A Longitudinal Assessment of Sleep Timing, Circadian

Phase, and Phase Angle of Entrainment across Human

Adolescence
Stephanie J. Crowley1*, Eliza Van Reen2, Monique K. LeBourgeois3, Christine Acebo2, Leila Tarokh2,5,6,
Ronald Seifer2,4, David H. Barker2,4, Mary A. Carskadon2,7
1 Biological Rhythms Research Laboratory, Department of Behavioral Sciences, Rush University Medical Center, Chicago, IL, United States of America, 2 E.P. Bradley
Hospital Sleep Research Laboratory, Department of Psychiatry and Human Behavior, Warren Alpert Medical School of Brown University, Providence, RI, United States of
America, 3 Sleep and Development Laboratory, Department of Integrative Physiology, University of Colorado Boulder, Boulder, CO, United States of America, 4 The
Bradley Hasbro Children’s Research Center, Providence, RI, United States of America, 5 Institute of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland,
6 University Hospital of Child and Adolescent Psychiatry, University of Bern, Bern, Switzerland, 7 Centre for Sleep Research, University of South Australia, Adelaide,
Australia

Abstract
The aim of this descriptive analysis was to examine sleep timing, circadian phase, and phase angle of entrainment across
adolescence in a longitudinal study design. Ninety-four adolescents participated; 38 (21 boys) were 9–10 years (‘‘younger
cohort’’) and 56 (30 boys) were 15–16 years (‘‘older cohort’’) at the baseline assessment. Participants completed a baseline
and then follow-up assessments approximately every six months for 2.5 years. At each assessment, participants wore a wrist
actigraph for at least one week at home to measure self-selected sleep timing before salivary dim light melatonin onset
(DLMO) phase – a marker of the circadian timing system – was measured in the laboratory. Weekday and weekend sleep
onset and offset and weekend-weekday differences were derived from actigraphy. Phase angles were the time durations
from DLMO to weekday sleep onset and offset times. Each cohort showed later sleep onset (weekend and weekday), later
weekend sleep offset, and later DLMO with age. Weekday sleep offset shifted earlier with age in the younger cohort and
later in the older cohort after age 17. Weekend-weekday sleep offset differences increased with age in the younger cohort
and decreased in the older cohort after age 17. DLMO to sleep offset phase angle narrowed with age in the younger cohort
and became broader in the older cohort. The older cohort had a wider sleep onset phase angle compared to the younger
cohort; however, an age-related phase angle increase was seen in the younger cohort only. Individual differences were seen
in these developmental trajectories. This descriptive study indicated that circadian phase and self-selected sleep delayed
across adolescence, though school-day sleep offset advanced until no longer in high school, whereupon offset was later.
Phase angle changes are described as an interaction of developmental changes in sleep regulation interacting with
psychosocial factors (e.g., bedtime autonomy).

Citation: Crowley SJ, Van Reen E, LeBourgeois MK, Acebo C, Tarokh L, et al. (2014) A Longitudinal Assessment of Sleep Timing, Circadian Phase, and Phase Angle
of Entrainment across Human Adolescence. PLoS ONE 9(11): e112199. doi:10.1371/journal.pone.0112199
Editor: Steven A. Shea, Oregon Health & Science University, United States of America
Received July 10, 2014; Accepted October 13, 2014; Published November 7, 2014
Copyright: ß 2014 Crowley et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its
Supporting Information files.
Funding: This work was supported by a grant from the National Institutes of Health (AA 13252) to MAC. The content is solely the responsibility of the authors
and does not necessarily represent the official views of the National Institutes of Health or the National Institute of Alcohol Abuse and Alcoholism. The National
Institute of Alcohol Abuse and Alcoholism and the National Institutes of Health had no involvement in designing the study, data collection, data analysis and
interpretation, writing of the manuscript, nor in the decision to submit the manuscript for publication. The authors alone are responsible for the content and
writing of the paper.
Competing Interests: The authors have read the journal’s policy and have the following competing interests: Dr. Acebo is a shareholder of Jazz Pharmaceuticals
plc and employee of Jazz Pharmaceuticals, Inc. who, in the course of this employment, has received stock options exercisable for, and other stock awards of,
ordinary shares of Jazz Pharmaceuticals plc. Her work on this project preceded this industry involvement and the study was not supported in any way by Jazz
Pharmaceuticals. Dr. Barker is a consultant for Insmed, Inc.; the article submitted is not related to this relationship. The other authors have indicated no competing
interests. This does not alter the authors’ adherence to PLOS ONE policies on sharing data and materials.
* Email: [email protected]

Introduction actigraphically-estimated sleep support these findings [8,9]. A

shift toward ‘‘eveningness’’ also emerges as youngsters age [10],
The transition through adolescence (the second decade) is often and this shift appears to be linked to pubertal development
accompanied by a shift toward later timing of sleep/wake [11,12]. For American students, this delay of sleep behavior is
behavior. Survey studies from around the globe report later often concurrent with the earliest school start times, reducing the
bedtimes on both school and non-school nights and later wake-up opportunity for sleep on school nights for many. Cross-sectional
times on non-school or vacation mornings as youngsters age and longitudinal studies of adolescent sleep length reflect this
[1,2,3,4,5,6,7]. Cross-sectional and longitudinal studies using circumstance, showing a consistent age-related reduction of total

