Having problems with message search?
Journal of Physiological Anthropology
Physiological Effects of Shinrin-yoku
(Taking in the Atmosphere of the Forest) in an
Yuko Tsuntesugu
Hideki Ishii
Hideki Hirano
Takahide Kagawa
And Yoshifumi Miyazaki
This article was presented at the 8th
International Congress of
Physiological Anthropology, 2006 (ICPA
2006), in
Received: September 30, 2006
Accepted: December 22, 2006
Correspondence to: Yuko Tsunetsugu, 1
Matsunosato,
Tsukuba,
Phone: [1]83–29–829–8310
Fax: [1]83–29–874–3720
e-mail: yukot@...
142 Physiological Effects of Shinrin-yoku
From:
Sent: Friday, October 24, 2008
6:38 AM
To: '
Cc:
Subject: FW: Forests and health.
No child left inside.
Abstract The physiological effects of
“Shinrin-yoku” (taking
in the atmosphere of the forest) were
examined by
investigating blood pressure, pulse rate,
heart rate variability
(HRV), salivary cortisol concentration,
and immunoglobulin A
concentration in saliva. Subjective
feelings of being
“comfortable”,
“calm”, and “refreshed” were also assessed by
questionnaire. The subjects were 12 male
university students
aged from 21 to 23 (meanSD: 22.01.0). The
physiological
measurements were conducted six times,
i.e., in the morning
and evening before meals at the place of
accommodation,
before and after the subjects walked a
predetermined course in
the forest and city areas for 15 minutes,
and before and after
they sat still on a chair watching the
scenery in the respective
areas for 15 minutes. The findings were as
follows. In the
forest area compared to the city area, 1)
blood pressure and
pulse rate were significantly lower, and
2) the power of the HF
component of the HRV tended to be higher
and LF/(LF[1]HF)
tended to be lower. Also, 3) salivary
cortisol concentration was
significantly lower in the forest area.
These physiological
responses suggest that sympathetic nervous
activity was
suppressed and parasympathetic nervous
activity was enhanced
in the forest area, and that
“Shinrin-yoku” reduced stress
levels. In the subjective evaluation, 4)
“comfortable”, “calm”,
and “refreshed” feelings were
significantly higher in the forest
area. The present study has, by conducting
physiological
investigations with subjective evaluations
as supporting
evidence, demonstrated the relaxing and
stress-relieving effects
of “Shinrin-yoku”. J Physiol
Anthropol 26(2): 135–142, 2007
http://www.jstage.jst.go.jp/browse/jpa2
[DOI: 10.2114/jpa2.26.135]
Keywords: therapeutic effect of forest,
natural environment,
heart rate variability, salivary cortisol,
relaxation
Introduction
Stress control is one of the most
important issues
confronting modern society. As reflected
by the word
“Technostress”, coined by Brod
in 1984, modern society is
becoming more complicated and highly
industrialized,
consequently causing many stress-related
disorders.
In “stressful” daily life, it
is a common experience that
contact with the natural environment or
natural objects
provides a feeling of relaxation or a
release from tension. A
study conducted by Lohr et al. (1996) is
considered to give
strong support to this contention. They
clarified that plants in a
windowless office environment contributed
to an improvement
in worker productivity. The authors have
tried to examine the
pleasant feelings that natural objects
induce in humans by
conducting physiological investigations.
From studies dealing
with the five senses separately, we have
clarified that the smell
of Japanese cedar wood lowered blood
pressure and regional
cerebral blood flow in the prefrontal area
(Miyazaki et al.,
1999), and the sound of murmuring water
lowered blood
pressure (Mishima et al., 2004). As
Frumkin (2001) showed
that a deep-seated connection between the
natural world and
humans was unsurprising from an
evolutionary perspective, we
assume that human physiological functions
have had to adapt
to the natural environment; thus, it is
somehow a stressor to
live in modern “artificial”
society. It should be natural for
people, having this background, to feel a
sense of comfort or
affinity with the natural environment.
Frumkin (2001) also pointed out that
certain kinds of
contact with the natural world could enhance
human health.
