Content uploaded by Gavin R H Sandercock
Author content
All content in this area was uploaded by Gavin R H Sandercock on Feb 10, 2014
Content may be subject to copyright.
Eur J Appl Physiol
DOI 10.1007/s00421-012-2318-8
123
ORIGINAL ARTICLE
The eVects of views of nature on autonomic control
V. F. Gladwell · D. K. Brown · J. L. Barton ·
M. P. Tarvainen · P. Kuoppa · J. Pretty ·
J. M. Suddaby · G. R. H. Sandercock
Received: 20 July 2011 / Accepted: 7 January 2012
© Springer-Verlag 2012
Abstract Previously studies have shown that nature
improves mood and self-esteem and reduces blood pres-
sure. Walking within a natural environment has been sug-
gested to alter autonomic nervous system control, but the
mechanisms are not fully understood. Heart rate variability
(HRV) is a non-invasive method of assessing autonomic
control and can give an insight into vagal modulation. Our
hypothesis was that viewing nature alone within a con-
trolled laboratory environment would induce higher levels
of HRV as compared to built scenes. Heart rate (HR) and
blood pressure (BP) were measured during viewing diVer-
ent scenes in a controlled environment. HRV was used to
investigate alterations in autonomic activity, speciWcally
parasympathetic activity. Each participant lay in the semi-
supine position in a laboratory while we recorded 5 min
(n= 29) of ECG, BP and respiration as they viewed two
collections of slides (one containing nature views and the
other built scenes). During viewing of nature, markers of
parasympathetic activity were increased in both studies.
Root mean squared of successive diVerences increased
4.2 §7.7 ms (t= 2.9, p= 0.008) and natural logarithm of
high frequency increased 0.19 §0.36 ms2Hz¡1 (t=2.9,
p= 0.007) as compared to built scenes. Mean HR and BP
were not signiWcantly altered. This study provides evidence
that autonomic control of the heart is altered by the simple
act of just viewing natural scenes with an increase in vagal
activity.
Keywords Environment · Nature · Cardiovascular ·
Autonomic control · Vagal activity
Introduction
Nature has wide ranging positive eVects, but the mecha-
nisms of these eVects are not understood, particularly at a
physiological level. Cohort study data show that viewing
natural landscapes has positive beneWts including:
improved general health perception (Moore 1982); reduced
need for pain relief (Ulrich 1984; Diette et al. 2003; Lechtzin
et al. 2010); improved concentration and attention (Berto
2005); improved cognition (Berman et al. 2008); and
improved self-esteem and mood (Pretty et al. 2007; Barton
et al. 2009).
Meta-analyses (Barton and Pretty 2010) and systematic
reviews (Bowler et al. 2010; Thompson Coon et al. 2011)
demonstrate the eYcacy of exposure to nature in improving
psychological well-being, but there is a paucity of studies
examining physiological eVects (Bowler et al. 2010), possi-
bly due to the diYculties of recording high-quality physio-
logical data outdoors. In addition, to date laboratory studies
of viewing natural scenes prove inconsistent (Bowler et al.
2010). When participants viewed slides of rural or built
scenes there were no signiWcant diVerences in heart rate
(HR) responses (Ulrich 1981). When exposed to a stressor
(elevating HR and blood pressure (BP)) prior to viewing
videos of diVerent environments, natural views were
deemed more ‘restorative’ because they elicited more rapid
Communicated by Susan A. Ward.
V. F. Gladwell (&) · D. K. Brown · J. L. Barton · J. Pretty ·
J. M. Suddaby · G. R. H. Sandercock
Department of Biological Sciences, Centre for Sport and Exercise
Science, University of Essex, Wivenhoe Park,
Colchester CO4 3SQ, UK
e-mail: vglad@essex.ac.uk
M. P. Tarvainen · P. Kuoppa
Department of Applied Physics, University of Eastern Finland,
P.O. Box 1627, 70211 Kuopio, Finland
Eur J Appl Physiol
123
returns to baseline HR (Ulrich et al. 1991; Laumann et al.
2003) and BP (Chang et al. 2008). When treadmill exercise
was incorporated while viewing slides, BP was lower after
a 5-min recovery period after viewing slides depicting natu-
ral scenes when compared with slides depicting built envi-
ronments (Pretty et al. 2005).
