Neuroscience Research 46 (2003) 23
/31

www.elsevier.com/locate/neures

Circadian fluctuation of time perception in healthy human subjects

Kenichi Kuriyama a,b, Makoto Uchiyama b,*, Hiroyuki Suzuki b,c, Hirokuni Tagaya b,
Akiko Ozaki b, Sayaka Aritake b, Yuichi Kamei d, Toru Nishikawa a,
Kiyohisa Takahashi e
a
Department of Psychiatry and Behavioral Science, Tokyo Medical and Dental University, Yushima, Bunkyo-Ku, 113-0034 Tokyo, Japan
b
Department of Psychophysiology, National Institute of Mental Health, National Center of Neurology and Psychiatry, Kohnodai, Ichikawa 272-0827,
Japan
c
Graduate School of Humanities, Nihon University, Sakurajosui, Setagaya-ku, Tokyo 156-8550, Japan
d
Department of Psychiatry, Kohnodai Hospital, NCNP, Kohnodai, Ichikawa 272-8516, Japan
e
National Center of Neurology and Psychiatry, Ogawahigashi-cho, Kodaira 187-8551, Japan

Received 6 September 2002; accepted 9 December 2002

Abstract

Previous studies suggested that various psychophysiological factors have influences on human time perception. In particular,
working memory loads, time of day, body temperature, and mood were known as important modifiers of time perception. The
purpose of this study is to elucidate factors affecting the short-term time perception under controlled condition. Fourteen healthy
young male adults participated in this study. Time perception sessions (TPS) were conducted 4 times at 0900, 1300, 1700 and 2100 h.
The TPS consisted of five 10-s time production trials under five different conditions (control trial, those with reward, and 3 different
dual-load working memory tasks). Subjective status was assessed using visual analogue scales (VAS). To verify a participants
vigilance state, an alpha attenuation coefficient (AAC) was calculated. Two-way repeated measures ANOVA for produced time
revealed a significant main effect of session, but no effect of task or interaction. Although produced time was not correlated with
AACs or VAS scores, there was a significant negative correlation between produced time and core body temperature. These results
suggest that human short-term time perception may be more influenced by circadian rhythm than working memory load or
psychophysiological status.
# 2003 Elsevier Science Ireland Ltd and the Japan Neuroscience Society. All rights reserved.

Keywords: Time perception; Circadian rhythm; Interval timing clock; Working memory; Mood; Core body temperature; Time production test

1. Introduction is located in the suprachiasmatic nuclei of the hypotha-

Animal studies have suggested that organisms have
two endogenous timekeeping systems; a circadian clock
that provides time of day and an interval timing clock
that measures passage of time like a stopwatch (Virgi-
nia, 1996). The circadian clock is driven by self-
sustaining oscillator with a period of about 24 h, which
* Corresponding author. Tel.: /81-47-375-4756; fax: /81-47-375-
4771.
E-mail address: macoto@ncnp-k.go.jp (M. Uchiyama).
0168-0102/03/$ - see front matter # 2003 Elsevier Science Ireland Ltd and the Japan Neuroscience Society. All rights reserved.
doi:10.1016/S0168-0102(03)00025-7
lamus and synchronized to the light
/dark cycle (Moore-
Ede et al., 1983). In contrast to the circadian clock, the
interval timing clock counts the number of signals that
generated in certain intervals (Gibbon and Church,
1981; Roberts and Holder, 1984). These clocks play a
crucial role in long lasting natural selection because the
ability to anticipate risks and opportunities that occur in
the environment could increase the chance for the
animal to survive.
It has been well-documented that humans can per-
ceive the passage of time without referring to an external
clock or stopwatch, whether duration to be perceived
24 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31
was longer (long-term time perception) or shorter
(short-term time perception). Previous studies on long-
term time perception, duration to be perceived was
widely ranged from minutes to hours (Mischel et al.,
1969; Watts and Sharrock, 1984; Zakay, 1992; Campbell
et al., 2001; Conti, 2001; Rammsayer et al., 2001),
whereas those on short-term time perception concerned
a perceived period of less than a minute (Treisman,
1963; Rammsayer and Vogel, 1992; Fortin et al., 1993;
Fortin and Breton, 1995; Ivry, 1996; Fortin and
Rousseau, 1998; Fortin and Masse, 1999; Casini and
Ivry, 1999; Rammsayer et al., 2001). Though various
mechanisms of long-term time perception have been
postulated, there was significant controversy on the
nature and functions because definitions of long-term
time perception were apparently diverse from study to
study, and because different study protocols were used
in the previous studies (Mischel et al., 1969; Watts and
Sharrock, 1984; Zakay, 1992; Campbell et al., 2001;
Conti, 2001; Rammsayer et al., 2001). In contrast, the
experimental settings in which short-term time percep-
tion was investigated were similar and comparable in the
previous studies (Treisman, 1963; Rammsayer and
Vogel, 1992; Fortin et al., 1993; Ivry, 1996; Fortin and
Rousseau, 1998; Rammsayer et al., 2001). Moreover,
recent progresses in functional imaging technique have
suggested that certain regions of the brain were acti-
vated when short-term time perception tasks were
loaded (Maquet et al., 1996; Harrington et al., 1998;
Pouthas et al., 1999; Schubotz et al., 2000), though
variations and changes of perceived time were not
monitored in these studies because of the methodologi-
cal limitations.
