Content uploaded by Stefania Maggi
Author content
All content in this area was uploaded by Stefania Maggi on Dec 30, 2014
Content may be subject to copyright.
educational ReseaRcheR
56
Educational Researcher, Vol. 40, No. 2, pp. 56–61
DOI: 10.3102/0013189X11402323
© 2011 AERA. http://er.aera.net
The authors have integrated the major findings on the sleep–wake
cycle and its performance correlates in adolescents. Basic research
shows that lack of synchronicity between early school start times and
the circadian rhythm of adolescents (and the sleep debt accumulated
as a result) involves several cognitive correlates that may harm the
academic performance of adolescent students. The authors therefore
examined findings from pilot interventions in which schools delayed
their start times; specifically, they examined the effects on students,
including potential pitfalls and strategies to consider for effective
scheduling change. There is sufficient evidence that adolescent stu-
dents would benefit from delaying school start times and that this
change can be implemented with tolerable consequences if ade-
quately strategized by school districts and communities.
Keywords: circadian rhythm; school starting times; sleep; student
performance
As children grow into adolescents, they tend to experi-
ence changes in the sleep–wake cycle, causing them to
exhibit a natural preference to go to bed later and wake
up later in the day (Carskadon, 1999, 2002; Dement & Vaughan,
1999). Although in the past this change in the sleep–wake cycle
was attributed to cultural and psychosocial factors (see the review
by Millman et al., 2005), most recent research points to a change
in specific biological processes during adolescence as a primary
determinant of this delayed sleep cycle (Carskadon, 1999, 2002;
Fischer et al., 2008), which has been coined delayed phase prefer-
ence (DPP; Mitru, Millrood, & Mateika, 2002).
Previous studies have shown a strong relationship between
various features of the sleep–wake cycle, affective states, and cog-
nitive performance (Schmidt, Collette, Cajochen, & Peigneux,
2007). Therefore, it is not surprising that researchers and educa-
tors alike are concerned with the study of DPP and how it may
influence the academic achievement of adolescents. More spe-
cifically, researchers and educators are exploring the possibility of
having schools accommodate DPP in adolescents by delaying
start times, for the sake of improving academic performance in
the student population.
In this article, we review the most current research on the
topic of DPP and discuss the impact of schools’ setting later start
times. More specifically, we focus on the following issues: rele-
vant features of the sleep–wake cycle; how the circadian typology
of an individual typically changes over the life span (i.e., DPP in
adolescents); the behavioral and cognitive correlates associated
with performing different tasks at different times of day and their
repercussions in an academic context; case studies; and finally,
considerations for implementing a change in start times.
However, the present review will not address in detail an eco-
nomic benefit–cost analysis because 1980 research linking DPP
to vital statistics regarding adolescent students has unequivocally
demonstrated the hidden and nonhidden, potentially incalcula-
ble costs to society (see reviews in Kryger, Roth, & Dement,
2000). We hope the reader will keep these incalculable costs in
mind as critical context for our article.
The Sleep–Wake Cycle
Blatter and Cajochen (2007) explain that there are two key phe-
nomena that characterize the sleep–wake cycle in humans: Sleep
readily occurs at specific times of the day, and although we are
prone to experiencing fluctuation in our level of sleepiness through-
out the day, the longer we go without sleep, the more difficult it
becomes to resist its onset. These features of the sleep–wake cycle
are thought to reflect two opposing yet simultaneous processes: the
circadian rhythm and the homeostatic drive for sleep. The circa-
dian rhythm, which oscillates within a period of approximately 24
hours, has been described as “a clock-like process independent of
whether the person is asleep or awake that is normally synchro-
nized with external time (i.e., time of day)” (Blatter & Cajochen,
2007, p. 201). The circadian rhythm can also be described as a
wake-promoting system, increasing wakefulness at a regulated time
of day (Blatter & Cajochen, 2007; Edgar, Dement, & Fuller,
1993). This circadian process acts in direct opposition to the
homeostatic drive for sleep, which represents a sleep-promoting
process. The homeostatic drive for sleep accumulates during time
spent awake and decreases during time spent asleep.