Animals, plants, landscapes, and
wilderness experience were
cited as components of the natural world
that can function to
enhance health. As one example of the
clinical trials
conducted, Urlich (1984) did a remarkable
study. He examined
Physiological Effects of Shinrin-yoku
(Taking in the Atmosphere of
the Forest) in an Old-Growth Broadleaf
in
Yuko Tsunetsugu1), Bum-Jin Park1), Hideki
Ishii2), Hideki Hirano3),
Takahide Kagawa1) and Yoshifumi Miyazaki1)
1) Forestry and Forest Products Research
Institute
2)
3) Ministry of the Environment
records on the recovery of patients in a
hospital for 10 years
and found that patients with tree views
had shorter
hospitalizations compared to patients with
brick-wall views. In
the field of environmental health, the
natural environment is
now seen as one of the factors which have
an impact on human
health, not in the traditional context of
causing harm by
exposure to environmental toxins, but from
the viewpoint of
potentially enhancing our well-being
through daily exposure to
the natural environment (Frumkin et al.,
2002). Through these
previous studies, it has been confirmed
that the natural
environment has a definite beneficial
effect on humans.
“Shinrin-yoku” is a word
coined by the Forestry Agency of the
Japanese government in 1982 to encourage
utilization of
national forests for enhancement of
physical and mental health.
It is a compound word made up of two
independent words
meaning “forest” and
“bathing”. Like sea bathing, to be in the
forest environment and take in the
atmosphere of the forest in
expectation of a potential curative or
therapeutic effect is
probably what the word
“Shinrin-yoku” intends to convey.
Thus, from this perspective, the effect of
“Shinrin-yoku”
should be considered as non-specific.
Though it would be reasonable to expect
that “Shinrinyoku”
has beneficial effects on human
physiology, there have
only been a few studies that have
attempted to prove this
through on-site experiments. Ohtsuka et
al. (1998) found that
the mean blood glucose level of diabetic
patients significantly
decreased after “Shinrin-yoku”
(walking 3 km to 6 km in the
forest). Ohira et al. (1999) examined the
immunological and
endocrine indices, EEG, and ECG of twenty
undergraduates in
a forest environment and in a non-forest
environment. Their
main finding was that NK cell activity and
immunoglobulin A,
G, and M were significantly increased
after staying 8 hours in a
forest environment. There were no
significant differences
between the forest and non-forest
environments in terms of
their effects on the other physiological
parameters or
psychological states (anxiety, mood, and
subjective stress).
They pointed out the necessity of further
studies, adding that
the bad weather and low temperature on the
day of their
experiment might have reduced the
pleasantness of “Shinrinyoku”.
The previous studies dealt with the
effects that occurred over
a period of one day. However, in our
previous studies, as
mentioned before (Miyazaki et al., 1999;
Mishima et al.,
2004), changes in physiological parameters
by the inhalation of
wood odor or the sound of murmuring water were
observed
within 60 to 90 seconds. Though these were
the results of
laboratory experiments, it can be expected
that the impacts of
“Shinrin-yoku” on
physiological response would be obtained
in a shorter time.
The aim of the present study was to clarify
the effect of
“Shinrin-yoku” as a daily
activity on healthy subjects. From
the viewpoint of preventive medicine or
health maintenance,
“Shinrin-yoku” should be
accessible and easy to implement.
Thus it was our intention to assess a
program that included
activities of short duration. The
evaluation was attempted by
measuring various parameters of autonomic
nervous activity,
the endocrine system, and the
immunological system.
Subjective feeling was also assessed using
a questionnaire. The
experiment reported in the present paper
was conducted as part
of a large-scale ongoing investigation on
“Shinrin-yoku” in
Materials and Methods
The experiment was conducted in a
deciduous broadleaf
forest mainly consisting of old-growth
beech in Nukumidaira
(Oguni,
For comparison, an area around Niigata
Station (
map in Figure 1. Hereinafter, the two
sites are referred to as the
forest area and the city area (Fig. 2).
The weather was fine in
both areas and the average temperature and
relative humidity
was 23.5°C and 73.4% in the forest area
and 26.4°C and
62.1% in the city area, respectively.