In the previous reports on short-term time perception,
it was indicated that psychological or cognitive factors
influenced human time perception. Some researchers
have reported that short-term memory or working
memory tasks that were loaded simultaneously with a
short-term time perception task had an effect on
perceived time (Fortin et al., 1993; Fortin and Breton,
1995; Fortin and Rousseau, 1998; Fortin and Masse,
1999). Enhancement of subjects attention to temporal
information has been reported to give more accurate
short-term time perception (Zakay, 1992; Casini and
Ivry, 1999; Rammsayer et al., 2001).
In chronobiological studies, it was reported that
short-term time perception seems to fluctuate across
the day (Aschoff, 1998; Aschoff and Daan, 1997;
Morofushi et al., 2001). Under the free-running condi-
tion, produced time (10-s) has been reported to change
along with core body temperature, which is a possible
marker of circadian rhythm. During the subjective day,
produced time (10-s) decreased from the morning into
the night (Aschoff, 1998). These findings might suggest
that short-term time perception may be modulated by a
circadian oscillator.
However, there may be another possible explanation.
That is, psychophysiological status such as mood,
alertness, or tiredness, or changes in the environment
such as light, posture or exercise may produce diurnal
fluctuation in perceived time, providing that diurnal
fluctuation of the time perception documented in the
previous studies may have been a consequence of non-
circadian origin.
Here we investigate short-term time perception in
healthy human and study factors affecting it. In the
present study, we focused on the effects of working
memory load and psychophysiological status on time
perception and those of diurnal fluctuations, as well as
the possible interaction among those factors. To differ-
entiate the influence of these factors clearly, we utilized a
regimen, in which a constant resting wakefulness with
semi-recumbent position was kept for 17 h, so that we
excluded potential confounding factors that may have
influences on the participants psychological and physi-
cal status and may mask the differences of the experi-
mental setting.
2. Methods
2.1. Participants
Fourteen healthy young male volunteers aged 18
/24
years (mean age9/S.D., 20.99/1.8 yr) participated in the
present study. A physician and a psychiatrist examined
them and found that no participants had a history of, or
suffered from, neurological or psychiatric disorders, or
had a history of using any psychoactive drugs. They
were asked to abstain from caffeine, nicotine, and
alcohol for a week prior to the experiment. They were
instructed to keep to a regular sleep
/wake schedule and
to record sleep logs for 2 weeks. We confirmed the
validity of their sleep logs by using an ambulatory wrist
activity recorder (Actiwatch-L, Mini-Mitter co., Inc.
Bend, OR) for a week prior to the experiment, and
found that the participants had regular sleep
/wake
habits without marked weekday
/weekend differences
in sleep schedule. The experimental protocol was
approved by the Intramural Research Board of Na-
tional Center of Neurology and Psychiatry, and each
participant gave his informed consent after the nature,
the purpose, and possible risks of the experiment had
been explained in detail. All the experiment was
performed at the time isolation laboratory of National
Center of Neurology and Psychiatry.
2.2. Experimental design
2.2.1. Overview
Experimental schedule is illustrated in Fig. 1. The
participant entered a two-day laboratory experimental
 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31 25

Fig. 1. A schematic schedule of the two-day study is shown. The participant was instructed to retire to temporal isolation facilities at 2300 h (Day 1)
and to stay in bed until 0700 h (Day 2) under complete darkness (B/0.1 lx). Thereafter, they were enforced to keep awake in a semi-recumbent
position under room light conditions (180 lx) until 2400 h (Day 2). Time production sessions (TPS) were carried out 4 times (0900, 1300, 1700 and
2100 h) on day 2. The TPS consisted of VAS, AAT, and five TPTs under five different psychological conditions; TPT without any tasks (control
task), that with a reward (5000 ) for accuracy of the perception, that with memory task of 8-digit numbers, that with memory task of 4 words, and
that with memory task of a figure of Bentons memory test. Three memory tasks were designed as dual-load tasks. The tasks were presented to the
participant in a randomized crossover manner. At the beginning to each TPS, VAS and AAT were administered.

