed for mid sleep on work days. However, participants from both samples exhibited similar social jetlag.

Table 1.

Estimated means (SE) of dependent variables expressed in standard time and in sun time in students from Warsaw and Heidelberg compared using ANCOVA.

Dependent variable	Time reference	Poland N = 291	Germany N = 279	Effect size of location (η_p2)	Significant covariates (η_p2)
MSFsc	Standard	4:39 (0:07)	5:06 (0:07)	0.016*
MSFsc	Sun	5:25 (0:07)	4:56 (0:07)	0.019**
MSF	Standard	5:16 (0:06)	5:24 (0:06)	0.002
MSF	Sun	6:02 (0:06)	5:14 (0:06)	0.063***
MSW	Standard	3:51 (0:03)	3:53 (0:03)	0.000	−R (0.017**)
MSW	Sun	4:37 (0:03)	3:43 (0:03)	0.178***	−R (0.017**)
Sleep onset free	Standard	0:43 (0:07)	0:54 (0:08)	0.002
Sleep onset free	Sun	1:29 (0:07)	0:44 (0:08)	0.036***
Sleep onset work	Standard	0:11 (0:04)	23:54 (0:04)	0.015**	−R (0.015**)
Sleep onset work	Sun	0:57 (0:04)	23:44 (0:04)	0.210***	−R (0.015**)
Sleep onset week	Standard	0:19 (0:04)	0:09 (0:04)	0.005	−R (0.011*)
Sleep onset week	Sun	1:05 (0:04)	23:59 (0:04)	0.179***	−R (0.011*)
Wake up free	Standard	9:49 (0:06)	9:53 (0:06)	0.000
Wake up free	Sun	10:35 (0:06)	9:43 (0:06)	0.064***
Wake up work	Standard	7:31 (0:03)	7:52 (0:04)	0.027***	−R(0.011*)
Wake up work	Sun	8:17 (0:03)	7:42 (0:04)	0.067***	−R(0.011*)
Wake up week	Standard	8:07 (0:03)	8:28 (0:04)	0.025***
Wake up week	Sun	8:53 (0:03)	8:18 (0:04)	0.065***
Get up week	Standard	8:15 (0:03)	8:45 (0:04)	0.043***
Get up week	Sun	9:01 (0:03)	8:35 (0:04)	0.033***
Breakfast week	Standard	8:56 (0:04)	9:24 (0:05)	0.027***
Breakfast week	Sun	9:42 (0:04)	9:14 (0:05)	0.024**
Contact week	Standard	8:54 (0:05)	9:26 (0:05)	0.029***
Contact week	Sun	9:40 (0:05)	9:16 (0:05)	0.015**
Outdoors week	Standard	9:50 (0:05)	10:28 (0:05)	0.038***	−Ch (0.013*)
Outdoors week	Sun	10:36 (0:05)	10:18 (0:05)	0.009	−Ch (0.013*)
Start Classes/work week	Standard	10:19 (0:04)	10:35 (0:05)	0.008
Start Classes/work week	Sun	11:05 (0:04)	10:25 (0:05)	0.048***
Lunch week	Standard	15:34 (0:04)	13:41 (0:04)	0.356***
Lunch week	Sun	16:20 (0:04)	13:31 (0:04)	0.553***
Finish Classes/work week	Standard	16:29 (0:08)	16:51 (0:07)	0.008
Finish Classes/work week	Sun	17:15 (0:08)	16:41 (0:07)	0.019**
Dinner week	Standard	20:02 (0:04)	19:14 (0:04)	0.102***	−R (0.015**)
Dinner week	Sun	20:48 (0:04)	19:04 (0:04)	0.348***	−R (0.015**)
Bedtime week	Standard	24:02 (0:03)	23:49 (0:04)	0.008	−R (0.012*)
Bedtime week	Sun	24:48 (0:03)	23:39 (0:04)	0.191***	−R (0.012*)
Sleep latency week		16.97 (1.02)	20.12 (1.09)	0.007
Social jetlag		1:34 (0:04)	1:40 (0:05)	0.002
Sleep duration week		7:48 (0:03)	8:19 (0:04)	0.050***	C (0.011*)
Sleep duration free		9:07 (0:07)	8:58 (0:07)	0.001
Sleep duration work		7:18 (0:04)	8:15 (0:04)	0.068***
CSM		32.35 (0.40)	33.55 (0.44)	0.007
CES-D 10		9.14 (0.32)	8.58 (0.35)	0.002	−A (0.014**)

