ArticlePDF AvailableLiterature Review


The interest in the systematic study of the circadian typology (CT) is relatively recent and has developed rapidly in the two last decades. All the existing data suggest that this individual difference affects our biological and psychological functioning, not only in health, but also in disease. In the present study, we review the current literature concerning the psychometric properties and validity of CT measures as well as individual, environmental and genetic factors that influence the CT. We present a brief overview of the biological markers that are used to define differences between CT groups (sleep-wake cycle, body temperature, cortisol and melatonin), and we assess the implications for CT and adjustment to shiftwork and jet lag. We also review the differences between CT in terms of cognitive abilities, personality traits and the incidence of psychiatric disorders. When necessary, we have emphasized the methodological limitations that exist today and suggested some future avenues of work in order to overcome these. This is a new field of interest to professionals in many different areas (research, labor, academic and clinical), and this review provides a state of the art discussion to allow professionals to integrate chronobiological aspects of human behavior into their daily practice.
Circadian Typology: A Comprehensive Review
Ana Adan,
Simon N. Archer,
Maria Paz Hidalgo,
Lee Di Milia,
Vincenzo Natale,
Christoph Randler
Department of Psychiatry and Clinical Psychobiology, School of Psychology, University of Barcelona, Spain,
Institute for Brain,
Cognition and Behavior (IR3C), Spain,
Faculty of Health and Medical Sciences, University of Surrey, Surrey, UK,
Laboratório de
Cronobiologia do Hospital de Clinicas de Porto Alegre, Brazil,
Departamento de Psiquiatria, Faculdade de Medicina,
Universidade Federal do Rio Grande do Sul, Brazil,
School of Management, Central Queensland University, QLD, Australia,
Department of Psychology, University of Bologna, Bologna, Italy,
University of Education, Heidelberg, Germany.
The interest in the systematic study of the circadian typology (CT) is relatively recent and has developed rapidly in the
two last decades. All the existing data suggest that this individual difference affects our biological and psychological
functioning, not only in health, but also in disease. In the present study, we review the current literature concerning
the psychometric properties and validity of CT measures as well as individual, environmental and genetic factors
that influence the CT. We present a brief overview of the biological markers that are used to define differences
between CT groups (sleepwake cycle, body temperature, cortisol and melatonin), and we assess the implications
for CT and adjustment to shiftwork and jet lag. We also review the differences between CT in terms of cognitive
abilities, personality traits and the incidence of psychiatric disorders. When necessary, we have emphasized the
methodological limitations that exist today and suggested some future avenues of work in order to overcome these.
This is a new field of interest to professionals in many different areas (research, labor, academic and clinical), and
this review provides a state of the art discussion to allow professionals to integrate chronobiological aspects of
human behavior into their daily practice.
Keywords: Age, Circadian typology, Chronotype, Cognitive performance, Morningnesseveningness, Personality,
Psychiatric disorders, Sex, Shiftwork
Circadian rhythmic expression differs among individuals
and may be classified with the dimension/concept of cir-
cadian typology (CT), which consists of three chrono-
types (morning- [MT], neither- [NT] and evening-type
[ET]). CT is determined using a number of self-
assessment questionnaires that have been validated in
several countries. MT subjects go to bed early and wake
up early, and achieve their peak mental and physical
performance in the early part of the day. By contrast,
ET subjects go to bed and wake up late, and perform at
their best toward the end of the day and evening hours.
The phase lags in circadian rhythmic functions
between extreme groups range from 2 to 12 h, and this
has been observed both in biological and behavioral
parameters. About 40% of the adult population is
classified in one of the two extreme groups, while 60%
are NT.
The int er est in the systematic study of the CT is r ela tively
recent and has dev eloped rapidly in the two last decades.
All the exis ting data suggest that this individual differ ence
affects our biological and psychological functioning, not
only in health, but also in disease. The epidemiology of
CT is influenced by individual factors such as age and
sex, which impact on CT across the lifespan. Mor eover,
environmental factors such as the perinatal photoperiod
or light exposur e also demonstr ate an influence on the
development of CT . Differ ences in rhythmic expr ession
between MT and ET can also be found in personality
traits, habits and life-styles. Finally , there is little doubt
about the association between circadian rhythmicity and
psychiatric disorders; most studied ar e mood and ea ting
disorders, and addictions. Ther e is mounting evidenc e
chro nodisruption aspects, linked to genetic vulner ability
and poor life-s t yle choices associated with social jet lag.
* The authors were listed in alphabetical order, except for the corresponding author.
Address correspondence to Ana Adan, Department of Psychiatry and Clinical Psychobiology, School of Psychology, University of Barcelona,
Passeig Vall dHebron, 171, 08035 Barcelona, Spain. E-mail:
Submitted June 28, 2012, Returned for revision July 18, 2012, Accepted July 24, 2012
Chronobiology International, 29(9): 1153 1175, (2012)
Copyright © Informa Healthcare USA, Inc.
ISSN 0742-0528 print/1525-6073 online
DOI: 10.3109/07420528.2012.719971
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
In the present study, we review the current literature in
terms of its central findings and offer suggestions to over-
come the methodological limitations. In particular, we
review the psychometric properties and validity of CT
measures as well as individual, environmental and
genetic factors that influence the CT. We present a brief
overview of the biological markers that are used to
define differences between CT groups (sleepwake
cycle, body temperature, cortisol, melatonin), and we
assess the implications for CT and adjustment to shift-
work and jet lag. We also review the differences
between CT in terms of cognitive abilities, personality
traits and the incidence of psychiatric disorders.
The publication of the MorningnessEveningness Ques-
tionnaire (MEQ, Horne & Östberg, 1976), the Diurnal
Type Scale (DTS, Torsvall & Åkerstedt, 1980) and the Cir-
cadian Type Questionnaire (CTQ, Folkard et al., 1979)
ushered a renaissance to better understand individual
differences and their role in explaining biological func-
tion. These scales can be considered as first-generation
CT measures and have served as catalysts for the develop-
ment of several other scales. Subsequent work has been
driven by the classical issues that surround measurement
(reliability and validity) and the need for shorter
measures that can be used in large-scale data collection.
MEQ and the Reduced MorningessEveningness
The MEQ (Horne & Östberg, 1976) is the most widely
used morningness measure. Since its publication, the
MEQ has been cited 1039 times (Scopus, February
2012). The MEQ has 19 items, and the answer options
include using a visual analog scale and choo sing
between four or five answer options.
The MEQ was administered to 150 adults aged 1832
and the sample was gender balanced. Forty-eight of
these participants were randomly selected and 18 were
found to be MT and 10 were ET. These participants
recorded their oral temperature for 3 wk and their
sleepwake behavior. The oral temperature peaked at
19:30 h in MT, 20:25 h in NT and 20:40 h in ET. Hour-
by-hour analysis revealed no significant differences
between these groups but large individual differences
were observed. MT were found to go to sleep 99 min
before ET and woke 114 min earlier than ET, but there
were no differences in sleep duration. Surprisingly,
Horne and Östberg (1976) did not report scale reliability,
their justification for dropping three items or the cut-off
points to define the CT.
A number of large studies have reported the MEQ to
be a reliable measure across several countries. The
reliability coefficient ranged between .78 and .86 (Adan
& Natale, 2002; Chelminski et al., 1997; Neubauer,
1992) and the stability of the MEQ is strong (.88 .89)
over a 3-mo period (Larsen, 1985; Neubauer, 1992).
Validation studies support the distinction between CT
and behavior. These studies are discussed in the
section on cognitive abilities and performance.
The MEQ has received some criticism however. The
scale contains 19 items and may be considered lengthy
in some situations. A second comment is that the bulk
of the variance can be explained by fewer items. Adan
and Almirall (1991) found four items had little discrimi-
natory power and Zickar et al. (2002) reported MT
items provided the best discrimination. In addition, the
MEQ measures a multidimensional construct and, there-
fore, the use of a total score may not be appropriate.
These criticisms underpinned the development
of the reduced MorningessEveningness Questionnaire
(rMEQ) (Adan & Almirall, 1991). This scale contains
five items and is considered a pure measure of MT. The
correlation between the rMEQ and the MEQ ranges
from satisfactory to excellent (.69.90) and also demon-
strates good convergent validity (Adan & Almirall, 1991;
Caci et al., 2009; Chelminski et al., 2000). Natale et al.
(2006) reported the rMEQ discriminated MT and ET on
the basis of objectively recorded moto r activity.
Composite Scale of Morningness and the Morning Affect
The Composite Scale of Morningness (CSM) was devel-
oped based on a psychometric assessment of the MEQ,
DTS and CTQ administered to 501 North American stu-
dents. The CSM consists of 13 items; nine from the
MEQ and four from the DTS (Smith et al., 1989).
Smith et al. (1989) reported scale reliability of .87 and
international studies have reported similar estimates.
Caci et al. (2005a) examined the reliability of the CSM
in five countries and estimates ranged between .65 and
.91. The lowest coefficient was found in Thailand, and
this may be explained by the fact that several items
were modified to reflect local activity timing. Using self-
report data, Smith et al. (1989) found significant differ-
ences between CT and bed/wake times, preferred class
times and times when students felt at their mental and
physical best. These results have been replicated else-
where (Caci et al., 1999; Randler, 2009a). Using objective
data, Guthrie et al. (1995) reported CT differences in
sleep behavior, study time and class performance.
The factor structure for the CSM is less clear (Di Milia
& Bohle, 2009). However, a brief and reliable morning
affect (MA) factor has been identified, and shows good
reliability (.76.85) in several countries (Caci et al.,
2009). The MA scale has been validated against self-
reported alertness in a student (Di Milia & Bohle, 2009)
and working sample (Di Milia & Muller, 2012). Further
studies are needed to confirm the utility of the MA scale.
Preferences Scale, CTQ and the Munich Chronotype
The Preferences Scale (PS) was developed to address
several concerns over the CSM (Smith et al., 2002). One
problem is the CSM items assume all people work a
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
diurnal schedule. Permanent night- and shiftworkers
may find it difficult to answer how they feel like when
they wake in the morning . Second, the varied response
formats and the different number of answer choices
may combine to increase the measurement error
(Zickar et al., 2002) and third, the reference to fixed
time points does not allow for cultural differences in
activity timing (Pornpitakpan, 1998).
The PS addresses these criticisms by using a scale that
does not make reference to time of day. Instead, partici-
pants rate their preference relative to most people on a
five-point scale. Across six countries, Smith et al. (2002)
reported reliability coefficients between .80 and .90.
The PS has demonstrated good convergent validity with
the CSM (.69.83) and .82 with the MEQ (Osland et al.,
2011). In terms of construct validity, the literature
reports significant differences in sleep/wake timing and
alertness by time of day (Bohle et al., 2001; Smith et al.,
2002). Di Milia (2005) was unable to replicate the
posited PS factor structure but developed a six-item
measure that was replicated in a student and
working sample.
The focus of the CTQ is to assess the amplitude and
stability of the circadian rhythms (Folkard et al., 1979).
The underlying hypotheses –“better adjustment might
be shown by people with; (a) low amplitude rhythms,
and (b) flexible or non-stable rhythms (p. 80). The
measurement properties of the CTQ were poor (Smith
et al., 1989) but the rCTI (Di Milia et al. 2004) appears
to be a more promising measure. The rCTI has been
replicated in a working sample, scale reliability is good
and has shown construct validity; vigorous ty pes were
significantly more alert across the day and flexible types
were significantly more alert in the evening (Di Milia
et al., 2005).
The Munich Chronotype Questionnaire (MCTQ) is the
most recent CT instrument measure. Roenneberg et al.
(2003) argue that assessing CT requires information
regarding genetic predisposition, specific timing of
sleep onset and offset by work or free days and light
exposure; information not collected by existing measures.
The MCTQ determines CT according to the mid-point of
sleep (onset and offset) calculated on days off (MSF).
Weekend sleep takes into account the fact that ET
accumulate a sleep debt which is repaid on weekends
(Roenneberg et al., 2007). The mid-point is considered
the best indicator of melatonin onset (Terman et al.,
Methodological Limitations
Most of the effort has been directed at improving the
reliability of CT instruments but the same rigor has not
been applie d to scale validation. We summarize the
main methodological limitations so that future studies
may provide better quality findings.
There appear to be no studies that have employed a
representative sample (Caci et al., 2005a). The MCTQ
has a large database and may be considered representative
but it is a self-selected sample rather than one that is ran-
domly drawn. A second limitation is many studies are
gender biased and rely on young student samples. These
sample characteristics limit the ability of CT scales to gen-
eralize to worker samples. A third limitation is a reliance
on cross-sectional studies collecting self-reported data.
Studies that collect the dependent and independent vari-
ables at the same time may inflate the risk of common
method variance. Future studies should employ a split
design methodology that results in the collection of inde-
pendent and dependent data at different time points
(Willis et al., 2008). Di Milia et al. (2008, 2012) have
shown that self-reported data are not necessarily biased
but obtaining some objective data is recommended.