PLOS ONE | www.plosone.org 1 November 2014 | Volume 9 | Issue 11 | e112199

 Sleep and Circadian Rhythms across Adolescence

sleep time [1,9,13,14,15,16,17,18,19]. Restricted weekday sleep DLMO to bedtime, particularly during the school year, with an
during the school year is often compensated by over-sleeping on older group (13–16 years) showing a wider phase angle compared
weekends, primarily through later wake-up times [1,4,20,21]. to a younger group (9–12 years) [31]. To date, no published
Many studies reporting sleep/wake behavioral timing across studies have examined developmental changes to circadian
adolescence have used cross-sectional comparisons of several age physiology or the temporal alignment between the circadian
groups. While cross-sectional studies are a crucial step to system and sleep/wake behavior using a longitudinal design in
understanding developmental differences, the findings can be adolescent humans.
misleading as this approach does not account for individual The current study is a descriptive longitudinal evaluation of
differences in developmental trajectories [22]. Longitudinal studies actigraphically-estimated weekday and weekend sleep onset and
allow one to better understand developmental trajectories of offset times, weekend-to-weekday differences in sleep onset and
behavior. Of the few studies reporting longitudinal patterns of offset times, circadian phase (measured using salivary DLMO
adolescent sleep/wake timing, two examined young adolescents phase) and phase angles of entrainment in a younger adolescent
(,10–13 years) [2,9]. Laberge and colleagues reported an age- cohort and an older adolescent cohort. We hypothesized that
related delay in parent-reported bedtimes on school and weekend participants in both cohorts would show later DLMOs, sleep
nights, and a delay in weekend wake-up time over four years in onsets on weekends and weekdays, and sleep offsets on weekends
young Canadian adolescents. Sadeh and colleagues found a over time and that weekend sleep onset and sleep offset differences
similar age-related delay in sleep onset time measured by would increase with age. We did not have specific hypotheses for
actigraphy in a group of young adolescents (9.9 to 11.2 years at the developmental changes of phase angles of entrainment.
baseline) followed over 3 years, and this change in sleep timing
predicted changes to self-assessed puberty ratings. Andrade and Methods
colleagues [20] reported a delay in self-reported weekend wake-up
time (on average between 34 and 39 minutes) in Brazilian Participants
adolescents (12–16 years); however, measurements were taken Two cohorts of adolescents living in the northeast United States
over one year only. Weekday and weekend bedtime and weekday (41u 499 N, 71u 249 W) participated in this study from 2002 to
wake time did not change over this year. The last longitudinal 2007: a younger cohort first assessed at age 9 or 10 years and an
study focused on the transition from high school to college [23]. older cohort whose baseline assessment occurred at age 15 or 16
Urner and colleagues measured actigraphically-estimated sleep/ years. Participants were screened for the following exclusion
wake patterns in Swiss students while in high school (aged 17 to 19 criteria at the initial assessment using questionnaires completed by
years) and again 5 years later when in university. Bedtime, wake- participants and a parent: chronic insufficient sleep concomitant
up time, and mid-sleep time shifted on average 34, 52, and with signs of excessive sleepiness, such as falling asleep inappro-
44 minutes later on school days after the transition to college, but priately; more than 3-hour variation in self-reported sleep times
sleep timing did not change on non-school days. In summary, across a week; personal or family history of a medical, psychiatric,
previous longitudinal studies focused on early [2,9] or late or sleep disorder; current illness; use of prescription medications or
adolescence [23] and the only longitudinal study that focused on over-the-counter medications known to affect sleep, alertness, or
a relatively wide age range, followed youngsters for one year only suppress melatonin; or physical or mental handicap. The Lifespan
[20]. Thus, longitudinal assessments of sleep/wake timing using Institutional Review Board approved the study, and participants
objective measures at more frequent intervals and spanning the received payment for taking part in each assessment. A parent
second decade of life are needed to understand developmental gave written informed consent, and participants co-signed to
trajectories of sleep/wake behavior during adolescence. indicate their assent to participate in the study. Participants gave
Delayed sleep timing during adolescence is partly driven by written informed consent if they were 18 years or older.
environmental factors that can displace sleep, such as part-time
work, homework, television-watching, or other media use
Procedures
[24,25,26]; however, changes to the homeostatic sleep [27] and
Procedures were the same for both cohorts, and assessments
circadian timing systems [28,29], particularly during pubertal
were scheduled to occur during the school year, excluding
development, can also explain delayed sleep/wake timing in these
vacations and holidays and not within one week of transition to
youngsters. Previous studies in adolescent humans [28,29] and
Daylight Saving Time. Participants were invited for a baseline
other young mammals [30] report a puberty-related delay of the
circadian timing system. In the cross-sectional studies of assessment and for follow-up assessments approximately every six
Carskadon and colleagues [28,29], for example, the melatonin months for 2.5 years thereafter (total planned assessments = 6). At
rhythm was later in participants who were late- or post-pubertal each assessment, participants wore an actigraph on their non-
compared to those who were pre- or early pubertal. Of note is that dominant wrist (Mini-motionlogger, Ambulatory Monitoring, Inc.,
all of these youngsters kept the same sleep/wake (and thus dark/ Ardsley, NY, USA) and kept a daily sleep diary in which they
light) pattern for at least one week before circadian phase documented their sleep pattern. This monitoring occurred for at
measurement, which indicates that changes to circadian timing least one week before coming to the laboratory to measure the
were not driven by differences in sleep/wake patterns. Given a salivary dim light melatonin onset (DLMO) phase on a weekday
fixed sleep/wake (dark/light) schedule and differences in circadian evening. Participants were free to select their sleep times, and they
timing, these data may also indicate developmental differences in were instructed to sleep at home alone and not remain awake all
the temporal alignment of sleep/wake behavior with the internal night. Activity data were collected in 1-minute epochs using a
circadian clock. Such temporal alignment is referred to as the Zero-Crossing Mode (ZCM) and filter setting 18 (the manufac-
phase angle of entrainment, i.e., the time interval between an turer’s setting for frequency bandpass of 2 to 3 Hz). Further
endogenous circadian marker (e.g., the dim light melatonin onset, verification of bedtimes and wake-up times were provided with
or DLMO) and a recurring external cyclic event (e.g., sleep onset twice-daily telephone calls to the laboratory’s time-stamped
or sleep offset). Indeed, a previous cross-sectional analysis of answering machine, immediately before going to bed and
adolescents reported an age-related difference in the interval from immediately after waking.