The subjects were 12 male university
students aged from 21
to 23 (meanSD: 22.01.0). The subjects were
assembled in
the afternoon on the day before the
experiment. Sufficient
information on the aim and process of the
experiment was
provided and written informed consent was obtained.
The
study was performed under the regulations
of the Institutional
Review Committee of the Forest and
previewed the experimental sites in the
forest and the city
areas. Then a measurement practice was
conducted at a place
of accommodation, which was located at
approximately the
same distance (about 60 minutes by car)
from both
136 Physiological Effects of Shinrin-yoku
Fig. 1 Location of the experimental sites.
experimental sites. Each subject stayed in
a single room in the
hotel and all subjects had the same meals
until the end of the
experiment.
The subjects were divided randomly into
two groups
consisting of six people each. On the
first day of the
experiment, one group was sent to the
forest area, and the other
was sent to the city area. On the second
day, each group went
to the other area to eliminate the order
effect.
In the morning, each group was taken to
the experimental
site and rested for a while in a nearby
resting room. In the
forenoon, the subjects walked a
predetermined course in each
area at an unhurried pace for 15 minutes.
In the afternoon,
after taking lunch in the resting room,
they sat on chairs
watching the scenery in each area for 15
minutes. The subjects
engaged in the walking and watching
individually.
Physiological measurements were conducted
six times a
day: (i) in the morning at the place of
accommodation before
breakfast (06:15–07:15); (ii) before
walking (10:40–11:30);
(iii) after walking (11:00–11:50);
(iv) before watching (14:10–
15:00); (v) after watching
(14:30–15:20); (vi) in the evening at
the place of accommodation before dinner
(18:00–19:00). The
R-R interval was measured continuously
during the walking
and watching in addition to these six
specific measurement
times.
The measured physiological parameters were
the R-R
interval of the electrocardiogram to
analyze heart rate
variability (HRV), systolic and diastolic
blood pressure, pulse
rate, salivary cortisol concentration, and
secretory
immunoglobulin A (s-IgA) concentration in
saliva. R-R
intervals were obtained by an ambulatory
electrocardiogram
monitor (Active Tracer AC301A, GMS
Corporation). They
were measured over 2 minutes during the resting
state with
eyes closed in the morning and the
evening, and before and
after walking and watching. R-R intervals
were also taken
continuously for 15 minutes during walking
and watching.
Systolic and diastolic blood pressure, and
pulse rate were
measured by a digital blood pressure
monitor using
oscillometric methods (HEM1000, Omron) on
the right upper
arm. Saliva for analysis of cortisol and
s-IgA was collected
using Salivette devices over a two-minute
period. The saliva
samples were frozen after collection and
subsequently
analyzed by SRL, Inc. In addition to the
physiological
measurements, subjective
“comfortable-uncomfortable” and
“calm-roused” feelings were
estimated using a 13-point scale.
“Refreshed” feeling was measured
with the Stress–Refresh
feeling test (Mackay et al., 1978). These
subjective tests using
the questionnaire were also carried out
six times a day at the
time of the physiological measurements.
R–R interval data were analyzed by
the maximum entropy
method (Memcalc, GMS, Ohtomo et al.,
1994). The power of
the low-frequency (LF; 0.04–0.15 Hz)
component and the
high-frequency (HF; 0.15–0.4 Hz)
component of the obtained
heart rate power spectrum were calculated
for each minute. It
is considered that HF reflects
parasympathetic nervous activity
and LF/(LF[1]HF) reflects sympathetic
nervous activity. A
paired t-test was used to compare the
physiological data
between the forest area and the city area.
The Wilcoxon signed
rank test was used to compare the data on
subjective feelings.
All statistical analysis was performed
using StatView version
5.0 (SAS Institute Inc). p
0.05 was considered to be
significant.