In general, according to this two-process model, an individual
will become sleepy when the homeostatic drive for sleep is exerting
School Start Times and the Sleep–Wake Cycle of
Adolescents: A Review and Critical Evaluation of
Available Evidence
Matthew Kirby, Stefania Maggi, and Amedeo D’Angiulli
Reviews/Essays
MaRch 2011 57
greater influence than the wake-promoting circadian process, and
the individual will feel awake when the opposite is true. This model
of sleep, involving these two opposing processes, was originally
developed to predict and explain human sleep regulation; however,
researchers are now using it to help explain task performance,
which has been found to be mediated by these processes (Blatter &
Cajochen, 2007).
Circadian Typology of Adolescence: Delayed Phase
Preference
Individuals exhibit different preferences regarding the best time
of day to be awake and to be asleep; this preference is called the
chronotype of an individual, the extremes of which are morning
preference and evening preference. Previous research has found
that the evening preference chronotype, which is most prevalent
in adolescents (Yoon, May, & Hasher, 1999), is correlated with
pubertal development; this means that adolescents who are older
and more mature (in terms of pubertal development) show
greater evening preference than do younger, less mature adoles-
cents (Crowley, Acebo, & Carskadon, 2007; Giannotti, Cortesi,
Sebastiani, & Ottaviano, 2002).
This DPP in adolescents is not a recent discovery; however,
it was previously believed to have been mediated exclusively by
behavioral factors such as social and academic pursuits (Crowley
et al., 2007). Although behavioral factors appear to play a role
in determining chronotype, as evidenced by reports of adoles-
cent social and entertainment activities (e.g., the Sleep in
America Poll by the National Sleep Foundation, 2006), there is
a growing body of evidence that implicates biological changes
at the onset of puberty as being major determinants of DPP
(Crowley et al., 2007). At least partially determined by intrinsic
biological factors, the lengthening of the circadian oscillation
period, a slower accumulation of the homeostatic drive for
sleep, and a greater requirement for sleep during adolescence
have all been cited as likely contributors to DPP (Carskadon,
1999; Carskadon & Acebo, 2005; Crowley et al., 2007; Taylor,
Jenni, Acebo, & Carskadon, 2005; Wright, Gronfier, Duffy, &
Czeisler, 2005). These findings indicate that educators and par-
ents may not be able to simply restrict or encourage certain
behaviors for the purpose of reducing DPP. Telling adolescents
to go to bed earlier will not necessarily attenuate DPP, as certain
contributing factors are intrinsic to the development of healthy
adolescents.
Cognitive Correlates of the Sleep Cycle and
Implications for School Start Times
Multiple studies have shown that various forms of cognitive per-
formance fluctuate according to the time of day (e.g., May, 1999;
Petros, Beckwith, & Anderson, 1990; Yoon et al., 1999). The
two-process model introduced earlier, composed of the homeo-
static drive for sleep (process S) and the circadian wakefulness
process (process C), provides a useful paradigm for analyzing the
various stages of the sleep–wake cycle and how they relate to
various measures of cognitive performance.
In general, the accumulation and intensification of process S,
representing the homeostatic drive for sleep (also known as sleep
pressure), is associated with a decrease in cognitive performance
and an increase in perceived sleepiness (Schmidt et al., 2007).
However, in spite of the accumulation of sleep pressure during
waking hours, cognitive performance can remain stable and even
increase throughout the day, depending on the relative magni-
tude of process C (Cajochen, Blatter, & Wallach, 2004). This
established relationship between processes S and C and cognitive
performance often holds true; however, research into specific
components of cognition has revealed that the relationship
between the sleep–wake cycle and cognitive performance is
largely task dependent. Variables such as the duration, difficulty,
and cognitive domain of a task have been shown to influence the
specific effects of the sleep–wake cycle on cognitive performance
(Bonnet, 2000; Schmidt et al., 2007).
Although more research needs to be conducted regarding the
separate components of cognition and the sleep–wake cycle
(because various components of cognition have been shown to
differ in their responses to the influence of the sleep–wake cycle),
there are some confirmed findings that may be helpful in deter-
mining optimal school schedules for adolescents:
1. Synchronicity effects (i.e., when optimal performance is
generally achieved at the peak time of day as determined by
chronotypology) have been found in performance on mea-
sures of attention (Cajochen, Khalsa, Wyatt, Czeisler, &
Dijk, 1999; De Gennaro, Ferrara, Curcio, & Bertini,
2001; Kirby & D’Angiulli, 2009; Schmidt et al., 2007) as
well as in explicit and working memory (see Schmidt et al.,
2007). However, research on implicit memory—memory
that is acquired and expressed without conscious
mediation—has not found the same synchronous relation-
ship between the circadian rhythm and performance (May,
Hasher, & Foong, 2005); this finding will be discussed in
a later section.