PLOS ONE | www.plosone.org 2 November 2014 | Volume 9 | Issue 11 | e112199

 Sleep and Circadian Rhythms across Adolescence

Saliva collection occurred in the laboratory under dim light onset time (decimal hours) minus DLMO phase (decimal hours))
conditions (,40 lux) using untreated Salivettes (Starstedt, and between DLMO phase and average weekday sleep offset
Germany) at 30-minute intervals. Participants were seated for five ((sleep offset time (decimal hours) +24) – DLMO phase (decimal
minutes before each of the twelve saliva samples collected hours)).
beginning 5 hours before the participant’s average bedtime (as Outcome measures by participant and assessment are available
calculated from daily telephone messages) and ending 30 minutes in Table S1.
after average bedtime. If participants ingested anything other than
water before the sample, they rinsed their mouths and brushed Statistical Analysis
their teeth with water. Saliva samples were centrifuged, frozen, The aim of this analysis was to describe developmental
and later assayed for melatonin using radioimmunoassay (RIA) test trajectories of sleep/wake timing, circadian timing, and phase
kits (ALPCO, Windham, NH, USA). An individual’s samples were angles of entrainment across adolescence. Therefore, statistical
analyzed in the same batch. The intra-assay coefficients of analyses for each outcome variable were carried out in three steps.
variation for low and high levels of salivary melatonin were In the first step, visual inspection of plots derived from non-
4.1% and 4.8%, respectively. The inter-assay coefficients of parametric localized regression was used to examine the shapes of
variation for low and high levels of salivary melatonin were 6.6% the developmental trajectories for each variable. In the second
and 8.4%, respectively. The functional least detectable dose of the step, mixed model analyses were used to test for differences in
assay (minimum salivary melatonin concentration measured with variables across age in order to account for the multiple
an intra-assay coefficient of variation of less than 10%) was assessments per participant. Separate models were fit for the
0.9 pg/mL. younger and older cohorts, a level-1 model testing for change in
A physician examined participants at each assessment to variables across chronological age using linear and quadratic
determine pubertal status using the criteria of Tanner [32], a trends and a level-2 model including a random effect for the
staging system based on secondary sexual characteristics with a intercept and linear trend, which allowed for variation among
range from stage 1 (i.e., child-like pre-pubertal) to 5 (post- participants in these two parameters. Non-significant quadratic
pubertal). Classification presented here is for pubic hair growth trends were dropped from final models. In the third step, two
[32]. Morningness/eveningness, a measure of when someone feels additional mixed models were run for each outcome variable in
at their peak level of functioning, was also measured at each order to examine the influence of Tanner stage and sex on the
assessment using the Morningness questionnaire of Smith and developmental trajectories: 1) Tanner stage was dichotomized (,
colleagues [33]. Scores on this scale range from 10 to 56, with a stage 3 vs. $ stage 3) and entered as a time-varying covariate at
lower score indicating eveningness. Alcohol and medication that level-1 for the younger cohort; Tanner stage was not examined in
would affect melatonin levels or sleep were prohibited throughout the older cohort because all participants were$ stage 3; 2) sex was
the assessment weeks. Caffeine and chocolate were prohibited entered in as a level-2 covariate and allowed to interact with the
after noon each day. linear and quadratic trends. All analyses were run using SAS
version 9.3 (SAS Institute Inc, Cary, NC).
Outcome measures
Actigraphic sleep data were analyzed using the Action-W2 Results
software (version 2.3.20, Ambulatory Monitoring, Inc., Ardsley,
NY, USA) to estimate sleep/wake using the validated ‘‘Sadeh’’ In the younger cohort, 38 participants contributed 179
algorithm [34]. Each sleep episode was inspected within a scoring observations (mean per participant = 3.98; range = 1 to 6;
interval spanning 15-minutes before participants reported trying to n = 17 completed all 6 assessments) across the ages of 9 to 13 years.
fall asleep to 15-minutes after reported wake-up time on their daily In the older cohort, 56 participants contributed 221 observations
sleep diary. The following variables were derived according to the (mean per participant = 4.74; range = 1 to 6; n = 15 completed all
procedures of Acebo and colleagues [35]: sleep onset time (the first 6 assessments) across the ages of 15 to 19 years. Demographic
minute of at least 3 consecutive minutes of sleep), sleep offset time information for individuals at each age year is presented in
(the last minute of at least 5 consecutive minutes of sleep before the Table 1.
end of the scoring interval), and sleep minutes (minutes of the sleep
interval scored as sleep). If periods of low activity persisted outside Developmental behavioral sleep trajectories
the scoring interval, members of the research team trained on The mean values for each actigraphic sleep variable by age are
actigraphy procedures reviewed the record to achieve consensus presented in Table 2. The non-parametric developmental trajec-
scoring. Nocturnal sleep onset, sleep offset, and sleep minutes for tories for actigraphic sleep variables by age are illustrated in
the 7 nights before salivary melatonin collection were aggregated Figure 1. Beginning with the younger cohort (left side of each plot
for weekday and weekend nights separately. Two weekend nights in Figure 1), weekday and weekend sleep onset times (Figure 1 A
were available for all assessments except 6 (n = 2 in the younger and B) shifted later between the ages of 9 and 13 years, with
cohort; n = 4 in the older cohort) in which only one weekend night significant linear trends for weekday (F(1,32) = 36.25, p,.01) and
was available. Five weekday nights were available for all weekend (F(1,32) = 36.04, p,.01) sleep onset times. The weekend-
assessments except 14; four nights were available in 13 assessments weekday sleep onset difference (Figure 1C), however, did not show
(n = 7 in the younger cohort; n = 6 in the older cohort) and three an age-related change in this younger cohort. Weekday sleep offset
nights were available for one assessment in the younger cohort. times shifted earlier in the younger cohort (Figure 1D) after age 11
Weekend and weekday differences for sleep onset and sleep offset (quadratic trend: F(1,100) = 6.60, p = .01). By contrast, weekend
were also computed. sleep offset times (Figure 1E) shifted later between 9 and 13 years,
DLMO phase, expressed in 24-hour clock time, was determined as shown by a significant linear trend (F(1,32) = 9.00, p,.01). The
by linear interpolation across the time points before and after the weekend-weekday differences for sleep offset (Figure 1F) increased
melatonin concentration increased to and stayed above 4 pg/mL from ages 9 to 13 years (linear trend: F(1,32) = 31.00, p,.01),
[29]. Phase angles were defined for each participant as the interval which reflects sleep offset times shifting earlier on weekdays and
between DLMO phase and average weekday sleep onset (sleep later on weekends in the younger cohort.

PLOS ONE | www.plosone.org 3 November 2014 | Volume 9 | Issue 11 | e112199

 Table 1. Demographics by individual at each age in the younger and older cohorts.

Age (younger cohort) Age (older cohort)

9 10 11 12 13 15 16 17 18+

(n = 15) (n = 35) (n = 30) (n = 24) (n = 11) (n = 43) (n = 44) (n = 38) (n = 16)

Sex, n female 7 15 12 13 6 20 21 20 10

PLOS ONE | www.plosone.org

Race, n Caucasian 10 27 24 20 10 29 32 26 11
Tanner Stage, n
1 8 19 6 1 0 - - - -
2 5 10a 15a 10a 2 - - - -
3 1 5 2b 6c 3c - - - -
4 - 1 2d 2 3d 7d 2 - -
5 - - 4e 5 2 34f 42f 38 16
Morningness/ 41.0 41.6 38.9 38.3 35.2 35.7 35.6 34.8 33.9
Eveningness, (4.5) (5.3) (6.2) (5.1) (2.8) (5.5) (6.3) (6.5) (8.4)
mean (SD)
School Start Time, 8:20 8:28 8:19 8:03 7:59 7:46 7:49 7:56 -
mean (SD in mins) (25) (28) (25) (17) (23) (28) (25) (40)