Fourth, too few studies have assessed the predictive
value of CT in operational settings over the longer term
(Kaliterna et al., 1995). Fifth, the literature has not
explored the inter-relationship between the various
circadian rhythm parameters, and perhaps a weighted
combination of them may better explain adjustment in
nightworkers (Smith et al., 1989). Finally, from an oper-
ational perspective, an unresolved problem is defining
the cut-off criteria to identify extreme CT.
There is evidence suggesting CT is influenced by individ-
ual factors, such as age and sex, as well as several
environmental factors including the photoperiod at
birth, the altitude/latitude of residence and the subjects
exposure to light.
After the end of adolescence, morningness scores tend to
increase with age (Kim et al., 2010; Merik ango et al., 2012;
Monk & Kupher, 2000; Paine et al., 2006; Park et al., 2002;
Taillard et al., 2004; Tonetti et al., 2008). The age-related
shift to morningness is observed after controlling for
demographic and socioeconomic factors, and correlates
with most of the circadian functions with biological and
behavioral circadian parameters (Klei et al., 2005; Mon-
grain et al., 2004; Monk et al., 2004; Taillard et al., 2011;
Zimmermann, 2011). As subjects grow older, there is a
tendency to go to bed and wake up earlier, and to
present the highest levels of activation at an earlier
time, and even more so from the age of 50 on.
Adolescence (1217 yr) is a critical period when a shift
in CT is seen to occur from MT to a more pronounced
tendency to ET (Achari & Pati, 2007; Borisenkov et al.,
2010; Kim et al., 2002; Randler, 2008a, 2011; Roenneberg
et al., 2004; Russo et al., 2007; Tonetti et al., 2008). Even-
ingness increases from 12 to 1520 yr before a shift to MT
becomes apparent. Eveningness may be considered as a
biological marker for the end of adolescence (Randler,
2011; Roenneberg et al., 2004). In women, the peak of
maximum eveningness appears earlier (Randler, 2011;
Tonetti et al., 2008). This phenomenon may be inter-
preted as associated with pubertal development
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
(Hagenauer et al., 2009), but it is also clearly influenced
by social and family factors, such as the school year or
parental control on the subjects schedule (Gau &
Soong, 2003; Randler et al., 2009). It is precisely at
these ages that the development of an extreme ET
should be controlled, since otherwise it could favor an
individual functioning with fewer healthy habits,
(Minors et al., 1989) and it could also interfere in the
adaptation to the socio-environmental demands of
the activity (Besoluk et al., 2011; Taylor et al., 2011;
Tzischinsky & Shochat, 2011).
The possibility that sex is related to CT has been
approached in several studies. Many studies report a
larger proportion of ET is found among males, while
MT is more commonly observed among females (Adan
& Natale, 2002; Borisenkov et al., 2012; Natale & Di
Milia, 2011; Randler, 2011; Roenneberg et al., 2004;
Tonetti et al., 2011). These results tend to be found in
large samples that have used the MEQ. However, some
studies have not found any gender differences (Paine
et al., 2006; Zimmermann, 2011), and others reported
ET was higher in females (Merikango et al., 2012).
The difference in favor of MT in adult women is in
accordance with the empirical observation that the
diurnal variations in women produce a phase advance
with respect to men (Adan & Sánchez-Turet, 2001; Park
et al., 2002), with the magnitude depending on the con-
sidered parameter. Moreover, the intrinsic circadian
period was significantly shorter in women than in men
and a significantly greater proportion of women have
intrinsic periods shorter than 24 h (Duffy et al., 2011).
This difference can be explained by the control of the cir-
camensual rhythmicity associated with the menstrual
cycle in women, which would act against the intensity
of the rhythmic control of circadian periodicity (Adan &
Natale, 2002). This argument is supported by the fact
that sex differences on CT disappear following meno-
pause in women (Roenneberg et al., 2004; Tonetti et al.,
2008). Although there are fewer data available, adoles-
cent samples also find more ET and a delay of sleep
timing in boys compared with girls (Borisenkov et al.,
2010, 2012; Tonetti et al., 2008).
Photoperiod at Birth, Longitude and Altitude
With respect to the influence of the photoperiod to which
the individual is exposed during the first months of life,
the circadian organization of the subjects born with a
short photoperiod (autumnwinter) tends to be more
MT, while for those with a long photoperiod (spring
summer), it tends to be more ET (Mongrain et al.,
2006b; Natale & Adan, 1999; Natale et al., 2002; Natale
& Di Milia, 2011). This is also observed in adolescent
samples (Borisenkov et al., 2012; Tonetti et al., 2011),
and is more clearly seen in boys or men than in girls
or women.
However, studies with Asians participants do not find
differences between adolescents and adults (Achari &
Pati, 2007; Harada et al., 2011; Takao et al., 2009), and
even the prevalence of sex is the opposite (Gaina et al.,
2006). This might be due to biological differences
(genetic and ocular photosensitivity) with respect to
Caucasians (Tonetti et al., 2011). Taking into account
the effect of the photoperiod in the endogenous rhythmic
expression, it should be advised that children born in
autumn/winter be exposed to light (natural or artificial)
despite the weather conditions, while preventing those
born in the spring/summer from having an excess of
light exposure (Adan et al., 2008).
The longitude and latitude of residence is also a rel-
evant factor in CT. The CT appears to be associated
with the geographical coordinates (eastwest and
northsouth) and in urban areas compared with rural
areas in both adults and adolesc ents (Borisenkov et al.,
2010, 2012; Natale & Di Milia, 2011; Randler, 2008a).
Broadly, there are more MT toward the East and North
and in rural municipalities.
Finally, in the same place of residence , ET obtained
lower levels of light exposure during the daytime and
higher during the nighttime, together with a more
indoor pattern and irregular life-style than MT (Gaina
et al., 2011; Harada et al., 2011; Martin et al., 2012;
Vollmer et al., 2012). With this, there is a decrease in
the strength of the light/dark zeitgeber, possibly related
to some alterations and disorders, which are commented
in the psychiatric disorder section of this review.
Circadian rhythmicity can be identified in a number of
biological markers. These markers include the sleep
wake cycle, body temperature and the hormones melato-
nin and cort isol. In this section, we will focus on these as
well as genetic-based evidence.
SleepWake Timing
Sleep is a widespread biological process that is known to
have underlying beneficial functions. Sleep is a complex
phenotype but has some well-defined characteristics that
show high levels of inheritance (e.g., Andretic et al.,
2008). An inevitable consequence of any inherited,
trait-like characteristic is that it will show inter-individual
variation in related phenotypic parameters measured
within a pop ulation due to associated inherited genetic
variation, and this is true for sleepwake timing and dur-
ation (Groeger et al., 2004; Roenneberg et al., 2007).
In humans, periods of sleep and wake are determined
by the interaction between a homeostatic process that
counts the build-up of sleep pressu re during wakefulness
and its dissipation during sleep, and the intrinsic circa-
dian clock that produces an oscillatory wake-promoting
signal, which has its peak just before sleep and its nadir
shortly before wake (Borbely, 1982). The importance of
the circadian oscillator in this interaction is well
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
understood because desynchronization of the two pro-
cesses is associated with sleep disruption during shift-
work (Akerstedt, 2003), jet lag (Jamieson et al., 2001),
and in free-running blind people (Leger et al., 1999).
It has been shown that the measurement of CT by
diurnal preference questionnaires is directly correlated
with an individuals intrinsic circadian rhythms. In the
development of the MEQ, Horne and Östberg (1976)
reported an earlier peak of temperature in MT. This has
been confirmed in subsequent, well-controlled forced
desynchrony (FD) studies, which showed that morning-
ness is associated with a shorter intrinsic circadian
period (Duffy et al., 2001; Hasan et al., 2012), as well as
an earlier core body temperature phase and earlier
wake times (Duffy et al., 2001).
Because of the interaction between sleep and circa-
dian systems and the correlation between circadian
rhythms and CT, it should be expected that inter-individ-
ual variability in CT would be associated with differences
in sleepwake timing. Indeed, a consistent finding from
studies that have measured CT in humans is that there
is a strong association between CT and sleepwake
timing, such that MT wake up and go to bed earlier
than ET (Horne & Östberg, 1976; Robilliard et al., 2002;
Taillard et al., 2004). In addition, it has also been reported
that differences in CT are associated with differences in
the regulation of sleep homeostasis as measured by the
EEG, such that MT have a higher percentage of stage 1
sleep, more spectral power in the sigma range during
NREM sleep (an indicator of cortical arousal) and a
faster decay of NREM slow-wave activity (Mongrain
et al., 2005a, 2005b, 2006a). These altered characteristics
of sleep homeostasis point toward a higher rate of sleep
dissipation in MT that may also represent an underlying
homeostatic contribution to sleepwake timing differ-
ences between MT and ET.
It should be noted, however, that the relationships
between CT, circadian rhythms and sleepwake timing
are more complex and also change with age. In young
MT, there is an earlier circadian phase (of melatonin
and core body temperature) compared with ET, but the
interval between the circadian phase and wake time is
longer in MT than in ET (Duffy et al., 1999). This
means that whereas ET wake up at a later clock time,
MT wake up at a later circadian phase. However, this
relationship changes with age, such that older MT have
an earlier phase than young MT, but also a shorter inter-
val between the circadian phase and wake time (Duffy
et al., 1999). This means that older MT are more similar
to younger ET in this respect. With age, the increase in
morning preference and earlier wake times is
accompanied by age-related changes in sleep capacity,
such that older people have reduced baseline daytime
sleep propensity and shorter durations of sleep in labora-
tory conditions, compared with younger people
(Klerman & Dijk, 2008). These age-related findings
again emphasize the important interactions between
sleep and circadian processes in the determination of CT.
Several investigations have found no overall associ-
ation between CT and sleep duration (e.g., Horne &
Östberg, 1976; Robilliard et al., 2002). However, one
on-going survey of CT using the MCTQ shows clear
workdayweekend differences in sleep duration, but
also demonstrates an interesting relationship between
sleep duration and sleep timing (Roenneberg et al.,
2007). Consistent with previous studies, the data show
no evidence for an overall correlation between sleep dur-
ation and sleep timing, with short (6 h) and long (9h)
sleepers having equal CT distributions, suggesting that
these two traits are indepe ndent. However, the data
nicely show the distinct effects of work, with a negative
correlation between CT and sleep duration during work-
days (late types sleep for less), and a positive correlation
at weekends (late types sleep for longer). Thus, CT differ-
ences in sleep debt accumulated during workdays can
affect sleep duration and timing.
Finally, the contribution of external, social factors to
CT should not be underestimated. Because the two pro-
cesses of circadian rhythms and sleep homeostasis inter-
act so precisely to determine sleep/wake timing,
desynchronization of the two can lead to metabolic and
cardiovascular problems and also circadian rhythm
sleep disorders, such as shiftwork disorder and delayed
sleep phase disorder (DSPD) (Luyster et al., 2012; Sack
et al., 2007). Social factors can lead to this desynchroniza-
tion during normal working lives. ET who go to bed late
but rise early because of work times or family commit-
ments will wake up at a time that is out of synch with
their circadian clock and also accumulate a sleep debt
during the week. This social jet lag is common in our
current society and has been shown to be associated
with metabolic disorders and depression (Levandovski
et al., 2011; Roenneberg et al., 2012; Wittmann et al.,
2006). Thus, there needs to be careful consideration of
CT and work schedules and how these interact to deter-
mine health outcomes.
Body Temperature
Several studies have shown that MT have an earlier circa-
dian temperature phase than ET measured both by rectal
(Duffy et al., 1999) and oral temperature (Gupta & Pati,
1994). This phase difference is around 2 h both in
normal daynight cond itions (Baehr et al., 2000)
and during a constant routine (Ke rkhof & Van Dongen,
This circadian phase difference is explained by
endogenous factors and not by differences in sleep sche-
dules. In normal daynight conditions, the phase angle,
defined as the interval between the trough of the body
temperature and wake up time, is shorter in ET compared
with MT. However, these differences tend to disappear in
experimental conditions (Mongrain et al., 2004). There-
fore, it is possible to conclude that social demands have
a large impact on when people wake up in relation to
their circadian clock, inducing the disadvantages this
may cause to ET.
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
Because women tend to have both a longer sleep dur-
ation than men (Natale et al., 2009) and a higher predis-
position to morningness (Adan & Natale, 2002), women
show a longer phase angle in comparison to men (Mon-
grain et al., 2004).
ET show a higher amplitude in circadian fluctuation of
body temperature in comparison to MT (Baehr et al.,
2000). The strength of the human circadian system is
thought to be correlated to its amplitude. This could be
one reason why ET are considered to have a higher toler-
ance for shiftwork than MT.