2. Performance on tasks measuring executive functions
(which control the content and execution of cognitive pro-
cesses and protect cognitive activity from distractions)
demonstrates significant modulations by time of day. For
adolescents, optimal performance on such tasks occurs
later in the day (May, 1999; Yoon et al., 1999).
3. Sleep debt (cumulative sleep loss) also has been shown to
contribute to an inability to concentrate, memory lapses,
difficulty in accomplishing tasks that require planning or
following a complex sequence of actions, and a decrease in
creative thought (Carskadon, 1999; Dinges & Kribbs,
1991).
Based on these findings, it would seem plausible that setting
early school start times for adolescents sufficiently impairs their
ability to effectively perform school-related tasks. Early start
times are causing adolescents to accumulate sleep debt during
the week (Carskadon, 1999; Noland, Price, Dake, & Telljohann,
2009)—some studies have reported that adolescents receive
approximately two hours of sleep less than the recommended
amount nightly as a result of early start times (Mitru et al.,
2002). Furthermore, these start times are causing students to
attend class before their circadian wake-promoting process is
fully engaged. This combination of factors is likely leading to
deficits in academic performance and attendance, as well as to
behavioral problems (Carskadon, 1999; Dahl, 1999; Owens,
Belon, & Moss, 2010; Wahlstrom, 2010). The results imply that
educational ReseaRcheR
58
we may be able to optimize adolescent academic achievement
and improve not only adolescents’ cognitive but also their affec-
tive states by delaying school start times so that their circadian
rhythms are synchronized with their daily school schedule.
Changing School Start Times: In Practice
Because of the growing evidence suggesting that the evening pref-
erence chronotype of adolescents conflicts with early school start
times, in the 1997–1998 school year the Minneapolis Public
School District changed the start times of seven high schools
from 7:15 a.m. to 8:40 a.m. Classes that were formerly dismissed
at 1:45 p.m. were rescheduled to end at 3:20 p.m. The
Minneapolis School Board asked the Center for Applied Research
and Educational Improvement to examine the impact of this
change in school start times over the long term (see Wahlstrom,
2002, for details). The findings of this study indicate that the
delay in school start times of 1 hour and 25 minutes in the
Minneapolis school district resulted in several positive changes in
the student population (Center for Applied Research and
Educational Improvement, 1998a, 1998b; Wahlstrom, 2002).
Perhaps most important is that the students, rather than simply
staying up later at night in response to having more time to sleep
in the morning, used the extra time in the morning for actual
sleep (Center for Applied Research and Educational Improvement,
1998b). Furthermore, the attendance rates for students in Grades
9, 10, and 11 improved from the 1995 school year (early start
time) to the 2000 school year, with the greatest increase in atten-
dance noted in Grade 9 students (Wahlstrom, 2002). Parents,
teachers, and school administrators also noted significant
improvements in student demeanor and a reduction in disciplin-
ary problems (Center for Applied Research and Educational
Improvement, 1998a; Wahlstrom, 2002), which is unsurprising
considering that the students slept 5 hours more per week when
school started later. These results support the conclusion that stu-
dents are emotionally and cognitively better suited for the later
start time.
Those who oppose the change to school start times may cite
the failure to find any significant increases in letter grades as a
lack of evidence that the change is worthwhile. However, letter
grades are not necessarily an accurate and objective measure of
policy efficacy; factors such as grade inflation, changes in curricu-
lum, teacher and administrator turnover, changes in the assess-
ment process, and other complicating factors compromise the
reliability of letter grades as a valid measure. Wahlstrom (2002)
notes that “there are other equally important measures of impact,
such as student physical and emotional well being, benefits asso-
ciated with teaching and learning, and improved family relation-
ships” (pp. 18–19), all of which showed marked improvement
after the change in start times.