Note: ***p < 0.001, **p < 0.01, *p < 0.05 (for η_p2 ≥ 0.01); Covariates: R: religion (1 Roman Catholic; 0 other); Ch: children; C: cohabitation with a partner (1 yes; 0 no); A: age. Sign in front of covariate indicates direction of relationship with dependent variable. MSF: mid sleep on free days; MSFsc: mid sleep on free days sleep corrected; MSW: mid sleep on work days; CSM: morningness preference; CES-D: depressiveness.

When circadian activity was expressed in sun time, a reverse pattern of differences appeared. Namely, according to sun time, Varsovians were waking up later on work days by 35 min and on free days by 52 min. Similarly, they got up 26 min later, ate breakfast 28 min later, had first contact with another person 24 min later, but went outside at a similar sun time as Heidelbergers. Then, Poles were undertaking the remaining activities in later sun time than Germans did; they started classes/work later by 40 min and finished later by 34 min, further they ate lunch later by 2 h 49 min, dinner by 1 h 44 min and went to bed later by 1 h 9 min. Consequently, Poles had later sleep onset on free days by 45 min and on working days by 1 h 13 min, what resulted in a later sleep timing on free days (MSFsc: 29 min) and on work days (MSW: 54 min).

Comparing timing of circadian functioning according to either of the two time frames (standard time vs. sun time), one may notice that in all the activities (except for going outdoors), when referring to sun time, Poles were later than Germans. On the other hand, using standard time as a reference, Poles compared to Germans were earlier in a number of activities.

Furthermore, Poles and Germans differed neither in CSM scores nor in depressiveness (CESD-10). Depressiveness was related only to the younger age in our participants.

Interesting results were obtained considering covariates (Table 1). Namely, it appeared that Roman Catholics as compared to others (mainly Protestants and atheists) were earlier on working days not only in waking up, falling asleep and mid sleep but also generally earlier in eating dinner and going to bed. Furthermore, having children was related to leaving home earlier in the morning. Having a relationship contributed to longer sleep duration across the week. These covariates explained over 1% of variance in the timing of abovementioned activities, thus their effects were small, but seem to have practical meaning (Cohen, 1992).

Discussion

The main aim of the study was to test whether suntime entrains the human circadian clock, and whether it is observable in sleep–wake and social rhythms, and in morningness preference. The study also aimed to test whether different reference points (morningness preference, circadian phase position expressed in standard time or in sun time) provide similar conclusions about chronotype of residents form different longitudes. Prior to the study, we established formal criteria that the results should meet to quantify the impact of sun time on circadian entrainment. If sun time does not play a role, then Varsovians’ and Heidelbergers' activities should occur at the same clock time. On the other hand, in sun time reference, Varsovians should be later than the residents of Heidelberg, since time of activities would be more distant (later) from the previous mid dark point (‘true midnight’), because in Warsaw mid dark occurred 56 min earlier than in Heidelberg.

The results showed that Varsovians and Heidelbergers did not differ in their morningness preference, as indicated by CSM scores. Further, the participants were partly entrained to sun time, as Varsovians compared to Heidelbergers were earlier in a number of activities when their timing was expressed in standard time, but not as much earlier as by 56 min expected if absolute entrainment to sun time occurred. Namely, Varsovians were generally earlier by 21–38 min (depending on circadian facet) than Heidelberger students when referring to standard time (except for lunch, dinner and bedtime). Precisely, they were earlier in MSFsc by 27 min, getting up by 30 min, having breakfast by 28 min, first contact with another person by 32 min, going outdoors by 38 min. The above results suggest that not only sleep timing but also social rhythm is partly synchronized with sun time, and according to our criteria particularly going outdoors and first contact with another person are more coupled to sun time than MSFsc. Interestingly, these activities are concentrated around morning hours, thus it is possible that starting a day is more affected by sun time than later activities. Namely, the two locations did not differ in clock time of starting and finishing classes/work, while times of eating lunch and dinner were later in Warsaw – thus, activities positioned further during a day seem to be more affected by social factors. These results are interesting, considering implications derived from phase response curve for light (Rosenthal et al., 1990), which shows importance of bright light exposure in the morning (e.g. sun light via going outdoors) in shortening tau, naturally longer than 24 h (Czeisler et al., 1999). Thus, the reason why the time of going outdoors is the most strongly coupled to sun time might be that it helps to reset circiadian clock to 24 h.