session. On the first day (Day 1), the participant arrived
at the laboratory at 1900 h. Setting of rectal temperature
sensor (inserted 10
/15 cm from the anus) and attach-
ment of EEG electrodes were completed between 2000
and 2200 h. The participant was instructed to retire at
2300 h and to stay in bed until 0700 h on the second
experiment day (Day 2) when he was awakened by the
technician. During this period, polysomnographic re-
cordings were conducted under complete darkness (B/
0.1 lx). Thereafter, the participant was enforced to keep
awake on a semi-recumbent bed under room light
conditions (180 lx) until 2400 h. During this period on
Day 2, time perception protocol (TPP) was carried out.
During the TPP, TP session (TPS) was conducted 4
times at 0900, 1300, 1700 and 2100 h. Iso-caloric meals
(470 kcal) were given an hour before every TPS (0800,
1200, 1600 and 2000 h). Non-sparkling mineral water
was supplied at any time according to the participants
requests. During the study, the ambient temperature and
humidity in the time isolation facility were controlled at
24.09/0.5 8C and 609/5%, respectively. Throughout the
study, no time cues or information on the exact numbers
of TPSs and meals were given to the participant. Rectal
temperature measurement and polygraphic recordings
were carried out continuously throughout the study.
Temperature data were measured every 2 min and
stored telemetrically in a computerized monitoring
system (Vital Sense, resolution 0.02 8C, Mini-Mitter
Co. Inc., Bend, OR). Polygraphic recordings were
performed throughout the study by using a polygraphic
recorder (Neurofax, Nihon Kohden, Tokyo, Japan).
Polygraphic recordings consisted of C3
/A1, C4
/A2
and O1
/A1 EEGs in conformity with the 10
/20
electrode system, electrooculograms (left and right),
chin surface electromyogram, and electrocardiogram.
During TPP, the participant was enforced to keep
awake quietly without any amusements. The investiga-
tors monitored the participants status via a video
monitoring system and polygraphic recordings, and
enforced them to stay awake. The participant was
restricted to meet people except for meals.
2.2.2. TP Sessions
The TPS consisted of psychological assessment by
visual analogue scales (VAS), alpha attenuation tests
(AAT), and five time production trials (TPT) under five
different task conditions: (1) TPT without any tasks
(control task), (2) one with reward (5000 ) for accuracy
of the produced time, (3) one with a memory task
consisting of 8-digit numbers, (4) one with a memory
task consisting of 4 different words (common nouns
written in three different font styles of the Japanese
cursive (hiragana ) syllabary), and (5) one with a memory
task of a figure of Bentons Memory test. The memory
task was designed as a dual-load regimen, so that each
task load would require processing in working memory
simultaneously with the TPTs (Baddeley and Hitch,
1974; Baddeley, 1982, 1986). These trials were given to
the participant in a randomized crossover manner. All
the tasks were performed on a laptop computer with a
14-in. color liquid crystal monitor. The participant was
instructed to produce 10 s by pressing a space key on the
computer. The answers of the memory tasks were
provided by pressing a key on the numerical keyboard
of the computer. The screen was placed at 50 cm from
the participants eyes. All the data were transmitted via
26 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31

a local network system and were stored in the host
computer. Prior to the study, they trained on TPTs to
understand the operation of the PC-based test battery;
between 1900 and 2000 h on Day 1, a TP session was
carried out after the investigator gave a detailed
explanation. Produced time in the training trials ranged
from 9.40 to 14.04 s (11.459/1.14, mean9/S.D.). For all
performance tasks, participants were instructed to per-
form as quickly as possible, but without sacrificing
accuracy.

Fig. 2. TPS was conducted under five different conditions. Produced time was measured as the interval between keypress to begin and keypress to
end. This procedure was operated under following conditions: (A) memory task condition. The participant hits a space key to start the trial, this
keypress triggered the presentation of memory sets. It disappeared between 5 and 15 s to avoid itself influenced on the participants judgment of a 10-
s time production. After the TPT, several choices of the memory targets were presented and urged to select the correct answer. The memory task of
numbers and figure of Bentons Memory test had four choices (a and b), and the task of words had nine choices (c). (B) Control task condition. TPT
without any additional tasks was carried out. (C) Reward task condition. At the beginning of TPT, instruction of reward condition was presented to
the participant. The reward was given to the participant when his produced time was within the range of 109/0.5 s. The results of time production
were not announced until the end of all the study. In this trial, no additional task was carried out. Each condition was presented in a randomized
manner. The intervals between trial conditions were set up 70
/100 s to avoid being used by the participant as a ruler interval.