The Minneapolis study serves as an important example of
changing school start times because it is the most thorough lon-
gitudinal study to date that demonstrates the lasting effects of
changing school start times. It is important to note that other
schools that have delayed their school start times have demon-
strated similar results, which include an increase in amount of
sleep, a positive change in attitudes, and increased alertness
(National Sleep Foundation, 2005a, 2005c, 2005e; Owens et al.,
2010).
Costs and Other Problematic Issues Associated With
Changing Start Times
The monetary costs associated with changing school start times are
difficult to generalize, as bus schedules and busing policies (i.e., to
whom does the school provide transportation?), the geographical
distribution of the student population, the number of schools in
the district, and other interdistrict differences all contribute to the
costs associated with the changing of school start times. The most
commonly cited monetary expense associated with changing
school start times is the cost of changing bus schedules; however,
for the aforementioned reasons, these costs can vary significantly
from district to district. For example, the suburban school district
of Edina, Minnesota, and the urban district of Minneapolis,
Minnesota, did not find that delaying start times increased trans-
portation costs, as they used the same bus routes as before changing
school start times and simply delayed the schedules accordingly
(Wahlstrom, 2002).
However, a school district in Fairfax County, Virginia (com-
posed of nearly 200 schools), estimated transportation costs asso-
ciated with changing school start times at anywhere between
$4 million and $40 million (Chandler, 2009). Thus there are
systematic differences between districts that can affect the mon-
etary costs. To determine the monetary costs of changing school
start times, a district-specific analysis must be conducted.
However, social costs and nonmonetary issues associated with
a change in school start times must also be considered before an
informed decision regarding changing start times can be made.
Researchers, educators, parents, and the general media have cited
some complications that may result from changing school start
times that should be considered before implementation:
1. Rearranging the transportation schedules has in some cases
resulted in high school students being picked up after other
student populations (i.e., middle school and elementary
school students; Kubow, Wahlstrom, & Bemis, 1999); as a
result, these high school students often arrived at school
late (Black, 2000), negating the effect of late start times in
reducing tardiness (Kubow et al., 1999).
2. Rush-hour congestion may become worse if school buses
are on the road later in the day in the morning and after-
noon, effectively lengthening the bus ride for many stu-
dents and potentially creating traffic problems for the rest
of the community (W.A.K.E., n.d.).
3. In many families, older siblings babysit younger family
members after school, but by keeping high school students
in school later in the day, families can no longer depend
on their high school–aged children to care for younger
siblings. This may force parents to seek other childcare ser-
vices, which may not be available or affordable (Wrobel,
1999).
4. Teachers’, administrators’, and other employees’ schedules
may be inconveniently disrupted (W.A.K.E., n.d.).
5. Extracurricular and athletic activities may be negatively
affected. Some schools have had to shorten extracurricular
activities as a result of the later school hours. Furthermore,
because of interschool differences in class schedules, stu-
dents from schools with later start times have sometimes
sought early dismissal from class (W.A.K.E., n.d.). In an
MaRch 2011 59
effort to reduce the negative impact on extracurricular and
athletic activities, some schools have rescheduled them to
take place in the morning, effectively nullifying the reason
for delaying start times in the first place.
6. Student participation in other non-school-related activities
may suffer. With delayed school times, students may find
it difficult to participate in out-of-school activities such as
music lessons, martial arts, dance, gymnastics, and so on
(W.A.K.E., n.d.).
7. Students who have part-time jobs may also be adversely
affected. Employers may be reluctant to hire students who
cannot start work earlier in the day, and later school hours
may limit the time when some students can earn much-
needed income (Wrobel, 1999).
It is important to point out that of the above complications,
numbers 2, 4, 5, and 6 are based on speculation from the Fairfax
County School District’s deliberations. Although all of these con-
cerns are valid and legitimate, there is evidence that some of those
voiced by districts in deliberations can be minimized. As will be
shown in the next section, school boards that have introduced
later start times have generally been able to mitigate these poten-
tial problems through prudent and informed decision-making.
Implementing Change
School boards that have successfully implemented later start
times for the benefit of their adolescent students accounted for
most of the aforementioned concerns before the inception of any
schedule change. We have divided the concerns into the follow-
ing three categories for the purpose of examining how these suc-
cessful changes were implemented: transportation issues,
non–student body scheduling issues, and extracurricular issues.