4
a
Participants transitioned from Tanner 1 to 2 at 10 years (n = 5), 11 years (n = 6), and 12 years (n = 5).
b
Participants transitioned from Tanner 1 to 3 at 11 years (n = 1).
c
Participants transitioned from Tanner 2 to 3 at 12 years (n = 1) and 13 years (n = 1).
d
Participants transitioned from Tanner 3 to 4 at 11 years (n = 2), 13 years (n = 1), and 15 years (n = 1).
e
Participants transitioned from Tanner 3 to 5 at 11 years (n = 2).
f
Participants transitioned from Tanner 4 to 5 at 15 years (n = 3) and 16 years (n = 1).
Notes: if more than one Morningness/Eveningness score was collected at each age, then the mean score was used; Tanner stage was unavailable for 1 participant at ages 9, 11, and 13 years, and for 2 participants at age 15 years.
doi:10.1371/journal.pone.0112199.t001

November 2014 | Volume 9 | Issue 11 | e112199

Sleep and Circadian Rhythms across Adolescence
 Table 2. Means (SDs) for actigraphic sleep and circadian outcomes by age in the younger and older cohorts.

Age (younger cohort) Age (older cohort)

9 10 11 12 13 15 16 17 18+

(n = 17) (n = 56) (n = 52) (n = 38) (n = 13) (n = 55) (n = 69) (n = 64) (n = 26)

Weekday sleep:

PLOS ONE | www.plosone.org

Onset time 21:28 21:44 21:59 22:07 22:30 22:34 22:53 23:04 00:14
(mins) (33) (30) (28) (28) (31) (44) (46) (45) (85)
Offset time 06:38 06:46 06:39 06:20 06:20 06:22 06:28 06:34 08:04
(mins) (35) (31) (29) (35) (22) (31) (33) (32) (70)
Total sleep time, h 8.24 8.22 7.87 7.50 7.21 7.13 6.91 6.90 7.32
(.85) (.74) (.68) (.82) (.70) (.86) (1.04) (.98) (1.37)
Weekend Sleep:
Onset time 22:03 22:26 22:38 22:49 23:20 23:19 23:47 00:01 00:45
(mins) (47) (46) (37) (48) (38) (52) (51) (61) (91)
Offset time 07:15 07:22 07:31 07:28 07:55 07:51 08:08 08:08 08:36
(mins) (54) (50) (49) (62) (25) (45) (46) (47) (73)
Total sleep time, h 8.16 8.15 8.04 7.81 7.90 7.68 7.51 7.43 7.20

5
(.98) (.98) (.93) (1.05) (.97) (1.15) (1.16) (1.25) (2.41)
Weekend-Weekday Difference:
Sleep Onset, h .58 .70 .65 .71 .83 .76 .90 .95 .51
(.64) (.56) (.48) (.57) (.57) (.72) (.73) (.79) (1.13)
Sleep Offset, h .62 .60 .87 1.12 1.58 1.47 1.67 1.58 .54
(.78) (.66) (.67) (.74) (.33) (.80) (.67) (.68) (1.13)
DLMO phase 20:42 20:39 20:32 20:50 21:27 20:34 20:50 20:57 21:53
(mins) (37) (47) (45) (50) (49) (57) (59) (58) (91)
Phase angle:
DLMO to Weekday .73 1.07 1.44 1.18 .99 2.05 2.06 2.13 2.17
Sleep Onset, h (.90) (.72) (.78) (1.06) (.73) (.66) (1.04) (.95) (1.19)
DLMO to Weekday 9.88 10.11 10.12 9.44 8.87 9.86 9.64 9.66 10.07
Sleep Offset, h (.78) (.67) (.69) (1.19) (.91) (.84) (.88) (.94) (1.21)

Notes: These data are from 94 participants (N = 38 in the younger cohort; N = 56 in the older cohort) who contributed on average 4.29 assessments range (1 to 6). Three observations at age 19 were included in the 18+ category.
doi:10.1371/journal.pone.0112199.t002

November 2014 | Volume 9 | Issue 11 | e112199

Sleep and Circadian Rhythms across Adolescence
 Sleep and Circadian Rhythms across Adolescence

Figure 1. Modeled developmental trajectories (bold line) and individual trajectories (thin lines) for actigraphically estimated sleep
onset and offset on weekdays (A and D) and weekends (B and E) in the proximal 7 days before DLMO phase was measured in both
cohorts. Sleep onset and offset differences between weekends and weekdays (C and F) illustrate when participants slept earlier (,0) or later (.0) on
weekends compared to weekdays. The younger cohort (9–13 years) is on the left and the older cohort (15–19 years) is on the right of each plot.
doi:10.1371/journal.pone.0112199.g001

Weekday and weekend sleep onset times also showed a delay in cohort, weekday sleep offset times delayed from 15 to 19 years,
the older cohort (right side of each plot in Figure 1) between 15 with a sharp delay after age 17 (Figure 1D). Significant linear
and 19 years indicated by significant linear trends for weekday (F(1,39) = 29.88, p,.01) and quadratic (F(1,123) = 35.81, p,.01)
(F(1,39) = 17.55, p,.05) and weekend (F(1,39) = 13.69, p,.01) trends were seen for weekday sleep offset time. Weekend sleep
sleep onset time (Figure 1A and B). In contrast to the younger offset times (Figure 1E) delayed from age 15 to 19 years in a linear

PLOS ONE | www.plosone.org 6 November 2014 | Volume 9 | Issue 11 | e112199

 Sleep and Circadian Rhythms across Adolescence

fashion (F(1,39) = 10.26, p,.01). Unlike the younger cohort, and phase angle to weekday sleep offset by age are illustrated in
weekend-weekday sleep offset timing differences decreased (Fig- Figure 2.
ure 1F), beginning from age 17 (linear trend: F(1,39) = 6.62, For the younger cohort, DLMO phase (Figure 2A) began to
p = .01; quadratic trend: F(1,120) = 7.30, p,.01), reflecting the show a delay after age 11, indicated by a significant quadratic
later weekday sleep offset times after age 17. trend (F(1,98) = 6.69, p = .01). Phase angle to sleep onset
(Figure 2B) grew wider between ages 9 and 13 years, and this
Developmental circadian timing trajectories increase was most prominent between 9 and about 11.5 years.
The mean values for each circadian outcome variable separated Significant linear (F(1,32) = 12.43, p,.01) and quadratic
by age are shown in Table 2. Non-parametric developmental (F(1,91) = 13.36, p,.01) trends were seen for this measure. Phase
trajectories for DLMO phase, phase angle to weekday sleep onset, angle to sleep offset time (Figure 2C) narrowed with age in the
younger cohort after age 11 years. This pattern reflects the earlier
timing of sleep offset on weekday mornings combined with DLMO
delay after age 11. Phase angle to sleep offset time showed
significant linear (F(1,32) = 7.21, p = .01) and quadratic
(F(1,91) = 17.30, p,.01) trends.
Similar to the younger cohort, DLMO phase became later in
the older cohort. The age-related delay was greater after age 17
(the inflection point in Figure 2A), indicated by a significant
quadratic trend for DLMO phase (F(1,112) = 3.99, p = .05). Phase
angle to sleep offset (Figure 2C) narrowed between about 15 and
17 years and then became wider again after age 17. Significant
linear (F(1,39) = 7.18, p = .01) and quadratic (F(1,106) = 7.00, p,
.01) trends were found for phase angle to sleep offset time. Phase
angle to sleep onset time showed no age-related change in the
older cohort (Figure 2B).