 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31
2.3. Measures
2.3.1. TP trials
Produced time was measured as the interval between
keypress in each different condition. The TPTs are
illustrated in Fig. 2. The memory tasks (Fig. 2A) was
designed to be loaded simultaneously during time
production. After pressing a space key to start the trial,
the participant was instructed to memorize the targets
displayed on the screen for 5
/15 s and to retain them
until the end of the trial. The participant was required to
select the correct targets among several choices. The
memory task for the numbers and for the figure of
Bentons Memory test had four choices (Fig. 2A (a and
b)), and the memory task for words had nine choices
(Fig. 2A c). The inter-trial interval ranged from 70 to
100 s. In the control task (Fig. 2B), the participant was
instructed directly to enter the TPT. In the reward task
(Fig. 2C), the amount of reward for accurate produced
time was presented prior to the TPT. A reward (5000)
was given to the participant when his produced time was
within the range of 109/0.5 s.
Information concerning the accuracy of time produc-
tion and the results of the memory tasks was not given
to the subject until the end of the study.
2.3.2. Alpha attenuation test
At the beginning of each TPS, AAT was adminis-
tered. In the AAT, three 1-min artifact-free EEG
recordings with eyes-open and those with eyes-closed
were obtained. The EEG data were digitized using a
sampling rate of 200 Hz. A half-amplitude low-fre-
quency filter was set at 0.5 Hz, and high frequency filter
at 50 Hz. The digitized data was stored on an optical
disk and analyzed offline with an EEG spectral analyses
package (FOCUS; MEGIS software, GmbH). Power
spectra of eyes-opened and eyes-closed EEG recordings
derived from O1 to A1 were analyzed by using a fast
Fourier transformation (FFT) within the alpha fre-
quency band (8
/12 Hz) using a bin size of 0.5 Hz. The
FFT analyses were conducted under 5.120-s Welch
tapered windows with 2.620-s overlap. This yielded 24
windows per 1-min epoch. The bins were averaged
across the 8
/12 Hz frequency range to produce an
estimate of alpha power in squared microvolts. The
ratio of the mean eyes-closed to mean eyes-opened alpha
power was defined as the alpha attenuation coefficient
(AAC).
2.3.3. Visual analogue scale
Subjective psychophysiological status was assessed
using VAS following the AAT in each TPS. The VAS
consisted of seven items, which were alertness, mood,
energy, tiredness, tension, motivation, and irritability.
The VAS was presented as a horizontal line 100 mm in
length, labeled for example Very Alert on the left and
27
Very Sleepy on the right. The participant was asked to
draw a vertical mark on the line at the point corre-
sponding to their present status. The VAS scores were
obtained by measuring the distance of the mark from
the left end of the line (higher scores being associated
with more intense feelings of each state).
2.4. Statistical analyses
To reduce inter-individual covariance, AACs and
VAS scores were standardized by using Fishers Z-
transformation, and core body temperature was trans-
formed to the mean deviations.
Two-factor 4 /5 (session-by-task) repeated measures
ANOVAs were used to analyze the produced time.
Greenhouse-Geisers adjustment factor (epsilon) was
calculated to adjust the degrees of freedom.
Endogenous circadian phase of the core body tem-
perature data was assessed by fitting of dual-harmonic
cosine curves (periods, 24 and 12 h) with a least squares
method (Uchiyama et al., 1995; Kubota et al., 2002;
Tagaya et al., 2002) using a software package (Kaleida-
Graph, Synergy Software, Reading, PA). The nadir of
the fitted curve was regarded as the circadian phase
marker. The nadir of the core body temperature was
05419/0122 (mean9/S.D.), suggesting the participants
circadian phase markers were normal range. Pearsons
correlation coefficients were calculated between pro-
duced time and other measures.
SPSS ver.10.0J for Windows (SPSS Inc., Chicago, IL)
was used for all the statistical analyses. All statistical
Fig. 3. The results of the TPTs of a representative participant are
shown. The horizontal axis represents time of day. The vertical axis
represents values of produced time. Black squares () represent
produced time under control task condition. Black circles (m) are
produced time with WM tasks of 8-digit numbers. Black triangles (?)
are produced time with WM tasks of 4 words. Open circles (k) are
produced time with WM tasks of figures. Open triangles (^) are
produced time with rewards. The produced time decreased as time goes
from 0900 to 1700 h regardless of presence of memory tasks or rewards
condition.