Results
Subjective “comfortable”
feelings in the forest area and the
city area at each measurement time are
shown in Figure 3. The
forest area was evaluated as producing
significantly more
comfortable feelings before and after
walking (p
0.05) and
after being seated and watching (p
0.01) than the city area
did. By comparing before and after
watching, it was clarified
that watching scenery in the forest area
significantly increased
feelings of comfort (p
0.05) while activities in the city area
significantly decreased feelings of
comfort (p
0.05 when
walking, p
0.01 when watching). Figure 4 shows the
results
for a “calm” feeling in the
forest and city areas. The forest area
was rated as producing significantly
calmer feelings after
walking and watching than the city area
did (p
0.05, p
0.01,
Tsunetsugu, Y et al. J Physiol Anthropol,
26: 135–142, 2007 137
Fig. 2 The scenery in the two experimental
sites.
upper:
respectively). As in the case of a
“comfortable” feeling,
watching in the forest area significantly
enhanced the feeling
of calm (p
0.05), whereas the activities in the city
environment lowered this feeling (p
0.01 when walking,
p
0.05 when watching). Figure 5 shows the
scores for a
“refreshed” feeling as gauged
by the Stress–Refresh test. The
score was significantly higher in the
forest area than in the city
area before and after walking and before
and after watching
(p
0.05 before and after walking, and before
watching,
p
0.01 after watching). Significant
differences between before
and after the activities were only
observed for the city area,
i.e., the city area caused a significant
decrease in feeling
refreshed (p
0.05).
Figure 6 shows the mean value of systolic
blood pressure in
each experimental site at each measurement
time. Systolic
blood pressure was significantly lower in
the forest area before
walking (p
0.05) and before (p
0.01) and after (p
0.05)
watching than in the city area. Figure 7
shows diastolic blood
pressure in the same sequence as systolic blood
pressure.
Diastolic blood pressures showed
significantly lower values in
the forest area before walking (p
0.05) and after watching
(p
0.01). In a comparison between before and
after, diastolic
blood pressure decreased following walking
in the city area
(p
0.05) and watching scenery in the forest
area (p
0.05).
Figure 8 shows the change in the pulse
rate. There was an
overall tendency for the pulse rate to be
lower in the forest area
138 Physiological Effects of Shinrin-yoku
Fig. 3 Subjective comfortable feeling
measured by questionnaire at six
measurement times in the forest area and
in the city area.
N
9 at “Before walking” and
“After walking”, N
11 at the other
times. *: p
0.05, **: p
0.01 by Wilcoxon signed rank test.
Fig. 4 Subjective calm feeling measured by
questionnaire at six
measurement times in the forest area and
in the city area.
N
9 at “Before walking” and
“After walking”, N
11 at the other
times. *: p
0.05, **: p
0.01 by Wilcoxon signed rank test.
Fig. 5 Subjective refreshed feeling
measured by the Stress–Refresh
feeling test at six measurement times in
the forest area and in the city
area.
N
9 at “Before walking” and
“After walking”, N
11 at the other
times. *: p
0.05, **: p
0.01 by Wilcoxon signed rank test.
Fig. 6 Changes in systolic blood pressure
determined at six
measurement times in the forest area and
in the city area.
N
9 at “Before walking” and
“After walking”, N
11 at the other
times. *: p
0.05, **: p
0.01 by paired t-test.
Fig. 7 Changes in diastolic blood pressure
determined at six
measurement times in the forest area and
in the city area.
N
9 at “Before walking” and
“After walking”, N
11 at the other
times. *: p
0.05, **: p
0.01 by paired t-test.
than in the city area except in the
morning, and a significant
difference between the two sites was
observed before walking.
Though it did not reach a significant
level, the p-value was less
than 0.06 after walking and in the
evening. Pulse rate
significantly decreased as a result of
watching scenery in the
city area (p
0.01).
To assess the activity in the autonomic
nervous system
separately from sympathetic nervous
activity and
parasympathetic nervous activity, we
investigated HRV by
means of frequency analysis of the
R–R interval. Figure 9
shows the minute-by-minute change in the
power of the HF
component of HRV over time. The data
sequence is as follows:
morning (at the place of accommodation),
before, during and
after walking, before, during and after
watching, and evening
(at the place of accommodation). The HF
power tended to be
higher in the forest area overall during
walking. A significant
difference between the forest area and the
city area was
observed at the 1st (p
0.01), 9th (p
0.05), and 15th
(p
0.05) minute during watching. Though the
HF power
tended to be higher in the forest area
than in the city area
before and after walking and watching, it
did not reach a
significant difference (p-value was less
than 0.06 at the 1st
minute in the measurement before walking
and at the 2nd
minute after watching). Figure 10 shows
the variation in
LF/(LF[1]HF) over time. The value in the
forest area was
significantly lower at the 2nd (p
0.05) and 11th (p
0.05)
minute during walking and the 1st (p
0.05) and 9th (p
0.05)
minute during watching.