Some studies have focused on the relationship between
CT and the cortisol awakening response. Bailey and Heit-
kemper, using both salivary cortisol sampling (Bailey and
Heitkemper, 1991) and plasma cortisol (Bailey & Heit-
kemper, 2001), revealed a higher amount of cortisol in
MT in the morning. In healthy, day-active adult men,
Kudielka et al. (2006, 2007) found an overall effect of
CT and salivary cortisol level after awakening. The acro-
phase of cortisol in serum was 55 min earlier in MT
(Bailey & Heitkemper, 2001), while Griefahn and
Robens (2008) reported higher cortisol levels in MT
after awakening but not during the rest of the day.
Randler and Schaal (2010) reported MT had higher corti-
sol levels immediately after awakening. Dockray and
Steptoe (2011) after different adjustments found no
effect of CT on cortisol in the evening, the cortisol awa-
kening response or total cortisol output over the
working day, whereas on the leisure day, total cortisol
output was greater in ET. In another study, ET showed
lower salivary cortisol levels and a flattened diurnal
curve (Oginska et al., 2010). As these results are some-
what inconclusive, further studies are needed addressing
weekend and weekday cortisol awakening responses,
both measured by spontaneous awakening and by
forced waking (alarm clock). Generally, it seems that cor-
tisol may enable people to prepare for the day and the
higher cortisol values in MT may explain why MT show
a better MA, e.g., feel less tired at awakening.
Melatonin is considered as a physiological cue for the
organism and its onset has been described as the best
predictor for sleep onset (Arendt, 2006; Rosenwasser,
2009). Moreover, the melatonin rhythm is argued to be
the best marker of the endogenous circadian pacemaker
(Benloucif et al., 2005). The onset, acrophase, and offset
of the melatonin profiles occurred approximately 3 h
earlier in MT than in ET, without differences in amplitude
(Gibertini et al., 1999; Griefahn et al., 2002; Mongrain
et al., 2004, 2005b). This was observed both in blood
and in salivary measurements. In a study with records
during extended wakefulness, an advance was also
observed in the time of salivary melatonin peak and
dim-light melatonin onset in the MT with respect to the
ET, although without differences in the area under the
curve (Taillard et al., 2011). There is an inverse relation-
ship between MEQ scores and the time of the melatonin
peak (Liu et al., 2000).
The maintenance of the melatonin circadian rhythm is
considered as a biological marker associated with suc-
cessful aging, while the flattening of the rhythm is
related to the appearance of neurodegenerative diseases
and the concomitant presence of psychiatric pathology
(Magri et al., 2005; Wu & Swaab, 2005).
A new physiological parameter has recently been
proposed for assigning the CT, based on daily secretion
patterns of melatonin and the growth hormone
(Nagane et al., 2011). The results suggest that the asyn-
chronicity and lack of peak secretion at midnight for
the two hormones reflect ET. This new parameter corre-
lates with psychosomatic complaints but requires further
validation studies.
Genetic Basis
We have already seen how circadian and sleep systems
interact to determine CT. Circadian rhythms are gener-
ated by a core set of circadian clock genes and proteins
that interact in a transcriptional/translational feedback
loop to determine the circadian period (Ko & Takahashi,
2006). Less is known about genes underlying sleep regu-
lation, but it is becoming increasingly evident that several
clock genes also have roles in sleep regulation and
homeostasis, which is consistent with the known inter-
action between the two systems (for a review, see
Franken & Dijk, 2009). As previously mentioned, some
characteristics of sleep are highly heritable, but twin
studies have also shown that CT has a very high overall
heritability of around 50% (Barclay et al., 2010; Kosken-
vuo et al., 2007). Therefore, because circadian and
sleep systems interact to determine CT, it should be
expected that variation in genes that have roles within
these systems will be associated with individual differ-
ences in CT phenotypes.
By far, the most successful approach to date in identi-
fying genetic polymorphisms that are linked with differ-
ences in CT has been via the candidate gene approach,
driven by systematic screens of clock gene polymorph-
isms. This approach has identified a catalog of poly-
morphisms in clock genes that show associations with
the CT phenotype, including CLOCK (Katzenberg et al.,
1998; Mishima et al., 2005), PER1 (Carpen et al., 2006),
PER2 (Carpen et al., 2005) and PER3 (Archer et al.,
2003, 2010; Johansson et al., 2003; Jones et al., 2007;
Lázár et al., 2012; Pereira et al., 2005). However, several
studies have failed to replicate some of these associations
in alternative populations: CLOCK (Barclay et al., 2011;
Chang et al., 2011; Johansson et al., 2003; Pedrazzoli
et al., 2007) and PER3 (Barclay et al., 2011; Osland
et al., 2011). The reasons for the failure to replicate
these findings are unclear, but may be related to differ-
ences in phenotyping methods, ethnicity differences,
age and sex differences, sample size, geographic location
and even time of year of study. A good example of how
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
age can be a confounding factor in these CT genotype/
phenotype associations, which can be relatively weak,
comes from a study that showed different levels of signifi-
cance among different age groups for a polymorphism
within PER3 (Jones et al., 2007). The association is stron-
gest for younger people and gradually reduces with age
until it becomes absent in middle-aged people between
40 and 49 yr old. Interestingly, this is also the age
where females no longer report greater morningness
scores than males of the same age in the MCTQ data
(Roenneberg et al., 2007).
In all of the above examples of associations between
CT and clock gene polymorphisms, it is not known
what the underlying biochemical or physiological mech-
anisms are that lead to the phenotypic differences.
However, some of the same polymorphisms have also
been linked with DSPD, in which sleep onset and offset
are delayed late into the night and day (CLOCK
Mishima et al., 2005; PER3 Archer et al., 2003, 2010;
Ebisawa et al., 2001; Pereira et al., 2005). For PER3,
these polymorphisms are hypothesized to affect the pro-
moter-driven gene expression levels of PER3 (Archer
et al., 2010), or are hypothesized to alter the phosphoryl-
ation state of the PER3 protein by removal of potential
Casein Kinase I (CKI) phosphorylation sites (Archer
et al., 2003; Ebisawa et al., 2001). The latter is important
because studies have shown that changes to the phos-
phorylation levels of PER protein can change the circa-
dian period (Lee et al., 2011). In addition, familial
mutations in PER2 (Toh et al., 2001) and CKI (Xu et al.,
2005) have been linked with advanced sleep phase dis-
order (ASPD) where it has been shown that greatly
advanced circadian rhythms and sleep timing are due
to changes in phosphorylation of PER protein.
Although genetic associations with CT are not always
reproducible, so far the association that has been repro-
duced most often is with the variable number tandem
repeat (VNTR) polymorphism in PER3. This primate-
specific (Jenkins et al., 2005), coding region polymorph-
ism in humans gives rise to four or five repeated
18-amino-acid motifs that contain multiple, putative
CKI phosphorylation sites (Archer et al., 2003). Because
CKI phosphorylation of PER was known to be an impor-
tant determinant of the circadian period and linked with
ASPD, it was hypothesized that the PER3 VNTR poly-
morphism would be associated with differences in CT
in humans (Archer et al., 2003). This was indeed the
case, such that MT were more likely to be homozygous
for the 5-repeat allele (PER3
) and ET and also
people with DSPD showing greater frequencies of
homozygotes (Archer et al., 2003; Jones et al.,
2007). The same association has been replicated in a
Brazilian study (Pereira et al., 2005) and more recently
in a separate UK-based study (Lázár et al., 2012).
Whereas previous studies had only genotyped subjects
with extreme CT phenotypes, the latter study investi-
gated the association in 675 subjects who were not
selected on the basis of their CT phenotype but
nevertheless found that increased morning preference
was associated with the PER3
genotype. As part of a
large battery of screening questionnaires, the study
employed both the MEQ and the MCTQ tools, in
addition to questionnaires and diaries to record sleep
timing. This approach enabled contrasts to be made
between workdays and rest days and revealed novel gen-
otype associations that were depende nt on this contrast.
subjects went to bed earlier, had an earlier mid-
point of sleep and woke up earlier than the other geno-
types. PER3
had a shorter time in bed during work-
days than PER3
but there was a significant interaction
between genotype and workdays versus rest days such
that PER3
actually spent the longest time in bed
during rest days. Interesting correlations were also ob-
served with the body mass index (BMI), such that if the
midpoint of sleep was later than 04:15 h during work-
days, PER3
individuals had a significantly higher
BMI than the other genotypes, and if time in bed was
greater than 9 h during workdays, they also had a lower
intelligence score. These observations underline the
importance of taking into account differences between
workdays and rest days, as previously emphasized for
other MCTQ data (Roenneberg et al., 2007).
While it was assumed that the PER3 VNTR would exert
its influence on CT via the circadian clock, this no longer
seems likely. Multiple human studies (Archer et al., 2008;
Hasan et al., 2012; Viola et al., 2007) have shown no evi-
dence for circadian differences between PER3 genotypes,
although a recent study in older people does report a
phase advance in melatonin in PER3
subjects com-
pared with PER
(Viola et al., 2011). The influence of
the PER3 VNTR on CT appears to be more closely
linked with mechanisms that regulate sleep timing and
homeostasis. PER3
people have earlier sleep/wake
times (Lázár et al., 2012) and show increased EEG theta
activity during wakefulness and greater slow-wave
activity power during NREM sleep, both markers of
greater homeostatic sleep pressure (for review, see Dijk
& Archer, 2009). PER3
people also show a greater cog-
nitive decline in response to sleep deprivation (Groeger
et al., 2008). A model can now be envisaged whereby
greater sleep pressure in the PER3
modulates diurnal
variation in performance and alertness via feedback
onto the circadian wake-promoting signal, leading to a
preference for morning activity in these people. This
model also fits well with what we know about the impor-
tant interaction between sleep and circadian systems and
emphasizes the point that CT is a complex phenotype
that is derived from multiple underlying genetic factors.
One intended purpose of CT scales is to identify individ-
uals that may better adjust to nightwork. However, the lit-
erature linking CT and adjustment is indicative of a
relationship but is far from convincing, while the
relationship with jet lag is even more problematic. In
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
this section, we review evidence for the validity of these
measures to predict adjustment in shiftwork and jet lag.
The major limitations include the absence of represen-
tative samples, few longitudinal designs, the absence of
control groups and a reliance on self-reported data. A
recent review of this literature identified 16 studies; all
these were cross-sectional and only two collected objec-
tive data (Saksvik et al., 2010). Furthermore, our review
of these studies revealed that six studies used a single-
item measure to assess CT, five used different versions
of the CTQ/I, four used the CSM, one study used the
MEQ and another used five items from the MEQ.
Seo et al. (2000) administered the MEQ to 561 shift-
workers across a number of heavy industries. MT found
the night shift more difficult. This group reported signifi-
cantly more sleepiness during the night shift, were more
likely to nap and the naps were of a longer duration. Kha-
leque (1999) studied 60 male shiftworkers and the results
suggested ET had less sleep interruption during day sleep
compared with MT. Small, large and epidemiological
studies each report an association between ET and night-
work (Adan & Almirall, 1991; Paine et al., 2006; Petru
et al., 2005).
Several studies report on the CSM and its relationship
with nightwork adjustment. McLaughlin et al. (2008)
reported morningness was a moderate predictor of toler-
ance to shiftwork. Older individuals were more likely to
be MT and reported greater difficulties concerning
sleep, fatigue and negative effect. Willis et al. (2008)
found ET had significantly greater work-family conflict
but no differences were found for burnout, depersonali-
zation and accomplishment. In a large-scale study of
nurses and industrial workers, morningness did not
predict sleep or social/domestic disturbance (Smith
et al., 1999). Di Milia and Muller (2012) showed that
MT shiftworkers were significantly more alert during
the morning hours but this pattern changed in the
evening/night hours when ET were more alert.
The CTQ and rCTI are also supported as predictors of
adjustment to shiftwork. Smith et al. (1999) reported that
irrespective of the actual shift schedule parameters and
type of job, workers with inflexible sleeping habits had
greater sleep disturbances and in turn resulted in
increased fatigue and emotional disturbances. Furnham
and Hughes (1999) found journalists with flexible sleep-
ing habits were better able to overcome drowsiness and
had greater well-being. In addition, they also reported
night journalists to be ET. Vela-Bueno et al. (2010) inves-
tigated adjustment in 265 aircraft technicians. Their
results indicated the non-adapted workers were
morning oriented, had rigid sleeping habits and were
languid. In a rare study, Kaliterna et al. (1995) collected
individual differences data from workers before com-
mencing shiftwork and at a 3-yr follow up. Workers
with rigid sleep habits and languid behavior were more
likely to report poor health and psychosomatic and diges-
tive complaints. It is important to note these correlations
were small but nevertheless significant.
The rCTI (Di Milia et al., 2005) is being used in a series
of longitudinal studies and results from the first wave of
data collection are now available. Natvik et al. (2011)
reported on the survey responses from 1500 nurses
working either two or three shifts. After controlling for
demographic factors, rigidity was associated with insom-
nia among the three shiftworkers and languidity was
associated with greater sleepiness, depression and
anxiety. In a group of experienced nurses, Saksvik
(2012) found flexibility but not languidity was positively
linked with shiftwork tolerance.