One may argue that all the activities we analysed are not real facets of the circadian rhythm, thus we consider MSFsc in more detail, as this marker has been used in a previous study (Roenneberg et al., 2007b). In our research, Varsovians’ MSFsc was earlier by 27 min when expressed in standard time, but later by 29 min when expressed in sun time. Furthermore, effect size for between-locations difference in MSFsc expressed in standard time was minimally smaller (η_p²= 0.016) than in MSFsc expressed in sun time (η_p²= 0.019). According to our criteria, such results suggest that the human circadian clock entrains to sun time but not completely, at least in early adulthood. Besides work/education schedules, a number of factors potentially disturbing entrainment to sun time could be considered, such as social contacts (Duffy et al., 1996), family relationships (Díaz-Morales et al., 2014), or electronic media use (Cain and Gradisar, 2010).

Moreover, the present results corroborate the observation that eastern people have earlier circadian phase position as indicated by MSFsc (Roenneberg et al., 2007b), and further add a conclusion that such dependency is present not only in comparisons within the same country but also when comparing different countries. Thus, it might be a rule that within a given time zone inhabitants living in the east have an earlier actual chronotype irrespective of national borders. Interestingly, the present study showed an earlier circadian phase position (expressed in standard time) in residents of eastern locations regarding not only MSFsc but also using other facets of circadian rhythmicity. Namely, this is the first study showing that eastern inhabitants, as compared to western ones, are earlier in other ways – that means social manifestations of circadian rhythm (Monk et al., 1990), like having breakfast, first contact with another person and going outside for the first time during the day.

Comparison of morningness levels between different geographic locations has been done typically using morningness preference questionnaires (Randler, 2008a,b). Considering this measure, the present results show that Polish and German students do not differ in morningness preference (as indicated by CSM), though the CSM scores are related to the timing of studied activities (Randler and Jankowski, 2014). Thus, it is possible that different longitude within the same time zone does not affect morningness preference in adulthood. This is in accordance with a previous study showing greater morningness in eastern habitancy in adolescents, which then disappears at the age of 17–18 years (Díaz-Morales and Randler, 2008). Consequently, it is likely that young adults entrain less to sun time in favor of adherence to social time (e.g. to follow the social demands of their peers). This issue should be assessed in future research in older adults, to find out whether it is only the stage of early adulthood when humans entrain less to sun time, or if such a trend is stable from puberty.

Remarkably, when circadian phase position was expressed in sun time, Varsovians, compared to Heidelbergers, were later in all the activities (except for going outdoors) by 28–54 min (lunch and dinner are not considered here, as they occur red later in Varsovians in both time references). Thus, the fact that eastern individuals were earlier when their activity was expressed in standard time, later when it was expressed in sun time, and similar in their morningness preference (CSM) prompts an important question in understanding chronotype – whether more morning-oriented individuals are those who sleep earlier according to clock hours or maybe those who function during the earlier phase of sunlight, or alternatively those who more prefer morning hours for activity.

No difference in depressiveness between the two countries was found. However, this result is not surprising taking into account that we found also no difference either in morningness preference (CSM scores), which is known to be related to depressiveness (Randler et al., 2012b), or in social jetlag, hypothesised to be a mediator between current circadian phase position and depression (Levandovski et al., 2011). On the other hand, this result may also suggest that depressiveness is related to some intrinsic aspect of morningness, for example temperament, as it has been previously suggested (Jankowski, 2014a), instead of to the circadian phase position against sun time or social time. The above conclusion seems strengthened by the observation that no differences in mood occurred despite the timing of presumed exposure to morning sunlight was more advantageous in Varsovians than Heidelbergers. Namely, Varsovians, compared to Heidelbergers, were going outdoors more immediately after waking up (1 h 43 min vs. 2 h 0 min, respectively) and at time of day when sunlight was brighter (2 h 56 min vs. 2 h 41 min after sunrise, respectively).