Tsunetsugu, Y et al. J Physiol Anthropol,
26: 135–142, 2007 139
Fig. 8 Changes in pulse determined at six
measurement times in the
forest area and in the city area.
N
9 at “Before walking” and
“After walking”, N
11 at the other
times. *: p
0.05, **: p
0.01 by paired t-test.
Fig. 9 Time-series change in the power of
the HF component of HRV in the forest area and in the city area.
N
5–12, *: p
0.05, **: p
0.01 by paired t-test.
Fig. 10 Time-series change in LF/(LF[1]HF)
of HRV in the forest area and in the city area.
N
5–12, *: p
0.05 by paired t-test.
Figure 11 shows the time course change in
the salivary
cortisol concentration. As was already
known, diurnal
variations where concentration is high in
the morning and
tends to decrease as time progresses were
observed in both
areas. Cortisol concentration was lower in
the forest area than
in the city area at all measurement times,
and there was a
significant difference between the two
sites before and after
walking, and after watching (p
0.05). The p-value was less
than 0.06 before watching. The present
study found no
significant difference in the s-IgA
concentration. No significant
diurnal variation was found in the s-IgA
concentration.
Discussion
The results for the forest area and the
city area are
summarized in Table 1. No indices showed
significant
differences between the forest area and
the city area for the
measurements taken in the morning and the
evening. The
differences were mainly observed before
walking and after
being seated and watching the scenery,
followed by after
walking, then before watching. All of the
indices except s-IgA
were generally in excellent agreement with
each other.
The results for the subjective evaluations
showed that
walking around or watching the scenery in
the forest area
created a feeling of comfort and calm.
Feeling refreshed was
higher in the forest area both when
walking and watching, and
both before and after the activities,
which can be interpreted as
indicating that the forest environment
itself induced the
refreshed feeling regardless of the kind
of activity and the
measurement time.
The findings for blood pressure and pulse
rate demonstrated
that activities in the forest area caused
a relaxed physiological
state compared to the city area. The
significant differences in
all of the three parameters (systolic and
diastolic blood
pressure, and pulse rate) before walking
suggest that the two
environments had already had different
impacts on physiology
before activities commenced. The
temperature in the period
10:00–12:00 was lower in the forest
area (about 22–24°C in
the forest and 26–27°C in the city),
and the forest area was felt
as significantly more
“comfortable” and “refreshing” before
walking. The whole environment in the
forest area was
considered to induce lower blood pressure
and pulse rate. This
might also be affected by the fact that
the subjects had stayed
in the forest environment for
1.5–2.5 hours from 9:00, which
corresponded to the measurement time of
“before walking”.
The reason for no significant differences
after walking seemed
to be that blood pressure and pulse rate
tended to decrease (a
significant decrease in diastolic blood
pressure) after walking
in the city area whereas they were rather
stable in the forest
area. Since pulse rate did not increase,
the 15-minute walking
conducted in the present study was assumed
to be light
exercise for these subjects. It has been
reported that moderate
140 Physiological Effects of Shinrin-yoku
Fig. 11 Changes in salivary cortisol
concentration at six measurement
times in the forest area and in the city
area.
N
9 at “Before walking” and
“After walking”, N
11 at the other
times. *: p
0.05 by paired t-test.
Table 1 Summary of the comparison between
the forest and city areas.