Sex and Tanner stage

Developmental trajectories did not differ by sex for any variable
except sleep onset on weekdays, which showed a more rapid delay
shift with age in females than in males in the younger cohort
(F(1,101) = 4.12, p = .05). Tanner stage added no statistically
significant information above chronological age other than for
sleep onset on weekdays in the younger group, in which Tanner
stage $ 3 was associated with an approximate 12-minute later
sleep onset on weekdays (F(1,100) = 5.11, p = .03).

Discussion
This descriptive longitudinal analysis aimed to examine sleep/
wake timing, circadian timing, and the temporal relations between
measures of the internal circadian clock and the self-selected sleep
patterns of younger and older adolescent cohorts that together
spanned the second decade of life (9–19 years). The developmental
trajectories observed in these two cohorts indicate that sleep/wake
timing shifted later as youngsters aged, except in the case of
weekday wake times, for which school schedules likely dictated the
morning schedule. Overall, we also found an age-related delay of
DLMO phase in both cohorts, with the largest (,1 h) shift
occurring between ages 11 and 13 years in the younger cohort and
between 17 and 19 years in the older cohort. An age-related
increase in phase angle to sleep onset was also seen in the younger
cohort, especially between ages 9 and about 11.5 years.
Participants in the older cohort fell asleep later relative to their
melatonin rhythm compared to the younger cohort, as evidenced
by a wider phase angle of DLMO to sleep onset. In addition,
however, phase angle of DLMO to sleep onset remained
consistent with age in the older cohort.

Self-selected sleep/wake timing across adolescence

Figure 2. Modeled developmental trajectories (bold line) and The age-related delay shift of weekday and weekend sleep onset
individual trajectories (thin lines) for DLMO phase (A), DLMO
phase angle to sleep onset (B), and DLMO phase angle to sleep
and weekend sleep offset in both cohorts of the current study is
offset (C). A negative DLMO phase angle to sleep onset indicates consistent with previous work. Roenneberg and colleagues [10] for
when the DLMO occurred after weekday sleep onset. example, reported a significant age-related delay of about
doi:10.1371/journal.pone.0112199.g002 2.5 hours in reported mid-sleep times of non-work/non-school

PLOS ONE | www.plosone.org 7 November 2014 | Volume 9 | Issue 11 | e112199

 Sleep and Circadian Rhythms across Adolescence

(‘‘free’’) days in a cross-sectional analysis of individuals ages 10 to across ages 9 to 11 years despite sleep onset time shifting later over
20 years. Other previous cross-sectional survey findings showed a this age range. If we apply the (presumed) properties of the phase
similar delayed pattern across the adolescent years [1,21], though response to light [46,47,48,49,50] such evening light exposure
not as large as that reported by Roenneberg and colleagues. The should produce a delay of circadian timing (i.e., later DLMO
younger cohort of the current study manifested a delay of sleep phase). Yet the younger cohort did not manifest this concomitant
onset by about 1 hour (see Table 2). Two previous longitudinal change, and we infer that DLMO phase was stabilized by morning
studies showed a similar 1-hour parent-reported bedtime delay rise-time consistency. One implication, therefore, is that the light
and actigraphic sleep onset delay during early adolescence, though phase response in the younger cohort is more sensitive in the
clock times were descriptively earlier [2] or later [9] than the morning than in the evening.
current sample. Such differences in clock times may be due to the On the other hand, weekday wake-up times did not appear to
way sleep timing was measured (parent report versus actigraphy) influence DLMO phase in the same manner after age 11. DLMO
or to cultural factors. Wake-up time on weekends delayed by about phase shifted later after age 11 in the younger cohort and showed a
40 minutes with age in our younger cohort, and the discrepancy delaying pattern between 15 and 17 years in the older cohort
between weekend and weekday wake-up times increased from despite earlier sleep offset times on weekdays. Later sleep onset,
about 37 minutes at age 9 to about 50 minutes at age 13, on and therefore exposure to ambient light later into the evening,
average (see Table 2). Laberge and colleagues [2] also reported may have driven a later DLMO phase at these ages. This finding is
such age-related changes to parent-reported wake-up times in their consistent with a previous study from this lab in which DLMO
longitudinal assessment of adolescents in Canada, though the phase was later in adolescents across a transition to an earlier
weekend wake time shift was descriptively longer (,90 to school start time accompanied by earlier wake-up times [51]. In
120 minutes, on average) by age 13. combination, these findings suggest that the sensitivity of the
The distinct changes in the developmental trajectories of sleep circadian system to morning phase-resetting light exposure
offset time observed in the current study (Figure 1D) likely reflect decreases and sensitivity to evening phase-delaying light increases
changes in the youngster’s social structure, the most prominent of with age during adolescent development. One study from juvenile
which are school start times. Previous work comparing sleep/wake mice showed a heightened phase delay response to evening light
schedules in response to differences in school start times showed exposure [30], supporting the hypothesis for a developmental
that sleep onset times remain unchanged, but wake-up times shift difference in the circadian clock’s response to light.
earlier or later in tandem with the school schedules [36,37]. In the
current study, weekday sleep offset was about 20 minutes earlier at Phase angles of entrainment across adolescence
age 12 compared to the previous assessment ages. This is the age The phase angle from DLMO to sleep onset was wider in the
when American students transition from elementary to middle older cohort compared to the younger cohort, meaning that the
school, and the school schedule often begins earlier in the day, a older adolescent group fell asleep at a later time on the melatonin
pattern experienced by our participants (see Table 1). In the older rhythm. This finding is consistent with previous cross-sectional
cohort, weekday sleep offset remained consistent at about 6:20– studies of adolescents [31]. In the context of findings from young
6:30 am from ages 15 through 17 years, and a significant delay children and adults, we note a pattern of larger phase angles to
occurred when these youngsters reached 18 years. Most of the sleep onset associated with increasing age. Thus, for example,
participants had completed high school by age 18, and no longer LeBourgeois and colleagues [52] recently reported a median
had the school schedule constraining sleep. A previous cross- DLMO phase to sleep onset interval of about 65 minutes in
sectional comparison [38] and longitudinal assessment [23] of toddlers aged 30–36 months sleeping on their habitual parent-
sleep times across the transition from high school to college also selected sleep schedules. Phase angles to sleep onset in our younger
showed a pattern of delayed wake-up times. cohort and in a previously studied group of 9–12 year olds [31]
The consistent early weekday sleep offset times across 9 to 17 was similar (,1 h) to these toddlers. Our older cohort, however,
years, followed by a delay at age 18 and 19 years indicates that the had a phase angle to sleep onset average closer to 2 h, similar to
school schedule may suppress a biologically-driven behavior to that reported for adults on self-selected sleep schedules [53,54,55].
sleep later. This implication is bolstered by trajectories of later We suggest that the adult-like temporal alignment between sleep
weekend sleep offset times in the two cohorts, as well as by the and the circadian timing system emerges during late adolescence
reduction of weekday-weekend sleep offset differences after age 17. and may be related to diminished parental involvement in setting
Roenneberg and colleagues [39] reported that the degree to which bedtimes [1,19,21,56,57]. Although we did not measure parental
weekend and weekday sleep timing differ increases over the second involvement in this study, one might presume that release from
decade of life, and they relate the phenomenon to the construct of parental-set bedtimes results in older adolescents choosing to go to
‘‘social jetlag’’ (i.e., the degree to which social and biological clocks bed later and thus falling asleep later with respect to their
conflict). Larger degrees of social jetlag have been associated with melatonin rhythm.
reduced alertness and performance [40,41,42,43], greater use of On the other hand, we suggest that developmental differences
alcohol, nicotine and caffeine, and an increased risk for depression affecting the homeostatic sleep system can produce longer phase
and obesity [39,44,45]. The current study’s findings support a angles. Previous work illustrates that the dynamics of the
concern that exaggeration of social jetlag and potential associated homeostatic system are altered during adolescent development.
health risks arise as adolescents’ biological tendencies to delay are While the decay of homeostatic sleep pressure across sleep does
confronted by an early school bell. not change [27,58,59], the accumulation of sleep pressure across
waking does. One cross-sectional study modeled the build-up of
Circadian timing across adolescence slow wave activity using sleep before and after 36 hours of sleep
The circadian timing system, as indexed by DLMO phase, deprivation and found an increase in the time constant of the
delayed with age in both cohorts; however, this age-related shift build-up in post-pubertal versus pre-pubertal teens [27]. In other
was non-linear, and the pattern of change was somewhat words, mature adolescents accumulated sleep pressure at a slower
inconsistent with predictions based on weekday sleep/wake rate across a waking interval compared to their younger peers.
(dark/light) timing. Thus, for example, DLMO phase was stable Longer sleep onset latency near bedtime in Tanner 5 adolescents