The DTS (Torsvall & Åkerstedt, 1980) reported strong
correlations between morningness and a number of
sleep complaints. MT reported less sleep problems
during the day shift while ET reported more problems.
This pattern was reversed during nightwork. Natvik
et al. (2011) found morningness was negatively related
with insomnia. Morningness by shift type was significant
overall for depressive symptoms and morningness was
associated with lower depressive symptoms in three shift-
workers. Saksvik (2012) reported morningness predicted
adjustment to nightwork among inexperienced nurses.
This finding is unexpected, but this group was younger
and had worked less than 20 night shifts across the year.
We found one published study (Gamble et al., 2011)
using a modified version of the MCTQ in a nursing
sample. Night shift nurses had a significantly later chron-
otype than day nurses. The interaction between shift type
and CT suggested the earlier types had better adaptation
to day work but worst adaptation on night shift. By con-
trast, the late CT had intermediate adaptation to both
shifts. We located a doctoral study (Juda, 2010) that dis-
cusses the development of the MCTQ
that applies
a series of corrections according to the shift being
worked. The results are interesting and require replica-
tion in larger samples.
Jet lag refers to the misalignment between the internal
clock and external time cues (Arendt, 2009). Thus, in
many ways, jet lag sufferers and shiftworkers report
similar impaired characteristics with the exception that
shiftworkers internal clocks are di srupted on a more fre-
quent basis. There is a good deal of literature advising tra-
velers on how to minimize jet lag (Eastman & Burgess,
2009). There is speculation that better adjustment may
be found in people with flexible sleeping habits and
that MT would find it easier to cope with eastward
travel, while ET cope better with westward flights (Water-
house et al., 2007). However, the study of CT and jet lag is
notable for the absence of studies.
Waterhouse and his colleagues (2002) studied 85 ath-
letes, coaches and academics on a flight from London to
Sydney. Participants completed the CSM, CTI and several
other measures. The findings suggested flexible types
were able to go to sleep earlier and stay asleep for the
first night only and no effect was found for rigidity of
sleeping habits. Morningness failed to predict tolerance
to jet lag, but this may be explained because 78 partici-
pants were intermediate types.
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
Flower et al. (2003) surveyed 100 participants that had
flown across five time zones at least once in the previous 6
mo. After adjusting for gender, the evidence was weak but
suggested flexible sleeping habits were associated with
less travel fatigue but that morningness was not a signifi-
cant predictor. Finally, in a review of circadian rhythm
sleep disorders, Sack et al. (2007) conclude d: no study
has examined the utility of the MEQ in assessing risk
for the development of jet lag disorder. Future studies
need to employ better designs before the link between
CT and jet lag can be outlined.
In spite of the work carrie d out in the past decades (Adan
et al., 2008; Cavallera et al., 2011; Guerrien et al., 1993), it
is still difficult to synthesize the relationship between CT
and cognitive performance because of the great number
of variables considered, and the heterogeneity of the
research methods adopted.
In this section, we aim to attract the readers attention
towards the possible explicative role of CT in the compre-
hension of cognitive functioning in humans. First, we
focus on methodological issues, and then we comment
on the role of CT in relation to time-of-day effect and to
the individual differences framework.
Methodological Issues
To remove the masking effect and dissect the contri-
bution of circadian and homeostatic influences on
performance rhythms, specific study protocols have
been developed: constant routine (CR) protocol; CR
with multiple naps; FD protocol (Blatter & Cajochen,
2007). However, these experimental protocols set
human participants in artificial conditions questioning
the applicability of the results to everyday life (Vanin
et al., 2012).
We suggest that circadian fluctuation data of cognitive
performance may be recorded and evaluated in the
normal daynight condition with greater accuracy if CT
is taken into account. The biological and behavioral par-
ameters studied in connection to the CT are now numer-
ous. As for body temperature (the gold standard
parameter for the circadian clock), it is known that both
extreme types reach the acrophase in the second half of
the day, even if the MT reach it around 2 h earlier than
ET (Kerkhof & van Dongen, 1996). On the contrary, for
both subjective and objective alertness, daytime trends
seem almost specular (Natale & Cicogna, 1996). The
MT reach the acrophase in the first half of the day,
whereas the ET reach it in the late afternoon. Therefore,
CT could be considered as a paradigm offered by the
nature in which the circadian pacemaker and homeo-
static processes are differently mixing and/or interacting.
Bearing in mind that a CT-based protocol cannot permit
to segregate the respective contributions of circadian and
homeostatic processes, it nevertheless has the advan-
tages of being very ecologic and gives realistic indications
about human behavior in normal daynight conditions
(Valdez et al., 2008).
Monk and Leng (1986) found a greater phase differ-
ence in a logic reasoning task than in a visual search
task in the extreme CT. On the basis of this evidence,
they posited that the nature of the task, and hence the
resource required to carry it out, might be critical in dif-
ferentiating the performance diurnal trend in the CT. In
particular, they suggested that tasks involving large
amounts of cognitive resources would better differentiate
the two extreme CTs, since these tasks are not simply
mediated by changes in arousal levels but also by
changes in the performance strategy. Interpreting the
results from complex task performance introduces a
new question: which is the cognitive function giving the
prevalent contribution to solve the task? Moreover,
time-of-day effects depend not only on the task complex-
ity but also on the way in which it is performed. Indeed,
researchers suggest that many time-of-day effects do not
reflect automatic changes arising from processing limit-
ation, but are due to the adoptio n of different strategies
over the course of the day (Folkard, 1990). It is not
clear whether the strategy changes observed over the
day reflect endogenous rhythms or are attempts to main-
tain competent performance in a sub-optimal state. The
mobilization of effort in a task is also modulated by the
demands of the task itself. The roles of endogenous
rhythms, exogenous factors and motivation have been
separately studied, but we have to keep in mind that it
is the combination of these factors and the individuals
interpretation of them which determines the nature of
the observed time-of-day effect on performance in
normal daynight conditions as well as in everyday life
(Adan, 1993; Clarisse et al., 2010; Smith, 1992).
Time of Day and Cognitive Performance
The interest in time-of-day effects on cognitive efficiency
has a long history (Laird, 1925). Kleitman (1963) showed
strong evidence for a parallelism between body tempera-
ture (the gold standard mark of the human circadian
system) and time-of-day effects for simple repetitive
tasks: the decrease in reaction time response was signifi-
cantly correlated to an increase in body temperature.
These data become known as the arousal model (Colqu-
houn, 1971). In this model, circadian performance vari-
ations are postulated to reflect an underlying circadian
rhythm in the basal arousal level. Because body tempera-
ture increases during the day, also the performance effi-
ciency should always increase during the day. Both
extreme CTs reach a higher arousal level in the second
half of the day, and thus for both extreme CTs, the
arousal model foresees better cognitive performances
in the second half of the day. However, in a study in
which a simple weak motor component task was sched-
uled every 2 h from 8:00 to 10:00 h, it was found that
ET improved their performance during the day,
whereas MT presented an opposite trend, with a phase
advance up to 12 h (Horne et al., 1980). As far as
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
memory tasks are concerned, some studies have shown a
phase advance of the best performance time ranging
between 2 and 6 h for MT (Adan, 1991; Anderson et al.,
1991; Natale & Lorenzetti, 1997; Petros et al., 1990).
Homeostatic processes are also involved in the modu-
lation of cognitive efficiency. During a normal day, the
most important effect is the so-called post-lunch impair-
ment, which has been found both in real life jobs and in
the laboratory. Lavie and Segal (1989), using the ultra-
short sleepwake paradigm, reported a much clearer
post-lunch dip in MT than for ET. The post-lunch dip
could be modulated by blood concentrations of naturally
occurring benzodiazepine, which reach relative higher
values at noon in MT in comparison to ET (Sand et al.,
It is unlikely that cognitive efficiency is solely mediated
by circadian (body temperature) or homeostatic (sleep
wake cycle) processes. From a behavioral point of view,
alertness derives from the interaction of these two pro-
cesses (Natale & Cicogna, 1996). Therefore, to explain
so large a phase difference in cognitive efficiency
between MT and ET, the role of alertness was then con-
sidered, which is a behavioral and not physiological vari-
able. It was concluded that cognitive performances are
more efficient when testing time is in synchrony with
individuals peak in alertness. This model is known as
the synchrony effect (May & Hasher, 1998), and states
that individuals who are more alert in the morning
tend to perform better in the morning than in the after-
noon, and individuals who are more alert in the
evening tend to perform better in the afternoon or
evening than in the morning. Data supporting the syn-
chrony effect were collected adopting a range of cognitive
tasks, including negative priming, false memory, recog-
nition and recall of prose and span materials, categoriz-
ation, judgment and control over distraction, and
working memory (Hasher et al., 2002; Hornik &
Miniero, 2009; Intons-Peterson et al., 1998, 1999; May,
1999; Rowe et al., 2009; Yang et al., 2007).
However, the synchrony effect did not globally affect
performance. In fact, synchrony effect was not documen-
ted in a constant way for all studies (Ciarkowska, 1997;
Fabbri et al., 2012; Gillooly et al., 1990; Natale et al.,
2003). In a recent study examining the effect of time of
day on problem solving, the results seem to indicate
that tasks involving creativity might benefit from a non-
optimal time of day (Wieth & Zacks, 2011).
On the basis of results involving CT as an independent
variable, it is possible to conclude that it is over simplistic
to link cognitive efficiency only to arousal or alertness.
Cognitive efficiency is not solely determined by under-
lying regulatory activation systems, but it is modulated
by compensatory mechanisms, such as motivational
factors or expectancy due to experience. It is possible to
conclude that in normal daynight conditions, tasks
requiring a high operational load could involve a cogni-
tive and motivational engagement that can offset the
decrease in efficiency induced by alertness changes.
This could explain why tasks involving a wider range of
cognitive resources present no relevant differences
between the two extreme typologies.
These last observations allow us to introduce the
importance for distinguishing among arousal and arou-
sability, which means that the arousal system can be dif-
ferentiated into tonic (slow adapting arousal state) and
phasic (which rapidly adapts to transient changes) acti-
vation. Current models of attention in cognitive science
postulate three primary functions for attentional mech-
anisms (i.e., alerting, orienting and executive control),
each of them presumably corresponding to underlying
distinct and independent neural networks (Fan et al.,
2002). The purpose of the alerting network is to increase
and sustain arousal and vigilance in order to better
prepare the organism for the det ection of forthcoming
stimuli (phasic attention). The orienting network special-
izes in selecting specific information from an array of
potentially relevant stimuli. The executive component
of attention mediates planning, decision making, error
detection, conflict resolution and inhibitory control.
The efficiency of the alerting, orienting and executive
components of attention is frequently measured with
the Attention Network Test (ANT) (Fan et al., 2002).
Only one study used the ANT paradigm in relation to
CT (Matchoock & Mordkoff, 2009). The orienting com-
ponent showed no time-of-day effect or CT effects,
whereas executive control was lower in the middle of
the day regardless of CT. By contrast, alerting scores sig-
nificantly increased in the second half of the day for MT
and NT, while results were stable over the day for ET.
Results relative to MT are not surprising because a
larger alerting effect is associated with a difficulty to
maintain alertness without a cue, while results relative
to ET are somewhat surprising (Fan & Posner, 2004).
On the whole, these results introduce a last possible
question: is CT associated with di fferences in cognitive
styles regardless of time of day?
Cognitive Styles
It is commonly believed that going to bed early and
getting up early would result in better performance
across the day. This belief is not supported by empirical
data (Gale & Martyn, 1998). The few studies that have
examined this question obtained small but significant
correlations in the opposite direction: ET showed higher
intelligence scores (Kanazawa & Perina, 2009 ; Roberts &
Kyllonen, 1999; Song & Stough, 2000). Killgore and Kill-
gore (2007) administered the MEQ and Wechsler Abbre-
viated Scale of Intelligence to 54 healthy volunteers and
found a small but significant negative correlation demon-
strating the better performance of ET on verbal ability.
In a study of adolescents, MT were reported to perform
better on final school examination (Randler & Frech,
2006). However, the authors correctly noted that school
start time is early, while adolescents biologically tend to
shift towards eveningness at this time (Tonetti et al.,
2008). Therefore, ET might be at a disadvantage
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
because they have gone to bed too late and so do not have
enough sleep.
A more interesting approach is to consider the cogni-
tive style. Cognitive style is not the ability but a preferred
way of information processing. This approach seeks to
understand the preferred strategies one adopts rather
than performance per se. Three kinds of styles have
been found: cognitive styles, learning styles and thinking
styles (Sterberg & Zhang, 2001). Cognitive styles indicate
attitudes and tendencies to activate distinctive mental
operations in a variety of situations. Thinking styles
concern which set of reasoning strategies an individual
is inclined to apply. Learning styles characterize how a
person faces study tasks. From the perspective of
cognitive styles, ET has been associated with field
independence (Sarmàny, 1984) and creative thinking
(Giampietro & Cavallera, 2007).