Transportation
As previously mentioned, the transportation issues that school
districts may encounter will vary according to the features of each
district. Thus it is difficult to predict the transportation issues
that any one district will encounter without a specific investiga-
tion into the features of that district. In instituting a change in
school start times, some U.S. districts have had to change ele-
mentary and middle school hours of operation as well, in an
effort to minimize costs (National Sleep Foundation, 2005a,
2005d). The transportation directors in the Minneapolis study
emphasized the importance of staggering the start times of high
schools, middle schools, and elementary schools to maximize the
efficiency of funds allocated to transportation (Center for Applied
Research and Educational Improvement, 1998a). This complete
restructuring of transportation scheduling and school hours
within a district has the potential to cause significant distur-
bances to the community as a whole.
Some U.S. school boards have been able to change their
school start times without an increase in transportation costs sim-
ply by delaying the bus schedule (Wahlstrom, 2002) or switching
elementary schedules with high school schedules (National Sleep
Foundation, 2005e). One school board saved $750,000 annually
by removing transportation altogether for high school students,
who instead used the public bus system (National Sleep
Foundation, 2005b). And busing is not an issue in some school
districts, as many high school students already use public trans-
portation. Because the factors that affect transportation vary
from region to region, some school boards have appointed a
transportation subcommittee to evaluate the options (e.g.,
National Sleep Foundation, 2005a), a practice that has proved
successful. An analysis of the public transportation system and
school schedules in each district is necessary if planners are to
understand the consequences that may result from changing
school start times, especially for students who are transported by
family members or who carpool.
Non–Student Body Daily Scheduling
Schedule disruptions for families, teachers, school administra-
tors, and school staff that can result from changed school start
times, including disruptions related to students’ extracurricular
activities and outside jobs, are of considerable concern. Schools
that have successfully implemented a start time change have
demonstrated that it is also important to ensure that the public is
well informed about and included in the decision-making pro-
cess (e.g., National Sleep Foundation, 2005a, 2005d, 2005e).
School boards that have failed to consult community members
before announcing a planned change to start times have met with
staunch opposition (National Sleep Foundation, 2005c). The
importance of allowing everyone who will be affected by a change
in school start times to participate in the decision-making process
is paramount if the initiative is to be accepted by the community
(Wahlstrom, 2002). The findings of the Minneapolis study show
that an overwhelming majority of teachers, administrators, and
parents were in favor of the change in start times following its
implementation (Center for Applied Research and Educational
Improvement, 1998a). However, the findings also suggest that
students with later start times tend to work less throughout the
week, on both schoolwork and paid jobs, than students with early
start times, although possibly for reasons other than the differ-
ences in start times (Center for Applied Research and Educational
Improvement, 1998b).
Extracurricular Activities
Many who oppose changing school start times cite the disruption
of extracurricular activities as a prohibiting factor. Some school
boards have successfully implemented a start time change without
disrupting extracurricular activities—and, ironically, without hav-
ing to schedule after-school activities before school—simply by
scheduling events later. In fact, results from the Minneapolis study
show that later start times did not significantly affect student par-
ticipation in after-school activities (Center for Applied Research
and Educational Improvement, 1998b). The only problem was that
some children were pulled out of class early for away-from-school
sporting events (e.g., Center for Applied Research and Educational
Improvement, 1998a; National Sleep Foundation, 2005e).
Perhaps the most important consideration is that the schools
that have successfully delayed school start times with minimal
complications had adequate time to prepare, which they spent
engaged in research, policy analysis, and a healthy discourse with
the public. However, it should be noted that, in any school dis-
trict where the start times are changed, it is likely that those
directly and indirectly involved in the school system will need to
make some degree of sacrifice for the benefit of the students.
educational ReseaRcheR
60
Potential Alternative or Enhancement to Changing
Start Times
As previously discussed, the influence of the circadian rhythm on
time-of-day modulations in performance of memory tasks varies
according to the task (Schmidt et al., 2007). More specifically, it
was found that implicit memory task performance was best at the
nonpeak time of day according to the circadian oscillation (May
et al., 2005). These findings suggest that a restructuring of class
schedules so that adolescents are practicing fluency-based skills—
such as reading aloud or rehearsing music—in the morning may
be more beneficial than trying to have them assimilate new
knowledge early in the day. Such arrangements may mitigate
some performance deficiencies that result from the misalignment
between the typical adolescent circadian rhythm and early classes.