Roenneberg et al. (2007b) found that entrainment to sun time depends on city size, with residents of cities with under 300,000 inhabitants showing greater entrainment to sun time, whereas those residing in cities with over 300,000 inhabitants showed only partial entrainment to sun time. The locations we sampled differ in size, with Heidelberg having 150,000 inhabitants and Warsaw 1,710,000 inhabitants, but results we obtained are concordant with those for big cities (Roenneberg et al., 2007b). Thus, we argue that effect of city size is due to specificity of populations living in big cities vs. small towns, so that people differ in occupations, work schedules etc. As our samples were homogeneous – students of social science not differing in demographic characteristics, while some existing differences were controlled for in the analyses by the ANCOVA techniques, and nonetheless we obtained results as for big cities, we argue that it is not city size per se which is a factor influencing entrainment to sun time (e.g. skyscrapers limiting light exposure), but rather a lifestyle of the studied population. Thus, our results seem limited to young adults relatively free to follow their preferred work schedule, as students both in Heidelberg and Warsaw had a lot of freedom in choosing the time of a day for their classes. This limitation also implies that further research should consider a number of separate comparisons between individuals showing similar lifestyle (office workers, high school pupils, etc.), rather than city size per se, in order to distinguish which groups entrain more or less to sun time. As another limitation, this study did not measure individual light exposure and thus cannot reveal if the daily light and dark variation actually differed between the two samples in the pattern presumed by sun time. Moreover, the circadian clock was not measured by physiological means like hormone or temperature profiles at different times of day, which could have given better evidence for the entrainment by sun time.

A more precise distinction of social groups weakly entrained to sun time seems increasingly important, as low coupling to sun time is considered a risk factor for a shift towards later circadian phase position (Roenneberg et al., 2007b). Namely, individuals less exposed to sun light experience too weak a Zeitgeber to entrain their internal time period to 24 h, as human tau is typically longer than 24 h (Czeisler et al., 1999). The results of the present research also confirm the above supposition – the studied population was only partially entrained to sun time, and showed late circadian phase position when compared to MSF and MSFsc distributions in a large sample (Roenneberg et al., 2007a). Considering late chronotype may predispose to a lowered well-being, education programmes promoting self-regulation of circadian functioning by its adjustment to sun time would be helpful. Our results show university students as potential recipients of such programmes, while the method we used indicates a possible way of testing sun time dependency in other subpopulations in future research.

Footnotes

Acknowledgement

The study was supported by a grant 2011/03/D/HS6/05760 from the National Science Centre, Poland.

References

Adan

Archer

Hidalgo

(2012) Circadian typology: A comprehensive review. Chronobiology International 29(9): 1153–1175.

Adan

Natale

(2002) Gender differences in morningness–eveningness preference. Chronobiology International 19(4): 709–720.

Bailey

Heitkemper

(2001) Circadian rhythmicity of cortisol and body temperature: Morningness–eveningness effects. Chronobiology International 18(2): 249–261.

Barclay

Eley

Mill

(2011) Sleep quality and diurnal preference in a sample of young adults: Associations with 5HTTLPR PER3 and CLOCK 3111. American Journal of Medical Genetics Part B 156(6): 681–690.

Borchers

Randler

(2012) Sleep–wake cycle of adolescents in Côte d'Ivoire: Influence of age gender religion and occupation. Chronobiology International 29(10): 1366–1375.

Cain

Gradisar

(2010) Electronic media use and sleep in school-aged children and adolescents: A review. Sleep Medicine 11(8): 735–742.

Cohen

(1992) A power primer. Psychological Bulletin 112(1): 155–159.