Learning and thinking styles have been related to an
individuals hemisphere preference (Springer &
Deutsch, 1997). Folkard (1990) suggested that diurnal
changes in performance may also reflect a change
across the day concerning the dominance of the left
hemisphere. Corbera and Grau (1993) studied 48 right-
handed women who performed verbal and spatial hemi-
field tachistoscopic tasks at four different times a day.
Changes in accuracy over the day showed a left-hemi-
sphere advantage at 12:00 h whereas right-hemisphere
use had an advantage at 19:45 h. Several studies have
also demonstrated a change over the day in hemispheric
dominance (Folkard, 1990; Iskra-Golec & Smith, 2006;
Natale et al., 2003; Shub et al. 1997) is a significant
factor when we try to understand the heterogeneity
between tasks in the literature regarding the time-of-
day effect. Using the Style Of Learning And Thinking
questionnaire, Fabbri et al. (2007) showed that MT is
associated with the left-thinking style and ET is associ-
ated with the right-thinking style. The left thinker
follows an analytic and sequential mode of reasoning
and he/she relies preferably on verbal-abstract represen-
tation. On the contrary, the right thinker tends to process
information in an intuitive, holistic, gestalt-type, syn-
thesized and visual-motor way (Torrance et al., 1988).
The concept of hemisphericity is obviously a sche-
matic way to represent the reality, but this concept
could allow us to give a summarized description of indi-
vidual differences. We hope that future research will
deeply analyze the relationship between CT and cogni-
tive style, suggesting that ET and MT tend to follow differ-
ent cognitive processes.
There ha v e been some previous attempts to summarize the
relationship between CT and personality (Cavallera &
Giudici, 2008; Tank ova et al., 1994). The early litera tur e
focused mainly on Ey sencks three-fa ctor model while
more recent studies hav e used differe nt models. The
purpose of this section is to first revie w the Eysenck
Personality Inv entory (EPI) literature and then assess
the rela t ionship between CT and more recent
personality inventories.
Eysenck Personality Inventory
Studies using the EPI have relied mainly on the extraver-
sion and neuroticism dimensions and less on psychotics.
Some studies have also examined the subcomponents of
extraversion, impulsivity and sociability (Eysenck &
Eysenck, 1975). Neuroticism is generally related to
emotional stability while psychoticism provides a link
to detect conduct disorders (Eysenck et al., 1985).
ET typically report higher scores in extraversion than
MT (Langford & Glendon 2002; Mitchell & Redman
1993; Tankova et al., 1994;), while Matthews (1988)
reported this relationship only in women, and some
studies yielded no relationship between CT and extraver-
sion (Mecacci & Rocchetti, 1998; Tankova et al., 1994).
However, no studies have found an inverse relationship
between ET and extraversion. These results suggest that
the relationship between ET and extraversion is the most
stable relationship obtained with the EPI. However, it is
still unclear which component of the extraversion dimen-
sion is responsible for this relationship. As Tankova et al.
(1994) point out in their review, some studies found ET
had higher scores for sociability and others for impulsivity,
while some detected no relationship.
The results for neuroticism show contradictory results.
Some have reported higher scores in ET (Mecacci & Roc-
chetti, 1998; Tankova et al., 1994) while two studies found
MT had higher scores (Langford & Glendon, 2002;
Tankova et al., 1994). Several studies reported no relation-
ship between CT and neuroticism (Mitchell & Redman
1993; Tankova et al., 1994). Studies based on the psychoti-
cism dimension of the EPI suggest ET score higher on this
dimension than MT (Mecacci & Rocchetti 1998; Mitchell &
Redman 1993; Tankova et al., 1994), and this may be
linked to psychopathology. Differences in results may be
related to different samples or study designs.
Big Five Model
The Big Five Model (or Five-Factor Model) is made up of
the following factors: extraversion, agreeableness, con-
scientiousness, neuroticism and openness (Costa &
McCrae, 1992). Extraversion is related to a high degree
of sociability, assertiveness and talkativeness. Agreeable-
ness is characterized by being helpful or cooperative
toward others. Conscientiousness refers to being disci-
plined, organized and achievement-oriented. Neuroti-
cism refers to emotional stability and anxiety. Openness
means a strong intellectual curiosity and a preference
for novelty.
ET were reported to be marginally but not significantly
more extroverted than MT (Jackson & Gerard, 1996). No
other study found a relationship between extraversion
and CT. There was a positive relationship between agree-
ableness and morningness (Hogben et al., 2007; Randler,
2008c; deYoung et al., 2007), but no relationship between
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
agreeableness and CT could be detected (Gray & Watson,
2002; Jackson & Gerard, 1996; Tonetti et al., 2009). In all
studies, conscientiousness showed a positive relationship
to morningness and is considered the best predictor of
morningness. In some studies, neuroticism was higher
in ET individuals (Randler, 2008c; Tonetti et al., 2009;
deYoung et al., 2007) but no relationship was found in
others (Gray & Watson, 2002; Hogben et al., 2007;
Jackson & Gerard, 1996). ET had a higher openness
(Hogben et al., 2007) while no relationship could be
detected in five other studies (Gray & Watson, 2002;
Jackson & Gerard, 1996; Randler, 2008c; Tonetti et al.,
2009; deYoung et al., 2007).
Temperament and Character Inventory
The Temperament and Character Inventory (TCI) has a
theoretical psychobiological basis (Cloninger, 1994).
These include four dimensions of temperament
(novelty seeking, harm avoidance, reward dependence
and persistence) and three dimensions of character
(self-directedness, cooperativeness and self-transcen-
dence). Temperament dimensions are supposed to be
inherited and related with specific neurotransmitters,
while character dimensions are supposed to develop
during life (Clo ninger, 1994). ET showed higher scores
in novelty seeking (Adan et al., 2010a; Caci et al., 2004;
Randler & Saliger, 2011) and lower in harm avoidance
(Adan et al., 2010a). There was no relationship in
reward dependence. Concerning persistence, higher
scores have been found in MT (Adan et al., 2010a; Caci
et al., 2004; Randler & Saliger, 2011). Concerning charac-
ter, ET had lower scores in self-directedness (Adan et al.,
2010a) and cooperation (Randler & Saliger, 2011), while
ET were higher in self-transcendence (Randler &
Saliger, 2011).
Alternative Five-Factor Model
The Alternative Five-Factor Model (AFFM) (Zuckerman,
2002) was used by Muro et al. (2009, 2011), and these
authors showed that MT subjects had significantly
higher scores than ET and NT subjects in activity, and
in its subscales general activity and work activity. Muro
et al. (2009) found an interaction: MT men showed
higher scores than ET and NT in neuroticismanxiety
and women had the opposite pattern. In women, Muro
et al. (2011) reported that MT women had higher
scores than ET and NT women on activity, and its sub-
scales general activity and work activity, while ET
women scored higher than MT women on aggression
hostility, impulsive sensation seeking and its subscale
sensation seeking. The higher level in activity in MT
might be related to higher cortisol levels, at least at awa-
kening (Kudielka et al., 2006, 2007; Randler & Schaal,
Other Personality Measurements
Díaz-Morales (2007) used the Millon Index of Personality
Styles. ET preferred imaginative, feeling-guided,
innovation-seeking thinking styles, while MT preferred
realistic, thought-guided and conservation-seeking
thinking styles. ET preferred unconventional behaving
styles while MT preferred dutiful behaving styles. By
using Zuckermans Sensation-Seeking Scale, Tonetti
et al. (2010) reported higher scores in ET compared
with MT in SSS-V total score and all subscales, except
boredom susceptibility. When comparing different
facets of impulsi vity, Adan et al. (2010b) reported that
MT obtained lower scores in dysfunctional impulsivity
than the NT and ET. By using the Portrait Values Ques-
tionnaire (Schwartz, 2003), Vollmer and Randler (2012)
found that morningness was correlated with higher
acceptance of social values (conservation and self-trans-
cendence), while eveningness was correlated with higher
preference for individual values (openness to change and
self-enhancement). Based on the Maudesley Personality
Inventory, Hsu et al. (2012) reported that ET of both sexes
scored higher than the MT in novelty seeking, harm
avoidance and neurotic personality characteristics, but
lower than MT in extraversion and social desirability.
Concerning reward dependence, ET scored lowest for
males, but there was no difference for females.
Morningness also was positively related to proactivity
(Randler, 2009b) and subjective well-being or life
quality (Biss & Hasher, 2012; Jankowski, 2012; Lázár,
2012; Randler, 2008b). ET showed a less adaptive
emotional profile than MT and NT, who showed a rela-
tively similar emotional pattern. Focus and order
(facets of Control), energy (facet of Volition), caution
(facet of Inhibition) and problem facing (facet of
Coping) were particularly low in ET and high in MT
(Ottoni et al. 2012). Finally, Díaz-Morales et al. (2008)
found that morningness is negatively related to indeci-
sion and procrastination, two behaviors that are typical
in individuals who score low in conscientiousness.
Personality Traits Relationships
As a summary, research on personality and CT indicates
that ET subjects are more extraverted, impulsive, novelty-
seeking and open-minded and tend more to psychoti-
cism (psychopathology), while MT ones tend to be
more introverted, conscientious, agreeable, persistent
and emotionally stable.
One interesting fact is that eveningness was relatively
strong related to extraversion when using the EPI, while
nearly no study focusing on the BFI found a significant
relationship between these two variables. However,
extraversion is one dimension that is quite similar
across EPI, BFI and AFFM (Aziz & Jackson 2001;
Zuckerman et al., 1993).
Nevertheless, some authors suggest the relationship
between CT and personality might have to do with the
different theoretical models used to assess personality
(Randler, 2008c; Tsaousis, 2010), rather than with the
different measures used to assess morningness, as none
of them seems to be psychometrically superior (Di
Milia et al., 2008). Indeed, it might depend on the
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
personality inventories used: the BFI is a psycholexical
approach (based on adjectives from a lexicon), while
EPQ, TCI and AFFM are based on biological theories/
models. Furthermore, it may also be dependent on the
aspect or subcomponent of extraversion measured e.
g., activity or impulsivity. Tankova et al. (1994) suggested
that it is probably due to the impulsive component
inherent to the extraversion dimension (Adan et al.,
2010b; Caci et al., 2005b), and especially dysfunctional
impulsivity (Adan et al. 2010b), which could be respon-
sible for the relationship between extraversion and even-
ingness. Indeed, Eysencks extraversion is positively
correlated with novelty seeking of the TCI and negatively
with harm avoidance (Zuckerman & Cloninger, 1996).
However, Muro et al. (2009) further suggested that pre-
vious correlations between morningness and Eysencks
extraversion were probably due to the activity component
inherent to extraversion, rather than impulsivity or socia-
bility components (Caci et al., 2005b; Larsen, 1985;
Tankova et al., 1994). The ZKPQ-activity component, in
turn, is correlated with the persistence component of
the TCI, which supports Muro et al.s (2009) suggestion
(Zuckerman & Cloninger, 1996).
Similarly, in most studies, ET scored higher in the
impulsive sensation-seeking dimension of the ZKPQ,
which is considered equivalent to the Big Fives con-
scientiousness factor reversed and very close to Clonin-
gers novelty seeking (Adan et al. 2010a; Tonetti et al.,
2010; Zuckerman & Cloninge r, 1996). Thus, it seems a
basic relationship between eveningness and novelty/
sensation seeking.
The morningness and eveningness dimension shares
some behavioral components with mental disorders.
Among the symptoms that are often associated with
the presence of psychiatric symptoms and CT are altera-
tions in appetite, in the sleep/wake cycle, in cognition and
in activity, which together can lead to social damage. The
majority of these studies have shown that ET is associ-
ated with a number of psychiatric symptoms.
CT have also been studied at the genetic level to deter-
mine whether clock genes are involved in the link
between CT and mental disorders, but controversial
results have been observed (Lee et al., 2010; Osland
et al., 2011; Serretti et al., 2010). The results from
genetic studies suggest the polymorphism may not be
the only factor responsible for the association between
mental disorders and CT. Instead, it is likely that this
relationship is generated by behavioral habits that
expose individuals to different Zeitgebers, which, in
turn, modify the internal clock and potentially predis-
pose individuals to mental disorders. As a test of this
hypothesis, one study reported that ET presented
higher levels of chronic work-related fatigue, exhibited
less regular social rhythms, and were exposed to lower
levels of light during their waking hours compared with
NT and MT (Martin et al., 2012). Additionally, the
relationship between depressive mood and eveningness
in rural populations was mediated by differences
between week and weekend sleep schedules (social jet
lag) (Levandovski et al., 2011). Social rhythms are
known to interfere with the synchronization of biological
rhythms and consequently cognitive and behavior altera-
tion. Thus, the study of social rhythms may lead to a
reflection on the temporal organization of society and
may support the development of a new approach to
treat mental disorders.