However, this kind of restructuring would not have any foresee-
able effect on the sleep deprivation that students experience as a
result of early school start times; therefore, the cognitive and
behavioral problems associated specifically with sleep debt would
not likely improve as a result of this change. Restructuring of
classes or activities to take into account the time-of-day differ-
ences in implicit memory performance may be most effective if
used in combination with delaying of school start times; however,
more research with regard to the implicit memory performance
of adolescents throughout the day needs to be conducted before
any claims to this effect can be validated.
Conclusion
There is a sufficient body of evidence showing that adolescents
experience changes in social context and physiology at the onset
of puberty that cause them to develop an evening preference
chronotype, predisposing them to go to sleep later at night and
wake up later in the morning than children and adults (Crowley
et al., 2007); this evening preference chronotype is not congruent
with early school start times. The asynchrony between school
start times and adolescent circadian rhythms can lead to an accu-
mulation of sleep debt (Carskadon, 1999, 2002), is associated
with cognitive impairments, and appears to adversely affect
behavior, subjective levels of sleepiness, subjective levels of
depression, and school-related variables such as attendance and
tardiness (Dahl, 1999; Dinges & Kribbs, 1991; Wahlstrom,
2002). In accordance with this body of research, some schools
have delayed school start times, with promising results.
Adolescents who attend schools with later start times have been
found to sleep more per night, resulting in a more positive tem-
perament, fewer behavioral problems, and an increase in corre-
lates of good academic performance (e.g., attendance, continuous
enrollment; Center for Applied Research and Educational
Improvement, 1998a; Owens et al., 2010; Wahlstrom, 2002).
Instituting a delay in school start times may be accompanied
by complications related to transportation costs in some school
districts, although this is a much lesser issue for high schools that
already require students to use alternative transportation meth-
ods (e.g., public transportation, car pools). In addition, changes
in start times are associated with disruptions of extracurricular
activities and other social and leisure activities in many school
districts. However, with sufficient strategizing and preparation,
the inconvenient consequences of changing school start times
can be attenuated. There is evidence that with adequate planning
and preparation, school boards have been able to delay school
start times at acceptable monetary cost (given the enormous
potential payoff ) and tolerable disruption of community
functioning.
REFERENCES
Black, S. (2000). A wake-up call on high-school starting times. Education
Digest, 66(4), 33–38.
Blatter, K., & Cajochen, C. (2007). Circadian rhythms in cognitive per-
formance: Methodological constraints, protocols, theoretical under-
pinnings. Physiology and Behavior, 90, 196–208.
Bonnet, M. H. (2000). Sleep deprivation. In W. C. Dement (Ed.),
Principles and practice of sleep medicine (3rd ed., pp. 53–71).
Philadelphia: Saunders.
Cajochen, C., Blatter, K., & Wallach, D. (2004). Circadian and sleep–
wake dependent impact on neurobehavioral function. Psychologica
Belgica, 44, 59–80.
Cajochen, C., Khalsa, S. B., Wyatt, J. K., Czeisler, C. A., & Dijk, D. J.
(1999). EEG and ocular correlates of circadian melatonin phase and
human performance decrements during sleep loss. American Journal
of Physiology, 277, 640–649.
Carskadon, M. (1999). When worlds collide: Adolescent need for sleep
versus societal demands. Phi Delta Kappan, 80(5), 348–353.
Carskadon, M. (2002). Adolescent sleep patterns: Biological, social, and
psychological influences. Cambridge, UK: Cambridge University Press.
Carskadon, M. A., & Acebo, C. (2005). Intrinsic circadian period in
adolescents versus adults from forced desynchrony. Sleep, 28(Abstract
supplement):A71.
Center for Applied Research and Educational Improvement. (1998a).
School start time study. Final report summary. Retrieved from http://
www.cehd.umn.edu/carei/Reports/summary.html
Center for Applied Research and Educational Improvement. (1998b).