PLOS ONE | www.plosone.org 8 November 2014 | Volume 9 | Issue 11 | e112199

 Sleep and Circadian Rhythms across Adolescence

compared to Tanner 1 adolescents following 14.5 and 16.5 hours Figures 1 and 2). Developmental trajectories differed by sex and
awake provides further support for this developmental difference Tanner stage for weekday sleep onset in the younger cohort only.
in sleep pressure [60]. Therefore, other factors – perhaps genetic and/or environmental
Figure 3 illustrates a model relating developmental changes in – likely influence an individual’s pattern of change in sleep/wake
circadian timing and sleep/wake homeostasis to phase angle of and circadian timing across adolescence. Future work to examine
DLMO and sleep onset that we propose to explain the phase angle these individual differences may inform our understanding of
differences between our age cohorts. Most assessments of the youngsters who could be at risk for developing circadian-based
younger adolescents occurred at a developmental stage (Tanner sleep complaints, such as a delayed sleep phase.
stages 1 to 3) associated with achieving maximum sleep pressure The current study had a number of strengths, including the
earlier relative to rise time. On the other hand, the older longitudinal design and use of actigraphy and melatonin onset to
adolescents were all late or post-pubertal (Tanner stages 4 and 5), a describe developmental trajectories of sleep and circadian rhythms
developmental stage at which sleep pressure builds more slowly and the temporal alignment between these two processes across
across the day. Our younger cohort exhibited an average of 15 h adolescence. The gap between the ages at the last assessment in
20 min between waking and falling asleep; whereas the waking the younger cohort and the first assessment in the older cohort is a
day length was 16 h 22 min in the older cohort. We propose that significant limitation. Furthermore, the sample size limited our
the differences in accumulation of homeostatic sleep pressure may ability to examine sleep and circadian timing trajectories by sex.
allow the older adolescent to stay awake for a longer period of time Recent data indicate sex differences in underlying circadian
after their DLMO, thus extending the observed phase angle to physiology and its temporal relationship to sleep [55,61,62].
sleep onset. These differences in day length can account for the Future longitudinal work examining sleep and circadian timing by
longer phase angle in older adolescents (about 2 h) versus younger sex may inform the developmental timing of sex differences in
adolescents (about 1 h), as noted above. these parameters. Finally, the current analysis was intended to be
descriptive. Examining the associations between the sleep and
Study limitations and future directions circadian timing trajectories and determining which may predict
The longitudinal design and analytic approach of the current the other over time requires a different analytic approach (e.g.,
study allowed us to observe individual variability in developmental structural equation modeling) than the one used here. Future
trajectories of sleep/wake behavior, DLMO phase, and phase analyses are planned to examine associations between sleep timing
angles of entrainment across early and late adolescence (see

Figure 3. A proposed model to explain phase angle to sleep onset differences in the younger (top) and older (bottom) adolescent
cohorts. Black horizontal bars illustrate average sleep times for each cohort (younger: 21:55–06:35; older: 23:02–06:40). Bold lines illustrate sleep
pressure accumulation and dissipation functions predicted by the homeostatic sleep system. The upward facing arrow indicates the average DLMO
phase for each age group (younger: 20:42; older: 20:54), and the right-facing block arrow shows the interval between DLMO phase and sleep onset
(phase angle to sleep onset). Based on previous modeling work, [27] the saturating exponential function reaches it maximum more quickly and
therefore at an earlier clock time in the younger cohort compared to the older cohort. We propose that the older adolescents are able to stay awake
for a longer period of time (,2 h) after DLMO phase compared to the younger adolescents (,1 h) because of this developmental difference in
homeostatic sleep pressure at the end of the waking day.
doi:10.1371/journal.pone.0112199.g003