Mood Disorders
Mood disorders, especially those related to depression,
are among the most prevalent mental impairments,
approximately 1015% (Lépine & Briley, 2011). The
World Health Organization estimates that unipolar
major depression will be the leading disease burden by
the year 2030. Both clinical and sub-clinical depressions
are related to a higher prevalence of disabilities. There
are different subtypes of depression, but certain
parameters are present in almost all of them, such as dis-
turbed sleep and changes in the circadian rhythms of
cortisol, ACTH, melatonin and other endocrine and
metabolic parameters (Soria & Urretavizcaya, 2009).
Figure 1 represents the main theory related to the
etiology of mood disorders that involves amino
neurotransmitters, such as noradrenaline, serotonin
and dopamine. These neurotransmitters are intrinsically
connected to melatonin production.
A clear example of the link between chronobiology
and depressive manifestation can be observed in seaso-
nal affective disorder (SAD). The circadian rhythm dis-
order may be caused by desynchronization between the
light/dark cycle and the human biological clock during
seasons with shorter photoperiods (Benedetti et al.,
2007; Lewy et al., 2006; Rosenthal et al., 1984). Similarly,
in bipolar disease, the oscillations between mania and
depression may occur in regular cycles. Furthermore,
inter-individual differences in sleep timing, duration
and morning or evening preferences are associated
with changes in circadian processes, sleep homeostatic
processes or both (Schmidt et al., 2012).
Several follow-up studies showed that circadian timing
is drastically delayed in patients with bipolar disorder
(Giglio et al., 2010; Soreca et al., 2009), SAD (Elmore
et al., 1993) and major depressive disorder (Drennan
et al., 1991; Soria & Urretavizcaya, 2009). Circadian mis-
alignment and sleep disruptions in patients with mood
disorders have been linked to abnormal daily patterns
of gene expression, hormonal secretion, body tempera-
ture, and cognitive and behavioral functions (Wulff
et al., 2010). An additional indication that circadian mis-
alignment contributes to depressive symptoms is the
finding that ET people are prone to depressive symptoms
(Chelminski et al., 1999; Giannotti et al., 2002; Hidalgo
et al., 2009, Kitamura et al., 2010; Selvi et al., 2011).
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
FIGURE 1. Molecular, physiological and clinical levels involved in circadian typology. At the molecular level, a set of core circadian clock
genes is involved with determining circadian typology (A). At the physiological level, light reception entrains the central circadian clock in
the suprachiasmatic nucleus, which drives circadian outputs such as melatonin synthesis (B). The neurotransmitters involved in the etiol-
ogy of mental disorders participate directly in the synthesis of melatonin (C). Factors such as stress trigger the induction of tryptophan
pyrrolase, capable of destroying tryptophan before it reaches the pineal gland (D). Tryptophan diminishes the production of leptin,
which is involved in food intake and weight gain, and increases risk of cardiovascular diseases. Leptin also increases the level of dopamine
(E). Molecular and physiological pathways interact to define a behavioral circadian typology phenotype that can be measured by self-assess-
ment scales and/or several biological outputs (F).
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
The internal desynchronization may be a major factor
with regard to mood. Studies in healthy individuals using
different measurement instruments have shown that the
depressive symptoms that present higher discriminate
coefficients among CT are those related to sadness,
inner tension, sleep reduction and pessimism (Hidalgo
et al., 2009). Thus, depressive symptoms present a poten-
tial confounding effect in CT studies. In a sample of
depressive patients, the sleep variables of CT and sleep
quality did not significantly predict suicidal ideation if
the effect of depressive symptoms was controlled (Selvi
et al., 2010).
ET has also been associated with higher levels of
depre ssion that may be modera ted by age (Kim et al.,
2010) and mediated by sexual hormones (for further infor-
mat ion, see biological differences section). High testos ter -
one levels lead to evening ness (Randler et al., 2012), and
menstrual symptoms are more significant in ET girls
(Negriff & Dorn, 2009). This evidence indicates that a
phase delay ma y be a part or consequence of the diagnosis
of mental disorders (such as depression, anxiety and ea ting
disorders) or may be an associa ted risk factor link ed by
sexua l development (Schneider et al., 2011; Schubert &
Randler, 2008). Therefore, the next step to clarify the caus-
ality relation is to dev elop transcultur al studies with a hom-
ogenous methodology in longitudinal stu dies.
The evidence that chronodisruption can be linked to
depression has opened a new approach for its treatment.
Photic treatment is based on the assumption that
exposure to light inhibits the production of melatonin,
thereby reducing the symptoms of depression (Lewy
et al., 1987; Wirz-Justice et al., 2005). Studies have also
found that physical exercise influences rhythms and
improves mood after weeks of early morning training
(Peiser, 2009). But, it will also be necessary to develop
randomized clinical trials to test the efficacy and the
effectiveness of chronobiological interventions, such as
melatonin, light exposure and social rhythm regulation,
as well as to test the best time of day for eating, sleeping
and exercising.
Eating Disorders
Leptin and ghrelin are hormones that act directly on the
timing of food ingestion and regulate appetite. Leptin is
a hormone produced in adipose tissue (Figure 1) that
acts directly under a timing system regulating food inges-
tion. It has been hypothesized that leptin is involved in the
etiology of depressive symptoms, which are very common
among obese patients. Obviously, there is a neuronal cor-
relation among mental disorders, despite the academic
effort to classify each syndrome into mental categories.
For example, mood and eating disorders appear to be con-
nected via hormones that regulate not only feeding but
also neuronal activity in the mesolimbic dopaminergic
pathway. In particular, leptin reduces the firing rates of
dopaminergic neurons, which indicates that leptin exerts
a direct influence on dopamine neurons via leptin recep-
tors (Dibner et al., 2010) (Figure 1).
Therefore, chronodisruption in appetite hormones
may provoke a disregulation in the aminergic pathway
and vice versa. A study that investigated the relationship
between leptin levels and ghrelin as well as symptoms of
depression, anxiet y and stress in women demonstrated a
significantly negative relationship between leptin and the
severity of depressive symptoms and anxiety (Lawson
et al., 2012). Females present a greater expression of
clock genes in adipose tissue, and consequently, it has
been hypothesized that this sexual dimorphism may
account for the different CT of males and female
(Gómez-Abellán et al., 2012).
Some syndromes are important human physiological
situations in which we can observe evidence of chrono-
biology manifestations through altered behavior. Nor-
mally, energy intake varies in intensity throughout the
day. In the natural environment, there is an increase in
food intake and energy consumption at night compared
with morning (de Castro, 2004). One example of abnormal
eating behavior due to temporal alteration is the night
eating syndrome (NES). The etiology of NES might be
explained by the desynchronization of temporal order
and internal difficulties in synchronizing with environ-
mental cycles, such as delayed eating times. Reinforcing
this hypothesis, one study has shown a positive correlation
between eveningness and NES (OReardon et al., 2004).
Recent chronobiological studies show that drug con-
sumption, even at moderate doses, has a negative effect
on circadian rhythmic expression. The amplitude of
some circadian rhythms is lower or even disappears in
extreme cases, which suggests a lesser quality of the
wake and sleep periods. A recent study by Reinberg
et al. (2010) found a glass of red wine during dinner
(20.5 g alcohol/24 h) is sufficient to desynchronize circa-
dian time organization. The effects of drug addiction on
the circadian rhythmic expression can persist for weeks
or even months after drug use has ceased (Adan, 2012).
The lack of regularity in circadian rhythms has been
mostly studied with alcohol (Conroy et al., 2012; Danel &
Touitou, 2004; Huang et al., 2010; Rupp et al., 2007),
tobacco (Adan et al., 2004; Adan & Sánchez-Turet, 1995,
2000) and cocaine dependence (McClung et al., 2005; Uz
et al., 2005). However, it seems that the data compiled to
date may be generalized to most types of drugs, indepen-
dently of their pharmacological effects on the organism
(Adan, 2012; Perreau-Lenz & Spanagel, 2008).
Adans (1994) pioneering study found significant
differences in the consumption of legal psychoactive
substances (nicotine, alcohol and caffeine). ET subjects
reported higher consumption sporadic and habitual
of the three substances. Subsequent international
studies have confirmed this observation, and this is of
special relevance for adolescent and adult populations
(Andershed, 2005; Gau et al., 2007; Giannotti et al.,
2002; Urbán et al., 2011; Wittmann et al., 2006). Prat
and Adan (2011) have also found ET subjects consumed
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
more illegal drugs (cannabis, ecstasy and cocaine) and
engaged in hazardous alcohol use than MT. This has
become a social and health problem with potentially
serious consequences in a short term (accidents,
unwanted pregnancies, assault) beyond the risk of devel-
oping an addictive disorder. These results alert us to the
need for further research on CT and heavy episodic
drinking, although in many cases drug poly-consump-
tion is present.
Although there are many other elements that also
participate in the onset and maintenance of drug con-
sumption, CT is certainly one of them. It has been hypoth-
esized that ET subjects begin drug consumption with
stimulants in order to adjust their degree of daytime acti-
vation to the socio-environmental demands when their
alertness is low, whereas the use of depressors tends to
appear toward the end of the day, when their degree of
activation exceeds that required for the demands. In the
initial phases of consumption, individuals who are prone
to chronobiological vulnerability (ET) may obtain superior
effects of regulation in their activation and reinforcement,
and this will favor behavior maintenance (Adan et al.,
2008; Adan & Sánchez-Turet, 2000; Prat & Adan, 2011).
However, the spiral of stress will appear with chronic con-
sumption, and it may be more intense and of more dra-
matic consequences in the ET individuals (Adan, 2012).
There is also evidence that circadian genes play a role in
addiction. The circadian gene Clock is a direct regulator
of the dopaminergic activity in the brain areas of reward
(Huang et al., 2010; Rosenwasser, 2010), Per1 has been
related to rewarding properties (McClung, 2007) and
Per2 has been proven essential to inhibit the sensitization
and reward to the effects of drugs (Perreau-Lenz & Spana-
gel, 2008; Perreau-Lenz et al., 2009).
It is important to remember that mood disorder,
obesity, metabolic syndrome and addiction are risk
factors for diabetes, hypertension and vascular disease,
among others. Thus, an effective health policy should
be based on primary prevention, which must consider
education and intervention to diminish chronodisrup-
tion and prevent the risk of prevalent disabilities. The
link between psychiatric disorders and CT is in its
infancy and functional neuroimaging studies carried
out in the future will advance our understanding and
go towards the integration of molecular and biochemical
aspects. Studies have already investigated the neuro-
structural basis for the relationship between sleep and
mental disorders, and the prefrontal cortex and striatum
have been implicated in differences among CT. As these
areas are related to mood variations, these structures
likely constitute one of the underlying mechanisms in
the relationship between depressive mood and ET
(Hasler et al., 2012). The other structure that has been im-
plicated in the correlation between sleepwake regu-
lation and CT is the posterior hypothalamic region,
which is also an important memory region (Schmidt
et al., 2012). These findings may explain the difference
between ET and MT in perceived performance on
memory tests, despite the lack of differences in actual
performance on the Word-Pairs Associated and Word
List with Emotional Content or Stroop tests (Hidalgo
et al., 2004; Schmidt et al., 2012).
The differences among CT should be understood from a
wide perspective, leading us to state that they are associ-
ated with different lifestyles. Although we should avoid a
simplistic shortcut of associating ET to some negative
aspects, the data point to the idea that an ET pattern is
a risk factor for some disorders, whereas MT is a protec-
tion factor. The habits of the MT are healthier in general
than those found in the ET. Although this is important all
along the life cycle, it seems more crucial in adolescence,
when the adult lifestyle is beginning to take shape, and in
the aging period, when health problems related to life
habits are more frequent. Hence, it may be of interest
to consider this variable in a bio-psycho-social health
model, in order to design more efficient preventive and
therapeutic approaches.
This is a new field of interest to professionals in many
different areas (research, labor, academic and clinic)
which must be formed and well informed on this knowl-
edge in order to integrate chronobiological aspects in
their daily practice as needed. Therapeutic approaches
to psychiatric disorders should take into account circa-
dian rhythmicity. In many cases, it may suffice to estab-
lish regular time patterns of sleepwake, meals and
daily activity with a tendency towards an MT pattern of
functioning. Other strategies, such as exposition to light
therapy and melatonin administration, according to
each case, may also be effective in clinical management
and as protective measures against relapse of mental
health disorders. Understanding the genetics underlying
CT could help to identify individuals who are at risk or
vulnerable to certain life styles or health disorders. Pre-
vention should also include these findings in their initiat-
ives. Involving chronobiology in prevention implies not
only promoting healthy leisure activities or what to
do, but rather placing such activities in appropriate
daytime temporal mome nts or when to do them.
The authors thank Prof. Yvan Touitou for considering
the importance of the issue and for providing an
opportunity for this review to be published in
Chronobiology International.
Declaration of Interest: The authors report no conflicts
of interest. The authors alone are responsible for the con-
tents and writing of the paper.