School start time study. Technical report: Vol. II. Analysis of student survey
data. Retrieved from http://www.cehd.umn.edu/carei/Reports/
summary.html
Chandler, M. A. (2009, January 6). Fairfax plan would delay high school
start at no cost. Washington Post. Retrieved from http://www.sleepdoc.
com/images/linkfiles/Fairfax%20plan%20would%20delay%20
High%20school%20start%20at%20no%20cost%20010709.pdf
Crowley, S. J., Acebo, C., & Carskadon, M. A. (2007). Sleep, circadian
rhythms, and delayed phase in adolescence. Sleep Medicine, 8,
602–612.
Dahl, R. E. (1999). The consequences of insufficient sleep for adoles-
cents: Links between sleep and emotional regulation. Phi Delta
Kappan, 80, 354–359.
De Gennaro, L., Ferrara, M., Curcio, G., & Bertini, M. (2001). Visual
search performance across 40 h of continuous wakefulness: Measures
of speed and accuracy and relation with oculomotor performance.
Physiology and Behavior, 74, 194–204.
Dement, W. C., & Vaughan, C. (1999). The promise of sleep: A pioneer
in sleep medicine explores the vital connection between health, happiness,
and a good night’s sleep. New York: Delacourt.
Dinges, D. F., & Kribbs, N. B. (1991). Performing while sleepy: Effects
of experimentally-induced sleepiness. In T. H. Monk (Ed.), Sleep,
sleepiness and performance. Human performance and cognition
(pp. 97–128). Oxford, UK: John Wiley.
Edgar, D. M., Dement, W. C., & Fuller, C. A. (1993). Effect of SCN
lesions on sleep in squirrel monkeys: Evidence for opponent processes
in sleep–wake regulation. Journal of Neuroscience, 13, 1065–1079.
Fischer, F. M., Radosevic-Vidacek, B., Koscec, A., Teixeira, L. R.,
Moreno, C. R., & Lowden, A. (2008). Internal and external time
MaRch 2011 61
conflicts in adolescents: Sleep characteristics and interventions. Mind,
Brain, and Education, 2, 17–23.
Giannotti, F., Cortesi, F., Sebastiani, T., & Ottaviano, S. (2002).
Circadian preference, sleep and daytime behaviour in adolescence.
Journal of Sleep Research, 11, 191–199.
Kirby, M., & D’Angiulli, A. (2009). Timing (not just amount) of sleep
makes the difference: Event-related potential correlates of delayed
sleep phase in adolescent female students. In N. A. Taatgen & H. van
Rijn (Eds.), Proceedings of the 31st Annual Conference of the Cognitive
Science Society. Austin, TX: Cognitive Science Society. Retrieved from
http://csjarchive.cogsci.rpi.edu/Proceedings/2009/index.html
Kryger, M. H., Roth, T., & Dement, W. C. (Eds.). (2000). Principles and
practice of sleep medicine (3rd ed.). Philadelphia: W. B. Saunders.
Kubow, P. K., Wahlstrom, K. L., & Bemis, A. E. (1999). Starting time
and school life: Reflections from educators and students. Phi Delta
Kappan, 80, 366–371.
May, C. P. (1999). Synchrony effects in cognition: The costs and a ben-
efit. Psychonomic Bulletin and Review, 6, 142–147.
May, C. P., Hasher, L., & Foong, N. (2005). Implicit memory, age, and
time of day: Paradoxical priming effects. Psychological Sciences, 16,
96–100.
Millman, R. P., Working Group on Sleepiness in Adolescents/Young
Adults, & AAP Committee on Adolescence. (2005). Excessive sleepi-
ness in adolescents and young adults: Causes, consequences, and
treatment strategies. Pediatrics, 115, 1774–1786.
Mitru, G., Millrood, D. L., & Mateika, J. H. (2002). The impact of
sleep on learning and behavior in adolescents. Teachers College Record,
104, 704–726.