Cole

Rabin

Smith

(2004) Development and validation of a Rasch-derived CES-D short form. Psychological Assessment 16(4): 360–372.

Czeisler

Duffy

Shanahan

(1999) Stability precision and near-24-hour period of the human circadian pacemaker. Science 284(5423): 2177–2181.

10.

Díaz-Morales

Escribano

Jankowski

(2014) Evening adolescents: The role of family relationships and pubertal development. Journal of Adolescence 37(4): 425–432.

11.

Díaz-Morales

Jankowski

Vollmer

(2013) Morningness and life satisfaction: Further evidence from Spain. Chronobiology International 30(10): 1283–1285.

12.

Díaz-Morales

Randler

(2008) Morningness–eveningness among German and Spanish adolescents 12–18 years. European Psychologist 13(3): 214–221.

13.

Duffy

Dijk

Hall

(1999) Relationship of endogenous circadian melatonin and temperature rhythms to self-reported preference for morning or evening activity in young and older people. Journal of Investigative Medicine 47(3): 141–150.

14.

Duffy

Kronauer

Czeisler

(1996) Phase-shifting human circadian rhythms: Influence of sleep timing, social contact and light exposure. Journal of Physiology 495(1): 289–297.

15.

Friborg

Bjorvatn

Amponsah

(2012) Associations between seasonal variations in day length (photoperiod) sleep timing sleep quality and mood: A comparison between Ghana (5°) and Norway (69°). Journal of Sleep Research 21(2): 176–184.

16.

Honma

Kohsaka

(1992) Seasonal variation in the human circadian rhythm: dissociation between sleep and temperature rhythm. American Journal of Physiology 262(5 pt 2): R885–R891.

17.

Jankowski

(2012a) Morningness–eveningness and temperament: The Regulative Theory of Temperament perspective. Personality and Individual Differences 53(6): 734–739.

18.

Jankowski

(2012b) Morningness/eveningness and satisfaction with life in a Polish sample. Chronobiology International 29(6): 780–785.

19.

Jankowski

(2013a) Morning types are less sensitive to pain than evening types all day long. European Journal of Pain 17(7): 1068–1073.

20.

Jankowski

(2014a) The role of temperament in relationship between morningness-eveningness and mood. Chronobiology International 31(1): 114–122.

21.

Jankowski KS (2014b) Composite Scale of Morningness: Psychometric properties, validity with Munich ChronoType Questionnaire and age/sex differences in Poland. European Psychiatry. DOI: 10.1016/j.eurpsy.2014.01.004.

22.

Lehmann

Spoelstra

Visser

(2012) Effects of temperature on circadian clock and chronotype: an experimental study on a passerine bird. Chronobiology International 29(8): 1062–1071.

23.

Leonhard

Randler

(2009) In sync with the family: Children and partners Influence the sleep-wake circadian rhythm and social habits of women. Chronobiology International 26(3): 510–525.

24.

Levandovski

Dantas

Fernandes

(2011) Depression scores associate with chronotype and social jetlag in a rural population. Chronobiology International 28(9): 771–778.

25.

Merikanto

Lahti

Kronholm

(2013) Evening types are prone to depression. Chronobiology International 30(5): 719–725.

26.

Monk

Flaherty

Frank

(1990) The Social Rhythm Metric. An instrument to quantify the daily rhythms of life. Journal of Nervous and Mental Disease 178(2): 120–126.

27.

Muro

Gomà-i-Freixanet

Adan

(2011) Circadian typology age and the Alternative Five-Factor Personality Model in an adult women sample. Chronobiology International 28(8): 690–696.

28.

NUMBEO (2013) Cost of living comparison between Heidelberg and Warsaw. Available at: www.numbeo.com/cost-of-living (accessed 9 July 2013).

29.

Olders

(2003) Average sunrise time predicts depression prevalence. Journal of Psychosomatic Research 55(2): 99–105.

30.

Osland

Bjorvatn

Steen

(2011) Association study of a variable-number tandem repeat polymorphism in the Clock Gene PERIOD3 and chronotype in Norwegian university students. Chronobiology International 28(9): 764–770.

31.