A.A. was supported by a grant from the Spanish Min-
istry of Science and Innovation (PSI2009-12300), S.N.A.
was supported by grants from the BBSRC (BB/F022883/
1) and AFOSR (FA-9550-08-1-0080), M.P.H. was sup-
ported by a grant from CAPES, CNPq, FAPERGS and
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
FIPE/ HCPA-Brasil and C.R. was supported by a
sabbatical leave.
Achari KV, Pati AK. (2007). Morningnesseveningness preference in
Indian school students as function of gender, age, and habitat.
Biol. Rhythm Res. 38:18.
Adan A. (1991). Influence of morningnesseveningness preference in
the relationship between body temperature and performance: a
diurnal study. Pers. Indiv. Differ. 12:11591169.
Adan A. (1993). Circadian variations in psychological measures: a new
classification. Chronobiologia 20:145162.
Adan A. (1994). Chronotype and personality factors in the daily con-
sumption of alcohol and psychostimulants. Addiction 89: 455462.
Adan A. (2012). A chronobiological approach to addiction. J. Subs. Use
(in press).
Adan A, Almirall H. (1991). Horne and Östberg morningnessevening-
ness questionnaire: a reduced scale. Pers. Indiv. Differ. 12:241253.
Adan A, Natale V. (2002). Gender differences in morningnessevening-
ness preference. Chronobiol. Int. 19:709720.
Adan A, Sánchez-Turet M. (1995). Smoking effects on diurnal vari-
ations of cardiovascular parameters. Int. J. Psychophysiol.
Adan A, Sánchez-Turet M. (2000). Effects of smoking on diurnal vari-
ations of subjective activation and mood. Hum. Psychopharmacol.
Clin. Exp. 15:287293.
Adan A, Sánchez-Turet M. (2001). Gender differences in diurnal vari-
ations of subjective activation and mood. Chronobiol. Int.
Adan A, Prat G, Sánchez-Turet M. (2004). Effects of nicotine depen-
dence on diurnal variations of subjective activation and mood.
Addiction 98:15991607.
Adan A, Natale V, Caci H. (2008). Cognitive strategies and circadian
typology. In Léglise AL (Ed.). Progress in circadian rhythms
research. Nova Biomedical Books. New York, NY: Nova Science
Publishers, Inc., pp. 141161.
Adan A, Lachica J, Caci H, Natale V. (2010a). Circadian typology and
temperament and character personality dimensions. Chronobiol.
Int. 27:181193.
Adan A, Natale V, Caci H, Prat G. (2010b). Relationship between circa-
dian typology and functional and dysfunctional impulsivity. Chron-
obiol. Int. 27:606619.
Akerstedt T (2003). Shift work and disturbed sleep/wakefulness. Occup.
Med. (Lond) 53:8994.
Andershed K-A. (2005). In sync with adolescence: the role of morning-
nesseveningness in adolescence. New York, NY: Springer.
Anderson MJ, Petros TV, Beckwith BE, Mitchell WW, Fritz S. (1991).
Individual differences in the effect of time of day on ling term
memory access. Am. J. Psychol. 104:241255.
Andretic R, Franken P, Tafti M. (2008). Genetics of sleep. Annu. Rev.
Genet. 42: 361388.
Archer SN, Robilliard D, Skene DJ, Smits MG, Williams A, Arendt J, von
Schantz M. (2003). A length polymorphism in the circadian clock
gene Per3 is linked to delayed sleep phase syndrome and extreme
diurnal preference. Sleep 26:413415.
Archer SN, Viola AU, Kyriakopoulou V, von Schantz M, Dijk DJ. (2008).
Inter-individual differences in habitual sleep timing and entrained
phase of endogenous circadian rhythms of BMAL1, PER2 and PER3
mRNA in human leukocytes. Sleep 31:608617.
Archer SN, Carpen JD, Gibson M, Lim GH, Johnston JD, Skene DJ, von
Schantz M. (2010). Polymorphism in the PER3 promoter associates
with diurnal preference and delayed sleep phase disorder. Sleep
Arendt J. (2006). Melatonin and human rhythms. Chronobiol. Int.
Arendt J. (2009). Managing jet lag: Some of the problems and possible
new solutions. Sleep Med. Rev. 13:249256.
Aziz S, Jackson CJ. (2001). A comparison between three and five factor
models of Pakistani personality data. Pers. Indiv. Differ.
Baehr E, Revelle W, Eastman CI. (2000). Individual differences in the
phase and amplitude of the human circadian temperature
rhythm: with an emphasis on morningnesseveningness. J. Sleep
Res. 9;117127.
Bailey SL, Heitkemper MM. (1991). Morningnesseveningness and
early morning salivary cortisol levels. Biol. Psychol. 32:181192.
Bailey SL, Heitkemper MM. (2001). Circadian rhythmicity of cortisol
and body temperature: Morningnesseveningness effects. Chrono-
biol. Int. 18:249261.
Barclay NL, Eley TC, Buysse DJ, Archer SN, Gregory AM. (2010).
Diurnal preference and sleep quality: same genes? A study of
young adult twins. Chronobiol. Int. 27:278296.
Barclay NL, Eley TC, Mill J, Wong CC, Zavos HM, Archer SN, Gregory
AM. (2011). Sleep quality and diurnal preference in a sample of
young adults: associations with 5HTTLPR, PER3, and CLOCK
3111. Am. J. Med. Genet. B Neuropsychiatr. Genet. 156B:681690.
Benedetti F, Barbini B, Colombo C, Smeraldi E. (2007). Chronothera-
peutics in a psychiatric ward. Sleep Med Rev
. 11:509522.
Benloucif S, Guico MJ, Reid KJ, Wolfe LF, LHermitage-Baleriaux M,
Zee PC. (2005). Stability of melatonin and temperature as circadian
phase markers and their relation to sleep in humans. J. Biol.
Rhythms 20:178188.
Besoluk S, Önder I, Deveci I. (2011). Morningnesseveningness prefer-
ences and academic achievement of university students. Chrono-
biol. Int. 28:118125.
Biss RK, Hasher L. (2012). Happy as a lark: morning-type younger and
older adults are higher in positive affect. Emotion 12:437441.
Blatter K, Cajochen C. (2007). Circadian rhythms in cognitive perform-
ance: methodological constrains, protocols, theoretical underpin-
nings. Physiol. Behav. 90:196208.
Bohle P, Tilley AJ, Brown S. (2001). A psychometric evaluation of the
Early/Late Preferences Scale. Ergonomics 44:887900.
Borbely A. (1982). A two process model of sleep regulation. Hum.
Neurobiol. 1:195204.
Borisenkov MF, Perminova EV, Kosova AL. (2010). Chronotype, sleep
length, and school achievement of 11- to 23-year-old students in
northern European Russia. Chronobiol. Int. 27:12591270.
Borisenkov MF, Perminova EV, Kosova AL. (2012). Impact of perinatal
photoperiod on the chronotype of 11- to 18-years-olds in northern
European Russia. Chronobiol. Int. 29:305310.
Caci H, Nadalet L, Staccini P, Myquel M, Boyer P. (1999). Psychometric
properties of the French version of the Composite Scale of Morn-
ingness in adults. Eur. Psychiatr. 14:284290.
Caci H, Robert P, Boyer, P. (2004). Novelty seekers and impulsive sub-
jects are low in morningness. Eur. Psychiatr. 19:7984.
Caci H, Adan A, Bohle P, Natale V, Pornpitakpan C, Tilley A. (2005a).
Transcultural properties of the composite scale of morningness:
the relevance of the morning affect factor. Chronobiol. Int.
Caci H, Mattei V, Bayle FJ, Nadalet L, Dossios C, Robert P, Boyer P.
(2005b). Impulsivity but not venturesomeness is related to morn-
ingness. Psychiatr. Res. 134:259265.
Caci H, Deschaux O, Adan A, Natale V. (2009). Comparing three morn-
ingness scales: age and gender effects, structure and cut-off cri-
teria. Sleep Med. 10:240245.
Carpen JD, Archer SN, Skene DJ, Smits M, von Schantz M. (2005). A
single-nucleotide polymorphism in the 5-untranslated region of
the hPER2 gene is associated with diurnal preference. J. Sleep
Res. 14:293297.
Carpen JD, von Schantz M, Smits M, Skene DJ, Archer SN (2006). A
silent polymorphism in the PER1 gene associates with extreme
diurnal preference in humans. J. Hum. Genet. 51:11221125.
de Castro JM. (2004). The time of day of food intake influences overall
intake in humans. J. Nutr. 134:104111.
Circadian Typology : A Comprehensive Review 
© Informa Healthcare USA, Inc.
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
Cavallera GM, Giudici S. (2008). Morningness and eveningness per-
sonality: a survey in literature from 1995 up till 2006. Pers. Indiv.
Differ. 44:321.
Cavallera GM, Boari G, Giudici S, Ortolano A. (2011). Cognitive par-
ameters and morning and evening types: two decades of research
(19902099). Percept. Mot. Skills 112: 649665.
Chang AM, Buch AM, Bradstreet DS, Klements DJ, Duffy JF. (2011).
Human diurnal preference and circadian rhythmicity are not
associated with the CLOCK 3111C/T gene polymorphism. J. Biol.
Rhythms 26:276279.
Chelminski I, Ferraro FR, Petros T, Plaud JJ. (1997). Horne and Östberg
questionnaire: a score distribution in a large sample of young
adults. Pers. Indiv. Differ. 23:647652.
Chelminski I, Ferraro FR, Petros T, Plaud JJ. (1999). An analysis of the
eveningnessmorningness dimension in depressive college
students. J. Affect. Disord. 52:1929.
Chelminski I, Petros T, Plaud JJ, Ferraro FR. (2000). Psychometric prop-
erties of the reduced Horne and Östberg. Pers. Indiv. Differ.
Ciarkowska W. (1997). Time of day effect on logical reasoning and heart
rate change. Stud. Psychol. 35:141163.
Clarisse R, Le Floch N, Kindelberger C, Feunteun P. (2010). Daily
rhythmicity of attention in morning- vs. evening-type adolescents
at boarding school under different psychosociological testing con-
ditions. Chronobiol. Int. 27:826841.
Cloninger CR. (1994). Temperament and personality. Curr. Biol .
Colquhoun WP. (1971). Biological rhythms and human performance.
London: Academic Press.
Conroy DA, Hairston IS, Arnedt JT, Hoffmann RF, Armitage R, Brower
KJ. (2012). Dim light melatonin onset in alcohol-dependent men
and women compared with healthy controls. Cronobiol. Int.
Corbera X, Grau C. (1993). Diurnal type and hemispheric asymmetry.
Cortex 29:519528.
Costa PT, McCrae RR. (1992). Revised NEO Personality Inventory (NEO
PI-R) and NEO Five-Factor Inventory (NEO-FFI): professional
manual. Odessa, FL: Psychological Assessment Resources.
Drennan MD, Klauber MR, Kripke DF, Goyette LM. (1991). The effects
of depression and age on the HorneOstberg morningness
eveningness score. J. Affect. Disord. 23:93
Danel T, Touitou Y. (2004). Chronobiology of alcohol: from chronoki-
netics to alcohol-related alterations of the circadian system. Chron-
obiol. Int. 21:923935.
Díaz-Morales JF. (2007). Morning and evening-types: exploring their
personality styles. Pers. Indiv. Differ. 43:769778.
Díaz-Morales JF, Ferrari JR, Cohen JR. (2008). Indecision and avoidant
procrastination: the role of morningness-eveningness and time
perspective in chronic delay lifestyles. J. General. Psychol.
Dibner C, Schibler U, Albrecht U. (2010). The mammalian circadian
timing system: organization and coordination of central and per-
ipheral clocks. Annu. Rev. Physiol. 17:517549.
Dijk DJ, Archer SN. (2009). PERIOD3, circadian phenotypes, and sleep
homeostasis. Sleep Med. Rev. 14:151160.
Di Milia L. (2005). A psychometric evaluation and validation of the Pre-
ferences Scale Chronobiol. Int. 22:679693.
Di Milia L, Bohle, P. (2009). Morningness or morning affect? A short
composite scale of morningness. Chronobiol. Int. 26:494509.
Di Milia L, Muller H. (2012). Does impression management impact the
relationship between morningnesseveningness and self-rated
sleepiness? Pers. Indiv. Differ. 36:19531964.
Di Milia L, Smith PA, Folkard S. (2004). Refining the psychometric
properties of the circadian type inventory. Pers. Indiv. Differ.
Di Milia L, Smith PA, Folkard S. (2005). A validation of the revised
circadian type inventory in a working sample. Pers. Indiv. Differ.
Di Milia L, Wikman R, Smith P. (2008). Additional psychometric
evidence and construct validity for a revised Preferences Scale of
Morningness. Chronobiol. Int. 25:776787.
Dockray S, Steptoe A. (2011). Chronotype and diurnal cortisol profile in
working women: differences between work and leisure days.
Psychoneuroendocrinology 36:649655.