28 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31
analyses were two-tailed. The level of statistical signifi-
cance was set at P B/0.05.
3. Results
3.1. Overview
The results obtained from a representative participant
are shown in Fig. 3. Produced time was longer than 10 s
in the first two sessions and was decreased toward the
evening (Fig. 3(a)). The difference in trial condition did
not seem to influence the produced time.
The overall error rates for memory tasks obtained in
all participants were 0.018% (1/56; at the fourth session)
for number, 0.018% (1/56; at the second session) for
Bentons figure, 0.018% (1/56; at the second session) for
word, confirming that most of the dual-loaded memory
tasks were properly performed.
3.2. Effect of trial condition and time of day
Two-way repeated measures ANOVA revealed a
significant main effect of session (F (3, 11) /7.27, P /
0.005), but no effect of trial condition (F (4, 10) /0.65,
P /0.56) or interaction (F (12, 2) /1.70, P /0.15).
Thus, produced times were averaged across condition.
Produced times decreased with progression of the
sessions (Fig. 4).
Fig. 4. Diurnal change of produced times are shown. The horizontal
axis represents time of day. The vertical axis represents values of
produced time. Two-way repeated measures ANOVA of produced
time revealed a significant main effect of session (F (3, 11)/7.27, P /
0.005). Each data point represents the mean produced time of all the
conditions. Bars represent standard errors of mean. The produced time
which were averaged across condition decreased from the morning till
evening.
Fig. 5. Relation between produced time and core body temperature
together with the regression line are shown. The horizontal axis
represents standardized scores of produced time. The vertical axis
represents mean deviated values of rectal temperature. Standardized
produced time and standardized core body temperature were signifi-
cantly negatively correlated (r//0.44, P B/0.001).
3.3. Correlation with core body temperature, AACs and
VAS scores
3.3.1. Core body temperature
A significant negative correlation was found between
produced times and core body temperature (r //0.44,
P B/0.001; Fig. 5).
3.3.2. Alpha attenuation coefficients
No significant correlation was found between pro-
duced times and AACs (r //0.061, NS).
3.3.3. Visual analogue scale scores
No significant correlation was found between pro-
duced times and VAS scores (Alertness: r //0.074,
Mood: r /0.283, Energy: r/0.038, Tiredness: r//
0.236, Tension: r //0.013, Motivation: r /0.158, Irrit-
ability: r/0.028, NS).
4. Discussion
In the present study, short-term time production
showed significant diurnal fluctuations, in which re-
markable over-production in the morning decreased
towards the evening. However, time production was
not influenced by the working memory tasks that were
simultaneously loaded to the participant. In addition,
the vigilance level measured by using AAT or VAS
scores did not affect the time production, whereas core
body temperature at the TPS showed a significant
negative correlation with produced time.
 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31
Recent studies have suggested that there are potential
underlying psychophysiological factors on the diurnal
fluctuations in human time production: (1) accumula-
tion of fatigue or sleepiness according to the length of
the wakefulness (Phillips, 1977; Rammsayer, 1989), (2)
an effect of circadian rhythm that was synchronized to
day
/night transition (Aschoff and Daan, 1997; Aschoff,
1998; Campbell et al., 2001; Morofushi et al., 2001), and
(3) differences in psychological status at the time
production session, especially mood (Mischel et al.,
1969; Watts and Sharrock, 1984).
We used an AAT to evaluate vigilance level at the
time production test. The AAC, a ratio of EEG alpha
power values with eyes-closed to that with eyes-opened,
has been widely utilized in human psychophysiological
studies and recognized as a robust measure to evaluate
the vigilance level objectively (Saletu and Grunberger,
1984; Niedermeyer et al., 1989). Studies on diurnal
fluctuations of vigilance have revealed that longer
sustained wakefulness was associated with lower vigi-
lance level (Higuchi et al., 2001). Our results showing
that AAC did not have a significant correlation with
produced time suggest that the diurnal fluctuation in
produced time was unlikely to be a consequence of
changes in vigilance level.
We conducted continuous measurement of rectal
temperature throughout the study, and found that a
significant negative correlation between the produced
time and rectal temperature. Rectal temperature mea-
sured under the strictly controlled conditions, in which
ambient temperature and illumination, body movement,
and posture of the participant, and feeding were kept
constant, was reported to represent an oscillation of the
circadian pacemaker (Czeisler et al., 1990), suggesting
that the diurnal fluctuations in produced time may
reflect changes in the output of the circadian pacemaker.