Physiological responses
Subjective feelings
Autonomic nervous system
Endocrine Immune
system system
Sys Dias Pulse
HF
LF/
BP BP Rate (LF[1]HF)
Cortisol s-IgA Comfort Calm Refreshed
Morning
Pre-walk F
C* F
C* F
C* F
C* F
C* FC* FC*
during
Post- walk walk F
C* FC* FC* FC*
Pre-watch F
C** FC* F
C* FC*
during during
Post-watch F
C* F
C** watch watch F
C* FC** FC* FC**
Evening
F:
differences were observed in the index at
the 1% significance level.
exercise decreases blood pressure compared
to the level before
exercise (Halliwill, 2001). The
significant decrease in diastolic
blood pressure in the city area was hence
considered to be the
effect of exercise. We assumed that diastolic
blood pressure
was lowered already at the measurement
time of “before
walking” in the forest area, so it
did not decrease as a result of
walking. The significantly lower blood
pressure after being
seated and watching the scenery in the
forest area showed that
the activity had the effect of reducing
stress. The fact that the
pulse rate of the subjects who returned
from the forest in the
evening tended to be lower (p
0.06) might indicate that the
effect of “Shinrin-yoku”
lasted for a certain amount of time.
The results for HRV showed that
parasympathetic nervous
activity tended to be dominant in the
forest area, which implies
that “Shinrin-yoku” had a
relaxing effect. A rapid decrease in
HF when doing submaximal exercise
(Yamamoto et al., 1991)
or ergometer exercise until exhaustion
(Tabusadani et al.,
2001) has been reported in previous
studies, but the decrease
in HF during walking in the city area in
the present study was
not considered to be caused only by the
exercise, since the
quantity of physical activity during
walking calculated from
the acceleration of the subject was at the
same level in each
area, and pulse rate did not show an
increase after walking (the
exercise load was not as high as in
previous studies). Thus the
decrease in HF components in the city area
was also
considered to relate to the perceived
mental stress (Dishman et
al., 2000; Hjortskov et al., 2004). From
this point of view, the
relatively high HF in the forest area
indicated a connection
with the subjective
“comfortable” and “calm” feelings.
Recently, it was reported that soothing
music caused a higher
HF component than stimulating music did
(Iwanaga et al.,
2005), and that an increase in HF power
was observed during
Zen meditation (Murata et al., 2004). The
outcomes in the
present study are supportive of the
results from those previous
studies. The effect of the environment was
more significant
during watching. As the subjects sat still
in a chair in both
areas, the results during watching can be
inferred to have a
close association with the mood state.
Parasympathetic
nervous activity was more dominant than in
the period of
walking in both areas, and moreover, it
was significantly more
dominant in the forest area than in the
city area. The standard
deviation of HF power was larger in the
forest area, possibly
because walking around or watching scenery
in the city area
caused a stressed, heightened state
equally in all subjects,
whereas the relaxing and calming effect of
“Shinrin-yoku” was
more variable and depended to a certain
extent on the
individual. Rosenwinkel et al. (2001)
pointed out that HRV
may be more applicable to assessing
parasympathetic nervous
activity than for assessing sympathetic
function. In the present
study, the difference between the forest
and city areas was not
as clear in LF/(LF[1]HF), but it can be
supposed that
sympathetic nervous activity was more
dominant in the city
area during walking and watching.
From the investigations of many previous
studies
(Kirschbaum and Hellhammer, 1989;
Ockenfels et al., 1995),
the result that cortisol concentration was
significantly lower in
the forest area before and after walking,
and after watching can
be interpreted as clearly demonstrating
the relaxation effect of
“Shinrin-yoku”. O’Connor
and Corrigan (1987) reported that a
significant increase in salivary cortisol
was elicited by
submaximal exercise, while Jin (1989)
found that moderate
exercise (practice of Tai Chi) decreased salivary
cortisol levels.
The significant decrease after walking in
the city area in the
present study might be attributable to the
effect of the exercise.
The reason why there was no significant
tendency toward an
increase or decrease in the IgA concentration
in saliva in the
present study is unclear, and further
consideration from the
viewpoint of individual variation, e.g., a
connection with
personality (Ohira et al., 1999), or a
relationship between kinds
of stress and the stress reaction mechanism
(Fujiwara and
Yokoyama, 1990) is needed.
The responses in the parameter reflecting
autonomic
Send Email