National Sleep Foundation. (2005a). Changing school start times:
Arlington, Virginia. Retrieved from http://www.sleepfoundation.org/
site/c.huIXKjM0IxF/b.2511895/k.FAA3/Changing_School_Start
_Times_Case_Studies.htm
National Sleep Foundation. (2005b). Changing school start times: Denver,
Colorado. Retrieved from http://www.sleepfoundation.org/site/c
.huIXKjM0IxF/b.2511895/k.FAA3/Changing_School_Start
_Times_Case_Studies.htm
National Sleep Foundation. (2005c). Changing school start times: Fayette
County, Kentucky. Retrieved from http://www.sleepfoundation.org/
site/c.huIXKjM0IxF/b.2511895/k.FAA3/Changing_School_Start
_Times_Case_Studies.htm
National Sleep Foundation. (2005d). Changing school start times: Jessamine
County, Kentucky. Retrieved from http://www.sleepfoundation.org/
site/c.huIXKjM0IxF/b.2511895/k.FAA3/Changing_School_Start
_Times_Case_Studies.htm
National Sleep Foundation. (2005e). Changing school start times: Wilton,
Connecticut. Retrieved from http://www.sleepfoundation.org/site/
c.huIXKjM0IxF/b.2511895/k.FAA3/Changing_School_Start
_Times_Case_Studies.htm
National Sleep Foundation. (2006). Sleep in America Poll. Retrieved
from http://www.sleepfoundation.org/
Noland, H., Price, J. H., Dake, J., & Telljohann, S. K. (2009).
Adolescents’ sleep behaviors and perceptions of sleep. Journal of School
Health, 79, 224–230.
Owens, J. A., Belon, K., & Moss, P. (2010). Impact of delaying school
start time on adolescent sleep, mood, and behavior. Archives of
Pediatrics and Adolescent Medicine, 164, 608–614.
Petros, T. V., Beckwith, B. E., & Anderson, M. (1990). Individual dif-
ferences in the effects of time of day and passage difficulty on prose
memory in adults. British Journal of Psychology, 81, 63–72.
Schmidt, C., Collette, F., Cajochen, C., & Peigneux, P. (2007). A time
to think: Circadian rhythms in human cognition. Cognitive
Neuropsychology, 24, 755–789.
Taylor, D. J., Jenni, O. G., Acebo, C., & Carskadon, M. A. (2005).
Sleep tendency during extended wakefulness: Insights into adoles-
cent sleep regulation and behavior. Journal of Sleep Research, 14,
239–244.
Wahlstrom, K. (2002). Changing times: Findings from the first longitu-
dinal study of later high school start times. NASSP Bulletin, 86(633),
3–21.
Wahlstrom, K. (2010). School start time and sleepy teens. Archives of
Pediatrics and Adolescent Medicine, 164, 676–677.
W.A.K.E.: Worried About Keeping Extra-curriculars. (n.d.). Disruptions.
Retrieved from http://sites.google.com/site/wakefairfax/disruptions/
Wright, K. P., Jr., Gronfier, C., Duffy, J. F., & Czeisler, C. A. (2005).
Intrinsic period and light intensity determine the phase relationship
between melatonin and sleep in humans. Journal of Biological Rhythms,
20, 168–177.
Wrobel, G. D. (1999). The impact of school starting time on family life.
Phi Delta Kappan, 80, 360–364.
Yoon, C., May, C. P., & Hasher, L. (1999). Aging, circadian arousal pat-
terns, and cognition. In D. Park & N. Schwartz (Eds.), Cognitive
aging: A primer (pp. 151–170). Philadelphia: Psychology Press.
AUTHORS
MATTHEW KIRBY is a researcher and research assistant at Carleton
University, Institute of Interdisciplinary Studies, 1125 Colonel By
Drive, Ottawa ON K1S 5B6, Canada; matthewfkirby@gmail.com. His
research focuses on educational psychology, specifically, cross-cultural
and environmental extra-curricular determinants of performance and
learning in students.
STEFANIA MAGGI is an assistant professor at Carleton University,
Institute of Interdisciplinary Studies, 1125 Colonel By Drive, Ottawa
ON K1S 5B6, Canada; stefania_maggi@carleton.ca. Her research focuses
on social determinants of child and adolescent mental health.
AMEDEO D’ANGIULLI is an associate professor at Carleton University,
Institute of Interdisciplinary Studies and Department of Neuroscience,
1125 Colonel By Drive, Ottawa ON K1S 5B6, Canada; amedeo@
connect.carleton.ca. His research focuses on the neurocognitive basis for
optimal learning and performance in the classroom and brain-based
educational practices.
Manuscript received December 5, 2009
Revision received February 2, 2011
Accepted February 9, 2011