In the present study, the experimental protocol was
carefully designed to minimize effects of non-circadian
factors. Thus, the fluctuation of core body temperature
obtained in the present study was likely to represent an
output of the circadian pacemaker. Aschoff, conducting
a time production test in human subjects exhibiting free-
run in the temporal isolation unit, has stressed that there
may be a circadian effect on short-term time perception,
though his study was not completely free from non-
circadian confounding factors that may have effects on
short-term time perception, such as psychological status,
or mental activities. In his experiments, longer or shorter
produced times were associated with lower or higher
core body temperatures, respectively (Aschoff, 1998). To
further confirm those potential effects of the circadian
pacemaker on short-term time perception which ob-
tained in his study and the present study, a longer
observation under the well-controlled condition is
necessary.
29
In contrast, about 70 years ago, when knowledge in
human circadian organization had not been established,
Hoagland described that a pathologically high skin
temperature in patients with various infectious diseases
was associated with prolongation of time estimation,
presumably regardless of circadian timing (Hoagland,
1933). Recently, a similar observation has been reported
by Hancock (1993). He found a significant but weak
negative correlation between manipulated intra-auricu-
lar temperature and produced times, though shortening
of produced time was not marked even when the intra-
auricular temperature was seriously elevated. These
studies may indicate that there are possible direct effects
of body temperature on time perception. However,
produced time obtained from patients suffering from
infectious diseases seemed to be influenced rather by
multiple pathological conditions than head temperature
alone, since infection may cause not only the elevation
of body temperature but also the releases of cytokines
that have marked effects on brain function (Dantzer,
2001; Owens and Babcock, 2002).
In the present study, participants were instructed to
memorize numbers, words, and figures before the
initiation of time production, and maintain them while
the participants were producing time. This procedure
was considered to provide a dual task loading. That is,
maintaining these memory targets was likely to occupy
the working memory areas in the brain (McCarthy et al.,
1994; Manoach et al., 1997). In the present study, the
simultaneous load on the working memory areas,
however, did not alter the time interval that the
participants produced. In contrast, Fortin et al. have
conducted a similar dual task procedure in searching
effects on produced time and found that memory tasks
presented during time production test were associated
with a prolongation of produced time as compared with
trials without memory tasks (Fortin and Breton, 1995;
Fortin and Masse, 1999). However, their results may
have been due to uncontrolled chronobiotic effects and
other behavioral confounding factors. Our results re-
vealed that the time of day had a more robust effect on
produced time in comparison with simultaneously
loaded memory tasks, even though there could have
been a scanty effect of such memory tasks that Fortin et
al. have reported.
Recently, functional imaging techniques were utilized
in identifying neural networks related to time produc-
tion (Harrington et al., 1998; Pouthas et al., 1999;
Schubotz et al., 2000). Maquet et al. have suggested that
time production process would be related to the function
of the prefrontal cortices (BA 10, 46) in their PET study
(Maquet et al., 1996). These areas have been reported to
have a functional relationship with working memory
process (Clifford et al., 2000). These functional imaging
studies suggested that time production process may be
related to a working memory function. Our results that
30 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31
working memory loads have no significant effect on an
accuracy of produced time suggest that the prefrontal
cortices might not be the interval timing clock itself,
though it played an important role as a part of neural
networks that were involved in participants attempt to
produce time, since their study did not concern the
accuracy of produced time due to different objectives of
their studies.
Mood had been believed as an important factor in
considering subjective passage of time. As one may
imagine, in ones daily life time is likely to pass faster in
a pleasant situation than in an unpleasant situation.
Mischel et al. and Watts et al. have reported that
subjective passage of time ran slower in the condition in
which an emotionally uncomfortable task was provided,
in comparison with that without such a task (Mischel et
al., 1969; Watts and Sharrock, 1984; Angrilli et al.,
1997). In the present study, we did not find any effects of
mood and other psychological measures on time pro-
duction. A possible explanation for the different results
may be attributed to our study design in which the
behavioral and psychological conditions were strictly
controlled, giving minimal changes in the psychological
statuses across 4 sessions. Moreover, enforced motiva-
tion for accuracy of produced time using economical
rewards did not affect the accuracy, providing another
explanation that produced time may be beyond voli-
tional control.
Acknowledgements
The authors would like to appreciate the assistance of
Kayo Shibui, M.D., Ph.D. and Kyuja Kim, M.D.,
Ph.D. in designing the study. This study was supported
in part by a Research Grant for Nervous and Mental
Disorders (11
/3) and a Health Science Grant
(12080701) from the Ministry of Health, Labor and
Welfare, and a Grant-in-aid for Scientific Research
(13470200) by the Ministry of Education, Science and
Culture.