Duffy JF, Dijk DJ, Hall EF, Czeisler CA. (1999). Relationship of endogen-
ous circadian melatonin and temperature rhythms to self-reported
preference for morning or evening activity in young and older
people. J. Invest. Med. 47:141150.
Duffy JF, Rimmer DW, Czeisler CA. (2001). Association of intrinsic cir-
cadian period with morningnesseveningness, usual wake time,
and circadian phase. Behav. Neurosci. 115:895899.
Duffy JF, Cain SW, Chang AM, Phillips KJ, Münch MY, Gronfier C,
Wyatt JK, Dijk DK, Wright KP, Czeisler CA. (2011). Sex differences
in the near-24-hours intrinsic period of the human circadian
timing system. Proc. Natl Acad. Sci. USA
Eastman CI, Burgess H. (2009). How to travel the world without jet lag.
Sleep Med. Rev. 4:241255.
Ebisawa T, Uchiyama M, Kajimura N, Mishima K, Kamei Y, Katoh M,
Watanabe T, Sekimoto M, Shibui K, Kim K, Kudo Y, Ozeki Y, Su-
gishita M, Toyoshima R, Inoue Y, Yamada N, Nagase T, Ozaki N,
Ohara O, Ishida N, Okawa M, Takahashi K, Yamauchi T. (2001).
Association of structural polymorphisms in the human period3
gene with delayed sleep phase syndrome. EMBO Rep. 2:342346.
Elmore SK, Dahl K, Avery DH, Savage MV, Brengelmann GL. (1993).
Body temperature and diurnal type in women with seasonal affec-
tive disorder. Health Care Women Int. 14:1726.
Eysenck HJ, Eysenck SBG. (1975). Manual of the Eysenck Personality
Questionnaire. London: Hodder & Stoughton.
Eysenck SBG, Eysenck HJ, Barrett P. (1985). A revised version of the
psychoticism scale. Pers. Indiv. Differ. 6:2129.
Fabbri M, Antonietti A, Gioirgetti M, Tonetti L, Natale V. (2007). Circa-
dian typology and style of thinking differences. Learn. Indiv. Diff.
Fabbri M, Mencarelli C, Adan A, Natale V. (2012) Time of day and cir-
cadian typology on memory retrieval. Biol. Rhythm Res (in press).
Fan J, Posner M. (2004). Human attentional networks. Psychiatr. Prax.
Fan J, MccCandliss BD, Sommer T, Raz A, Posner MI. (2002). Testing
the efficiency and independence of attentional networks. J. Cogn.
Neurosci. 14:340347.
Flower DJC, Irvine D, Folkard S. (2003). Perception and predictability
of travel fatigue after long-haul flights: a retrospective study.
Aviat. Space Environ. Med. 74:173179.
Folkard S. (1990). Circadian performance rhythms: some practical and
theoretical implications. Phil. Trans. R. Soc. Lond. B 327:543553.
Folkard S, Monk TH, Lobban MC. (1979). Towards a predictive test of
adjustment to shift work. Ergonomics 22:7991.
Franken P, Dijk DJ. (2009). Circadian clock genes and sleep homeosta-
sis. Eur. J. Neurosci. 29:18201829.
Furnham A, Hughes K. (1999). Individual difference correlates of
nightwork and shift-work rotation. Pers. Indiv. Differ. 26:830848.
Gaina A, Sekine M, Kanayama H, Takashi Y, Hu L, Sengoku K, Kagami-
mori S. (2006). Morningevening preference: Sleep pattern spec-
trum and lifestyle habits among Japanese junior high school
pupils. Chronobiol. Int. 23:607621.
Gaina A, Sekine M, Kakayama H, Takashi Y, Hu L, Sengoku K, Kagami-
mori S. (2011). Morningevening preference: sleep pattern espec-
trum and lifestyle habits among Japanese junior high school
pupils. Chronobiol. Int. 23:607621.
Gale C, Martyn C. (1998). Larks and owls and health, wealth and
wisdom. Brit. Med. J. 317:16751677.
Gamble KL, Motsinger-Reif AA, Hida A, Borsetti HM, Servick SV, Ciar-
leglio CM, Robbins S, Hicks J, Carver K, Hamilton N, Wells N,
Summar NK, McMahon DG, Johnson CH. (2011). Shift work in
nurses: contribution of phenotypes and genotypes to adaptation.
PLoS One 6:e18395.
 A. Adan et al.
Chronobiology International
Chronobiol Int Downloaded from by Libreria Joaquin Ibanez on 10/10/12
For personal use only.
Gau SS, Soong WT. (2003). The transition of sleepwake patterns in
early adolescence. Sleep 26:449454.
Gau SS, Shang CY, Merikangas KR, Chiu YN, Soong WT, Cheng AT
(2007). Association between morningnesseveningness and behav-
ioral/emotional problems among adolescents. J. Biol. Rhythms
Giampietro M, Cavallera G. (2007). Morning and evening types and
creative thinking. Pers. Indiv. Differ. 42:453463.
Giannotti F, Cortesi F, Sebastiani T, Ottaviano S. (2002). Circadian pre-
ference, sleep and daytime behaviour in adolescence. J. Sleep Res.
Gibertini M, Graham C, Cook MR. (1999). Self-report of circadian type
reflects the phase of the melatonin rhythm. Biol. Psychol. 50:1933.
Giglio LM, Magalhães PV, Andersen ML, Walz JC, Jakobson L, Kapc-
zinski F. (2010). Circadian preference in bipolar disorder. Sleep
Breath. 14:153155.
Gillooly PB, Smolensky MH, Albright DL, Hsi B, Thorne DR. (1990).
Circadian variation in human performance evaluated by Walter
Reed Performance Assessment Battery. Chronobiol. Int. 7:143153.
Gómez-Abellán P, Madrid JA, Ordovás JC, Garauleta M. (2012). Chron-
obiological aspects of obesity and metabolic syndrome. Endocrinol.
Nutr. 59:5061.
Gray EK, Watson D. (2002). General and specific traits of personality
and their relationship to sleep and academic performance. J.
Pers. 70:177206.
Griefahn B, Robens S. (2008). The cortisol awakening response: a pilot
study on the effects of shift work, morningness and sleep duration.
Psychoneuroendocrinology 33:981988.
Griefahn B, Künemund C, Golka K, Their R, Degen G. (2002). Melato-
nin synthesis: a possible indicator of intolerance to shift work.
Am. J. Ind. Med. 42:427436.
Groeger JA, Zijlstra FR, Dijk DJ. (2004). Sleep quantity, sleep difficulties
and their perceived consequences in a representative sample of
some 2000 British adults. J. Sleep Res. 13:359371.
Groeger JA, Viola AU, Lo JC, von Schantz M, Archer SN, Dijk DJ. (2008).
Early morning executive functioning during sleep deprivation is
compromised by a PERIOD3 polymorphism. Sleep 31:11591167.
Guerrien A, Leconte-Lambert C, Leconte L. (1993). Time of day effects
on attention and memory efficiency: is the chronopsychology a
method for studying the functioning of the human subjects?
Psychol. Belg. 33:159170.
Gupta S, Pati AK. (1994). Characteristics of circadian rhythm in six vari-
ables of morning active and evening active healthy human sub-
jects. Ind. J. Physiol. Pharmacol. 38:101107.
Guthrie JP, Ash RA, Bendapudi V. (1995). Additional validity evidence
for a measure of morningness. J. App. Psych. 80:186
Hagenauer MH, Perryman JI, Lee TM, Carskadon MA. (2009). Adoles-
cent changes in the homeostatic and circadian regulation of sleep.
Dev. Neurosci. 31:276284.
Harada T, Kobayashi R, Wada K, Nishihara R, Kondo A, Akimitsi O,
Noji T, Taniwaki N, Nakade M, Krejci M, Takeuchi H. (2011).
Effects of birth season on circadian typology appearing in Japanese
young children aged 2 to 12 years disappears in older students aged
18 to 25 years. Chronobiol. Int. 28:638642.
Hasan S, Santhi N, Lazar AS, Slak A, Lo J, von Schantz M, Archer SN,
Johnston JD, Dijk DJ. (2012). Assessment of circadian rhythms in
humans: comparison of real-time fibroblast reporter imaging
with plasma melatonin. Faseb J. 26:24142423.
Hasher L, Chung C, May CP, Foong N. (2002). Age, time of testing, and
proactive interference. Can. J. Exp. Psychol. 56:200207.
Hasler BP, Germain A, Nofzinger EA, Kupfer DJ, Krafty RT, Rothenber-
ger SD, James JA, Bi W, Buysse DJ. (2012). Chronotype and diurnal
patterns of positive affect and affective neural circuitry in primary
insomnia. J Sleep Res. (in press)
Hidalgo MP, Zanette CB, Pedrotti M, Souza CM, Nunes PV, Chaves ML.
(2004). Performance of chronotypes on memory tests during the
morning and the evening shifts. Psychol. Rep. 95:7585.
Hidalgo MP, Caumo W, Posser M, Coccaro SB, Camozzato AL, Chaves
ML. (2009). Relationship between depressive mood and chrono-
type in healthy subjects. Psychiatry Clin. Neurosci. 63:283290.
Hogben AL, Ellis J, Archer SN, von Schantz, M. (2007). Conscientious-
ness is a predictor of diurnal preference. Chronobiol. Int.
Horne JA, Östberg O. (1976). A self-assessment questionnaire to deter-
mine morningnesseveningness in human circadian systems.
Int. J. Chronobiol. 4:97110.
Horne JA, Brass CG, Petitt AN. (1980). Circadian performance differ-
ences between morning and evening types. Ergonomics 23:2936.
Hornik J, Miniero G. (2009). Synchronyeffects on customers responses
and behaviors. Int. J. Res. Mark. 26:3440.
Hsu CY, Gau SS-F, Shang C-Y, Chiu Y-N, Lee M-B. (2012). Associations
between chronotypes, psychopathology, and personalityamong in-
coming college students. Chronobiol. Int. 29:491501.
Huang M-C, Ho C-W, Chen C-H, Liu S-H, Chen C-C, Leu SJ. (2010).
Reduced expression of circadian clock genes in male alcoholic
patients. Alcohol. Clin. Exp. Res. 34:18991904.
Intons-Peterson MJ, Rocchi P, West T. McLellan K, Hackney A. (1998).
Aging, optimal testing times, and negative priming. J. Exp. Psychol.
Learn. Mem. Cogn. 24:362376.
Intons-Peterson MJ, Rocchi P, West T. McLellan K, Hackney A. (1999).
Aging, testing at preferred or non preferred times (testing optimal-
ity), and false memory. J. Exp. Psychol. Learn. Mem. Cogn. 25:2340.
Iskra-Golec I, Smith L. (2006). Ultradian and asymmetric rhythms of
hemispheric processing speed. Chronobiol. Int. 23:12291239.
Jackson LA, Gerard DA. (1996). Diurnal types, the big five personality
factors, and other personal characteristics. J. Soc. Behav. Person.
Jamieson AO, Zammit GK, Rosenberg RS, Davis JR, Walsh JK. (2001).
Zolpidem reduces the sleep disturbance of jet lag. Sleep Med.
Jankowski S. (2012). Morningness/eveningness and satisfaction with
life in a Polish sample. Chronobiol. Int. 29:780785.
Jenkins A, Archer SN, von Schantz M. (2005). Expansion during
primate radiation of a variable number tandem repeat in the
coding region of the circadian clock gene period3. J. Biol.
Rhythms 20:470472.
Johansson C, Willeit M, Smedh C, Ekholm J, Paunio T, Kieseppa T,
Lichtermann D. N. Praschak-Rieder, A. Neumeister, L. Nilsson G,
Kasper S, Peltonen L, Adolfsson R, Schalling M, Partonen T.
(2003). Circadian clock-related polymorphisms in seasonal
affective disorder and their relevance to diurnal preference.
Neuropsychopharmacology 28:734739.
Jones KH, Ellis J, von Schantz M, Skene DJ, Dijk DJ, Archer SN. (2007).
Age-related change in the association between a polymorphism in
the PER3 gene and preferred timing of sleep and waking activities.
J. Sleep Res. 16:1216.
Juda M. (2010). The importance of chronotype in shift work research.
Institute for Medical Psychology, Ludwigs-Maximilians-
Universität, Munich, Unpublished doctoral thesis.
Kaliterna, L., Vidacek, S., Prizmic, Z., & Radosevic-Viadacek, B. (1995).
Is tolerance to shiftwork predictable from individual difference
measures. Work Stress 9:140147.
Kanazawa S, Perina K. (2009). Why night owls are more intelligent.
Pers. Indiv. Differ. 47:685690.
Katzenberg D, Young T, Finn L, Lin L, King DP, Takahashi JS, Mignot
M. (1998). A CLOCK polymorphism associated with human
diurnal preference. Sleep 21:569576.
Kerkhof GA, Van Dongen HPA. (1996). Morning-type and evening-
type individuals differ in the