PLOS ONE | www.plosone.org 9 November 2014 | Volume 9 | Issue 11 | e112199

 Sleep and Circadian Rhythms across Adolescence

and circadian timing trajectories, but that this analysis was beyond to thank Denise Maceroni for saliva assays and Jennifer Maxwell for
the scope of the current descriptive paper. assistance with data reduction and compilation. The assistance of the staff
at the E.P. Bradley Sleep Lab, and Brown University students in data
collection is gratefully acknowledged.
Supporting Information
Table S1 Outcome measures by participant and assess- Author Contributions
ment.
Conceived and designed the experiments: MKL CA RS MAC. Performed
(XLS) the experiments: SJC EVR MKL CA MAC. Analyzed the data: SJC CA
LT RS DHB. Wrote the paper: SJC EVR MKL CA LT RS DHB MAC.
Acknowledgments Interpretation of results: SJC EVR LT DHB MAC.
The authors thank our consultants on the project, Dr. Tim Roehrs, Dr. J.
Todd Arnedt, Dr. Robert Swift, and Dr. Peter Monti. Also, we would like

References
1. Wolfson AR, Carskadon MA (1998) Sleep schedules and daytime functioning in 27. Jenni OG, Achermann P, Carskadon MA (2005) Homeostatic sleep regulation in
adolescents. Child Development 69: 875–887. adolescents. Sleep 28: 1446–1454.
2. Laberge L, Petit D, Simard C, Vitaro F, Tremblay R, et al. (2001) Development 28. Carskadon MA, Acebo C, Jenni OG (2004) Regulation of adolescent sleep:
of sleep patterns in early adolescence. J Sleep Res 10: 59–67. Implications for behavior. Ann NY Acad Sci 1021: 276–291.
3. Giannotti F, Cortesi F, Sebastiani T, Ottaviano S (2002) Circadian preference, 29. Carskadon MA, Acebo C, Richardson GS, Tate BA, Seifer R (1997) An
sleep and daytime behaviour in adolescence. J Sleep Res 11: 191–199. approach to studying circadian rhythms of adolescent humans. J Biol Rhythms
4. O’Brien EM, Mindell JA (2005) Sleep and risk-taking behavior in adolescents. 12: 278–289.
Behav Sleep Med 3: 113–133. 30. Hagenauer MH, Perryman JI, Lee TM, Carskadon MA (2009) Adolescent
5. Yang CK, Kim JK, Patel SR, Lee JH (2005) Age-related changes in sleep/wake changes in the homeostatic and circadian regulation of sleep. Dev Neurosci 31:
patterns among Korean teenagers. Pediatrics 115: 250–256. 276–284.
6. Crowley SJ, Acebo C, Carskadon MA (2007) Sleep, circadian rhythms, and 31. Crowley SJ, Acebo C, Fallone G, Carskadon MA (2006) Estimating dim light
delayed phase in adolescence. Sleep Med 8: 602–612. melatonin onset (DLMO) phase in adolescents using summer or school-year
7. Randler C (2008) Differences in sleep and circadian preference between Eastern sleep/wake schedules. Sleep 29: 1632–1641.
and Western German adolescents. Chronobiol Int 25: 565–575. 32. Tanner J (1962) Growth at Adolescence. Oxford: Blackwell.
8. Wolfson AR, Carskadon MA, Acebo C, Seifer R, Fallone G, et al. (2003) 33. Smith CS, Reilly C, Midkiff K (1989) Evaluation of three circadian rhythm
Evidence for the validity of a sleep habits survey for adolescents. Sleep 26: 213– questionnaires with suggestions for an improved measure of morningness. J Appl
216. Psychol 74: 728–738.
9. Sadeh A, Dahl RE, Shahar G, Rosenblat-Stein S (2009) Sleep and the 34. Sadeh A, Sharkey KM, Carskadon MA (1994) Activity-based sleep-wake
Transition to Adolescence: A Longitudinal Study. Sleep 32: 1602–1609. identification: An empirical test of methodological issues. Sleep 17: 201–207.
10. Roenneberg T, Kuehnle T, Pramstaller PP, Ricken J, Havel M, et al. (2004) A 35. Acebo C, Sadeh A, Seifer R, Tzischinsky O, Wolfson AR, et al. (1999)
marker for the end of adolescence. Curr Biol 14: R1038–1039. Estimating sleep patterns with activity monitoring in children and adolescents:
11. Carskadon MA, Vieira C, Acebo C (1993) Association between puberty and how many nights are necessary for reliable measures? Sleep 22: 95–103.
delayed phase preference. Sleep 16: 258–262. 36. Wolfson AR, Spaulding NL, Dandrow C, Baroni EM (2007) Middle school start
12. Frey S, Balu S, Greusing S, Rothen N, Cajochen C (2009) Consequences of the times: the importance of a good night’s sleep for young adolescents. Behav Sleep
timing of menarche on female adolescent sleep phase preference. PLoS ONE 4: Med 5: 194–209.
e5217. 37. Boergers J, Gable CJ, Owens JA (2014) Later school start time is associated with
13. Strauch I, Meier B (1988) Sleep need in adolescents: a longitudinal approach. improved sleep and daytime functioning in adolescents. J Dev Behav Pediatr 35:
Sleep 11: 378–386. 11–17.
14. Thorleifsdottir B, Bjornsson JK, Benediktsdottir B, Gislason T, Kristbjarnarson
38. Lund HG, Reider BD, Whiting AB, Prichard JR (2010) Sleep patterns and
H (2002) Sleep and sleep habits from childhood to young adulthood over a 10-
predictors of disturbed sleep in a large population of college students. J Adolesc
year period. J Psychosom Res 53: 529–537.
Health 46: 124–132.
15. Iglowstein I, Jenni OG, Molinari L, Largo RH (2003) Sleep duration from
39. Roenneberg T, Allebrandt KV, Merrow M, Vetter C (2012) Social jetlag and
infancy to adolescence: reference values and generational trends. Pediatrics 111:
obesity. Curr Biol 22: 939–943.
302–307.
40. Taylor A, Wright HR, Lack LC (2008) Sleeping-in on the weekend delays
16. Fredriksen K, Rhodes J, Reddy R, Way N (2004) Sleepless in Chicago: Tracking
circadian phase and increases sleepiness the following week. Sleep and Biological
the effects of adolescent sleep loss during the middle school years. Child
Rhythms 6: 172–179.
Development 75: 84–95.
17. Olds T, Blunden S, Petkov J, Forchino F (2010) The relationships between sex, 41. Yang CM, Spielman AJ, D’Ambrosio P, Serizawa S, Nunes J, et al. (2001) A
age, geography and time in bed in adolescents: a meta-analysis of data from 23 single dose of melatonin prevents the phase delay associated with a delayed
countries. Sleep Med Rev 14: 371–378. weekend sleep pattern. Sleep 24: 272–281.
18. Maslowsky J, Ozer EJ (2014) Developmental Trends in Sleep Duration in 42. Yang CM, Spielman AJ (2001) The effect of a delayed weekend sleep pattern on
Adolescence and Young Adulthood: Evidence From a National United States sleep and morning functioning. Psychol Health 16: 715–725.
Sample. J Adolesc Health 54: 691–697. 43. Burgess HJ, Legasto CS, Fogg LF, Smith MR (2012) Can small shifts in
19. National Sleep Foundation 2014 Sleep in America Poll: Sleep in the Modern circadian phase affect performance? Appl Ergon 44: 109–111.
Family. Washington, D.C.: Available: http://sleepfoundation.org/sleep-polls- 44. Levandovski R, Dantas G, Fernandes LC, Caumo W, Torres I, et al. (2011)
data/2014-sleep-the-modern-family. Accessed July 9, 2014. Depression scores associate with chronotype and social jetlag in a rural
20. Andrade MM, Benedito-Silva AA, Domenice S, Arnhold IJ, Menna-Barreto L population. Chronobiol Int 28: 771–778.
(1993) Sleep characteristics of adolescents: a longitudinal study. J Adolesc Health 45. Wittmann M, Dinich J, Merrow M, Roenneberg T (2006) Social jetlag:
14: 401–406. misalignment of biological and social time. Chronobiol Int 23: 497–509.
21. National Sleep Foundation 2006 Sleep in America Poll. Washington, D.C.: 46. Minors DS, Waterhouse JM, Wirz-Justice A (1991) A human phase-response
Available: http://sleepfoundation.org/sleep-polls-data/sleep-in-america-poll/ curve to light. Neurosci Lett 133: 36–40.
2006-teens-and-sleep. Accessed July 9, 2014. 47. Khalsa SBS, Jewett ME, Cajochen C, Czeisler CA (2003) A phase response
22. Kraemer HC, Yesavage JA, Taylor JL, Kupfer D (2000) How can we learn curve to single bright light pulses in human subjects. J Physiol 549.3: 945–952.
about developmental processes from cross-sectional studies, or can we? 48. Kripke DF, Elliott JA, Youngstedt SD, Rex KM (2007) Circadian phase
Am J Psychiatry 157: 163–171. response curves to light in older and young women and men. J Circadian
23. Urner M, Tornic J, Bloch KE (2009) Sleep patterns in high school and university Rhythms 5: 4.
students: a longitudinal study. Chronobiol Int 26: 1222–1234. 49. St Hilaire MA, Gooley JJ, Khalsa SB, Kronauer RE, Czeisler CA, et al. (2012)
24. Carskadon MA (1989-1990) Adolescent sleepiness: increased risk in a high-risk Human phase response curve to a 1 h pulse of bright white light. J Physiol 590:
population. Alcohol, Drugs and Driving 5/6: 317–328. 3035–3045.
25. Van den Bulck J (2004) Television viewing, computer game playing, and 50. Revell VL, Molina TA, Eastman CI (2012) Human phase response curve to
Internet use and self-reported time to bed and time out of bed in secondary- intermittent blue light using a commercially available device. J Physiol 590:
school children. Sleep 27: 101–104. 4859–4868.
26. Gradisar M, Wolfson AR, Harvey AG, Hale L, Rosenberg R, et al. (2013) The 51. Carskadon MA, Wolfson AR, Acebo C, Tzischinsky O, Seifer R (1998)
sleep and technology use of Americans: findings from the National Sleep Adolescent sleep patterns, circadian timing, and sleepiness at a transition to early
Foundation’s 2011 Sleep in America poll. J Clin Sleep Med 9: 1291–1299. school days. Sleep 21: 871–881.