Park

Kripke

Cole

(2007) More prominent reactivity in mood than activity and sleep induced by differential light exposure due to seasonal and local differences. Chronobiology International 24(5): 905–920.

32.

Randler

(2008a) Differences in sleep and circadian preference between eastern and western German adolescents. Chronobiology International 25(4): 565–575.

33.

Randler

(2008b) Morningness–eveningness comparison in adolescents from different countries around the world. Chronobiology International 25(6): 1017–1028.

34.

Randler

(2008c) Psychometric properties of the German version of the Composite Scale of Morningness. Biological Rhythm Research 39(2): 151–161.

35.

Randler

Díaz-Morales

(2007) Morningness in German and Spanish students: A comparative study. European Journal of Personality 21(4): 419–427.

36.

Randler C and Jankowski KS (2014) Evidence for the validity of the Composite Scale of Morningness based on students from Germany and Poland – relationship with sleep wake and social schedules. Biological Rhythm Research. DOI: 10.1080/09291016.2014.884307.

37.

Randler

Ebenhöh

Fischer

(2012a) Chronotype but not sleep length is related to salivary testosterone in young adult men. Psychoneuroendocrinology 37(10): 1740–1744.

38.

Randler

Schaal

(2010) Morningness–eveningness habitual sleep–wake variables and cortisol level. Biological Psychology 85(1): 14–18.

39.

Randler

Stadler

Vollmer

(2012b) Relationship between depressive symptoms and sleep duration/chronotype in women. Journal of Individual Differences 33(3): 186–191.

40.

Roenneberg

Kuehnle

Juda

(2007a) Epidemiology of the human circadian clock. Sleep Medicine Reviews 11(6): 429–438.

41.

Roenneberg

Kuehnle

Pramstaller

(2004) A marker for the end of adolescence. Current Biology 14(24): R1038–R1039.

42.

Roenneberg

Kumar

Merrow

(2007b) The human circadian clock entrains to sun time. Current Biology 17(2): R44–R45.

43.

Roenneberg

Wirz-Justice

Merrow

(2003) Life between clocks: Daily temporal patterns of human chronotypes. Journal of Biological Rhythms 18(1): 80–90.

44.

Roeser

Obergfell

Meule

(2012) Of larks and hearts – morningness/eveningness heart rate variability and cardiovascular stress response at different times of day. Physiology and Behaviour 106(2): 151–157.

45.

Rosenthal

Joseph-Vanderpool

Levendosky

(1990) Phase-shifting effects of bright morning light as treatment for delayed sleep phase syndrome. Sleep 13(4): 354–361.

46.

Schneider

Randler

(2009) Daytime sleepiness during transition into daylight saving time in adolescents: Are owls higher at risk? Sleep Medicine 10(9): 1047–1050.

47.

Selvi

Aydin

Atli

(2011) Chronotype differences in suicidal behavior and impulsivity among suicide attempters. Chronobiology International 28(2): 170–175.

48.

Smith

Folkard

Schmieder

(2002) Investigation of morning–evening orientation in six countries using the preferences scale. Personality and Individual Differences 32(6): 949–968.

49.

Smith

Reilly

Midkiff

(1989) Evaluation of three circadian rhythm questionnaires with suggestions for an improved measure of morningness. Journal of Applied Psychology 74(5): 728–738.

50.

Terman

McMahan

(2012) Chronotherapy, New York: Avery.

51.

Tonetti

Sahu

Natale

(2012) Circadian preference in Italy and India: A comparative study in young adults. Personality and Individual Differences 53(3): 355–358.

52.

White

Terman

(2003) Effect of iris pigmentation and latitude on chronotype and sleep timing. Chronobiology International 20(6): 1993–1995.

53.

Wittmann

Dinich

Merrow

(2006) Social jetlag: Misalignment of biological and social time. Chronobiology International 23(1–2): 497–509.

Differences in sun time within the same time zone affect sleep–wake and social rhythms,but not morningness preference: Findings from a Polish–German comparison study

Abstract

Keywords

Introduction

Materials and methods

Questionnaires/measurement instruments

Participants

Procedure and variables

Results

Discussion

Footnotes

Acknowledgement

References