References
Angrilli, A., Cherubini, P., Pavese, A., Mantredini, S., 1997. The
influence of affective factors on time perception. Percept. Psycho-
phys. 59 (6), 972
/982.
Aschoff, J., Daan, S., 1997. Human time perception in temporal
isolation: effects of illumination intensity. Chronobiol. Int. 14 (6),
585
/596.
Aschoff, J., 1998. Human perception of short and long time intervals:
its correlation with body temperature and the duration of wake
time. J. Biol. Rhythms 13 (5), 437
/442.
Baddeley, A.D., Hitch, G.J., 1974. Working memory. In: Bower, G.H.
(Ed.), The Psychology of Learning and Motivation: Advances in
Research and Theory, vol. 8. Academic Press, New York, pp. 47
/
89.
Baddeley, A.D., 1982. Implications of neuropsychological evidence for
theories of normal memory. Philos. Trans. R. Soc. Lond. B Biol.
Sci. 298 (1089), 59
/72.
Baddeley, A.D., 1986. Working Memory. Oxford University Press,
New York.
Campbell, S.S., Murphy, P.J., Boothroyd, C.E., 2001. Long-term time
estimation is influenced by circadian phase. Physiol. Behav. 72 (4),
589
/593.
Casini, L., Ivry, R.B., 1999. Effects of divided attention on temporal
processing in patients with lesions of the cerebellum or frontal lobe.
Neuropsychology 13 (1), 10
/21.
Clifford, B.S., Susan, I., Richard, F., 2000. Integration of sensory and
motor function: the association areas of the cerebral cortex and the
cognitive capabilities of the brain. In: Eric, R.K., James, H.S.,
Thomas, M.J. (Eds.), Principles of Neural Science, fourth ed..
Principles of Neural Science, pp. 357
/361.
Conti, R., 2001. Time flies: investigating the connection between
intrinsic motivation and the experience of time. J. Pers. 69 (1), 1
/
26.
Czeisler, C.A., Johnson, M.P., Duffy, J.F., Brown, E.N., Ronda, J.M.,
Kronauer, R.E., 1990. Exposure to bright light and darkness to
treat physiologic maladaptation to night work. N. Engl. J. Med.
322, 1253
/1259.
Dantzer, R., 2001. Cytokine-induced sickness behavior: mechanisms
and implications. Ann. N.Y. Acad. Sci. 933, 222
/234.
Fortin, C., Rousseau, R., Bourque, P., Kirouac, E., 1993. Time
estimation and concurrent nontemporal processing: specific inter-
ference from short-term-memory demands. Percept. Psychophys.
53 (5), 536
/548.
Fortin, C., Breton, R., 1995. Temporal interval production and
processing in working memory. Percept. Psychophys. 57 (2),
203
/215.
Fortin, C., Rousseau, R., 1998. Interference from short-term memory
processing on encoding and reproducing brief durations. Psychol.
Res. 61 (4), 269
/276.
Fortin, C., Masse, N., 1999. Order information in short-term memory
and time estimation. Mem. Cognit. 27 (1), 54
/62.
Gibbon, J., Church, R.M., 1981. Journal of experimental psychology:
animal behavior processes. J. Exp. Psychol. Anim. Behav. Process.
7 (2), 87
/107.
Hancock, P.A., 1993. Body temperature influence on time perception.
J. Gen. Psychol. 120 (3), 197
/216.
Harrington, D.L., Haaland, K.Y., Knight, R.T., 1998. Cortical
networks underlying mechanisms of time perception. J. Neurosci.
18 (3), 1085
/1095.
Higuchi, S., Liu, Y., Yuasa, T., Maeda, A., Motohashi, Y., 2001.
Diurnal variations in alpha power density and subjective sleepiness
while performing repeated vigilance tasks. Clin. Neurophysiol. 112
(6), 997
/1000.
Hoagland, H., 1933. The physiological control of judgements of
duration. Evidence of a chemical clock. J. Gen. Psychol. 9, 267
/
287.
Ivry, R.B., 1996. The representation of temporal information in
perception and motor control. Curr. Opin. Neurobiol. 6 (6),
851
/857.
Kubota, T., Uchiyama, M., Suzuki, H., Shibui, K., Kim, K., Tan, X.,
Tagaya, H., Okawa, M., Inoue, S., 2002. Effects of nocturnal
bright light on saliva melatonin, core body temperature and sleep
propensity rhythms in human subjects. Neurosci. Res. 42, 115
/122.
Manoach, D.S., Schlaug, G., Siewert, B., Darby, D.G., Bly, B.M.,
Benfield, A., Edelman, R.R., Warach, S., 1997. Prefrontal cortex
fMRI signal changes are correlated with working memory load.