PLOS ONE | www.plosone.org 10 November 2014 | Volume 9 | Issue 11 | e112199

 Sleep and Circadian Rhythms across Adolescence

52. Lebourgeois MK, Carskadon MA, Akacem LD, Simpkin CT, Wright KP, Jr., 58. Tarokh L, Carskadon MA, Achermann P (2012) Dissipation of sleep pressure is
et al. (2013) Circadian phase and its relationship to nighttime sleep in toddlers. stable across adolescence. Neuroscience 216: 167–177.
J Biol Rhythms 28: 322–331. 59. Campbell IG, Darchia N, Higgins LM, Dykan IV, Davis NM, et al. (2011)
53. Martin SK, Eastman CI (2002) Sleep logs of young adults with self-selected sleep Adolescent changes in homeostatic regulation of EEG activity in the delta and
times predict the dim light melatonin onset. Chronobiol Int 19: 695–707. theta frequency bands during NREM sleep. Sleep 34: 83–91.
54. Burgess HJ, Eastman CI (2005) The dim light melatonin onset following fixed 60. Taylor DJ, Jenni OG, Acebo C, Carskadon MA (2005) Sleep tendency during
and free sleep schedules. J Sleep Res 14: 229–237. extended wakefulness: insights into adolescent sleep regulation and behavior.
55. Van Reen E, Sharkey KM, Roane BM, Barker D, Seifer R, et al. (2013) Sex of J Sleep Res 14: 239–244.
college students moderates associations among bedtime, time in bed, and 61. Cain SW, Dennison CF, Zeitzer JM, Guzik AM, Khalsa SB, et al. (2010) Sex
circadian phase angle. J Biol Rhythms 28: 425–431. differences in phase angle of entrainment and melatonin amplitude in humans.
56. Carskadon MA (1990) Patterns of sleep and sleepiness in adolescents. J Biol Rhythms 25: 288–296.
Pediatrician 17: 5–12. 62. Duffy JF, Cain SW, Chang A, Phillips AJK, Munch MY, et al. (2011) Sex
57. Loessl B, Valerius G, Kopasz M, Hornyak M, Riemann D, et al. (2008) Are difference in the near-24-hour intrinsic period of the human circadian timing
adolescents chronically sleep-deprived? An investigation of sleep habits of
system. PNAS 108: 15602–15608.
adolescents in the Southwest of Germany. Child Care Health Dev 34: 549–556.

PLOS ONE | www.plosone.org 11 November 2014 | Volume 9 | Issue 11 | e112199