Neuroreport 8 (2), 545
/549.
Maquet, P., Lejeune, H., Pouthas, V., Bonnet, M., Casini, L., Macar,
F., Timsit-Berthier, M., Vidal, F., Ferrara, A., Degueldre, C.,
Quaglia, L., Delfiore, G., Luxen, A., Woods, R., Mazziotta, J.C.,
 K. Kuriyama et al. / Neuroscience Research 46 (2003) 23
/31
Comar, D., 1984. Brain activation induced by estimation of
duration: a PET study. Neuroimage 3 (2), 119
/126.
McCarthy, G., Blamire, A.M., Puce, A., Nobre, A.C., Bloch, G.,
Hyder, F., Goldman-Rakic, P., Shulman, R.G., 1994. Functional
magnetic resonance imaging of human prefrontal cortex activation
during a spatial working memory task. Proc. Natl. Acad. Sci. 91
(18), 8690
/8694.
Mischel, W., Grusec, J., Masters, J.C., 1969. Effects of expected delay
time on the subjective value of rewards and punishments. J. Pers.
Soc. Psychol. 11 (4), 363
/373.
Moore-Ede, M.C., Czeisler, C.A., Richardson, G.S., 1983. Circadian
time-keeping in health and disease. I. Basic properties of circadian
pacemakers. N. Engl. J. Med. 309, 469
/476.
Morofushi, M., Shinohara, K., Kimura, F., 2001. Menstrual and
circadian variations in time perception in healthy women and
women with premenstrual syndrome. Neurosci. Res. 41 (4), 339
/
344.
Niedermeyer, E., Krauss, G.L., Peyser, C.E., 1989. The electroence-
phalogram and mental activation. Clin. Electroencephalogr. 20 (4),
215
/227.
Owens, T., Babcock, A., 2002. Immune response induction in the
central nervous system. Front. Biosci. 1 (7), 427
/438.
Phillips, J.R., 1977. Relationship of field-dependence */independence
to posture and judgment of time duration. Percept. Mot. Skills 44
(3 Pt. 1), 931
/940.
Pouthas, V., Maquet, P., Garnero, L., Ferrandez, A.M., Renault, B.,
1999. Neural bases of time estimation: a PET and ERP study.
Electroencephalogr. Clin. Neurophysiol. 50 (Suppl.), 598
/603.
Rammsayer, T.H., 1989. Dopaminergic and serotoninergic influence
on duration discrimination and vigilance. Pharmacopsychiatry 22
(Suppl. 1), 39
/43.
31
Rammsayer, T.H., Vogel, W.H., 1992. Pharmacologic properties of
the internal clock underlying time perception in humans. Neurop-
sychobiology 26 (1
/2), 71
/80.
Rammsayer, T.H., Hennig, J., Haag, A., Lange, N., 2001. Effects of
noradrenergic activity on temporal information processing in
humans. Q. J. Exp. Psychol. B 54 (3), 247
/258.
Roberts, S., Holder, M.D., 1984. What starts an internal clock? J. Exp.
Psychol. Anim. Behav. Process. 10 (3), 273
/296.
Saletu, B., Grunberger, J., 1984. Early clinical pharmacological trials
with a new anti-epileptic, milacemide, using pharmaco-EEG and
psychometry. Methods Find Exp. Clin. Pharmacol. 6 (6), 317
/330.
Schubotz, R.I., Friederici, A.D., von Cramon, D.Y., 2000. Time
perception and motor timing: a common cortical and subcortical
basis revealed by fMRI. Neuroimage 11 (1), 1
/12.
Treisman, M., 1963. Temporal discrimination and the indifference
interval. Implications for a model of the internal clock. Psychol.
Monogr. 77 (13), 1
/31.
Tagaya, H., Uchiyama, M., Shibui, K., Kim, K., Suzuki, H., Kamei,
Y., Okawa, M., 2002. Non-rapid-eye-movement sleep propensity
after sleep deprivation in human subjects. Neurosci. Lett. 323, 17
/
20.
Uchiyama, M., Mayer, G., Okawa, M., Meier-Ewert, K., 1995. Effects
of vitamin B12 on human circadian body temperature rhythm.
Neurosci. Lett. 192, 1
/4.
Virginia, M., 1996. Setting a biological stopwatch. Science 271, 905
/
906.
Watts, F.N., Sharrock, R., 1984. Fear and time estimation. Percept.
Mot. Skills 59 (2), 597
/598.
Zakay, D., 1992. The role of attention in childrens time perception. J.
Exp. Child. Psychol. 54 (3), 355
/371.