To try and understand the physiological eVects of nature,
one area of interest is the control of the autonomic nervous
system (ANS). The ANS is important in the maintenance of
homeostasis and also in normal and stress–responsive phys-
iology. One way of investigating ANS control is studying
heart rate variability (HRV). HRV is a well-established
non-invasive tool which gives an indication of the changes
in vagal and sympathetic control of the heart.
Walking or sitting in a natural (forest) environment has
previously been shown to lower HR and BP when com-
pared with a built environment control (Park et al. 2010).
Park and colleagues suggest that an increase in vagally
mediated HRV with simultaneous decreases in sympathetic
components is responsible for the observed reductions in
HR and BP. Another study with similar engagement with
natural environments showed that BP was reduced with
a trend to reduced urinary noradrenaline inferring that
this was driven by a decrease in sympathetic stimulation
(Li et al. 2011).
However, in the outdoor environment, it is diYcult to
control external factors including the weather. In addition,
other elements, including smells and sounds may have pos-
itive or negative inXuences on the ANS. To date, there are
only a handful of studies that have explored ANS (Park
et al. 2010; Li et al. 2011) and these have not been able to
elucidate whether the act of just viewing nature can alter
ANS function. Furthermore, these previous studies have
also included exercise which is likely to alter the physiolog-
ical responses.
This is the Wrst study to explore the underlying physio-
logical mechanisms of nature by isolating viewing
nature in a controlled environment and comparing the ANS
responses to viewing built environments. The use of a con-
trolled environment also allowed the recording of Wnger BP
continually, as well as respiration, both of which can inXu-
ence HRV measures.
Our hypothesis was that the simple act of viewing nature
would alter ANS control with vagal measures of HRV
enhanced during viewing nature compared with built views.
Methods
Following ethical approval from the University Ethics
Committee, 35 (22 females) volunteers (comprising staV
and students from the University (mean (SD): age 39.7
(12.1) years) were recruited following an electronic adver-
tisement. Six individuals were excluded due to taking med-
ication which interfered with HR (n= 1), irregular heart
rhythms (n= 4) or severe obesity prohibiting valid readings
(n= 1). All remaining participants (n= 29) were free from
known disease. Participants attended the University labora-
tory on one occasion.
All participants provided informed consent and com-
pleted a health questionnaire (PAR-Q). All testing proce-
dures were carried out between 09:00 h and 14:00 h in a
quiet room with a constant temperature of 22–23°C to stan-
dardise for potential eVects of time of day and temperature,
as recommended when conducting autonomic experiments
(Tukek et al. 2003). Participants were asked to abstain from
food for 2 h and caVeine for 12 h prior to the start of their
tests and not to undertake strenuous physical activity in the
previous 24 h, as such activities may inXuence autonomic
regulation (Sidery and Macdonald 1994; Stubbs and Mac-
donald 1995). A diary was kept to ensure compliance.
ECG (modiWed Lead II conWguration) and continuous
BP (Portapres, FMS, Finapres Medical Systems BV, The
Netherlands) were measured. In addition, respiratory rate
and depth were recorded using a respiratory belt transducer
placed around the lower part of the chest. This strap con-
tains a piezo-electric device that responds linearly to
changes in length induced by chest movement due to
breathing. Breathing rate and depth over each 5-min seg-
ment were analysed to ensure that there were no signiWcant
alterations in these parameters.
Participants rested in the semi-supine position to allow
their HR and BP to stabilise and remained in this position
for the rest of the experiment. All data were sampled at
1,000 Hz and collected by a Powerlab 8SP (Model ML785,
ADInstruments, UK), using Chart 4 software (ADInstru-
ments, UK).
Testing commenced after 15 min of rest to ensure stabi-
lisation of HR. Participants were shown two collections of
slides during the same session. One slideshow contained
natural scenes and the second set incorporated built or
urban scenes lacking greenery (Fig. 1). Participants were
asked to imagine they were in the environment. Slides were
projected on to a screen (1.8 m £1.8 m) situated in front of
the participant.
Half of the participants viewed the natural environment
set of slides Wrst followed by the built environment slides in
a randomised crossover design. The other participants
viewed the built slides Wrst. Ten minutes between the two
sets was allowed, with participants remaining quiet and still
in a semi-supine position whilst looking at a blank screen.
Each slideshow (18 slides) lasted for 5 min with each slide
shown for 17 s. Slides within a slideshow were always
shown in the same order. HR, BP and respiration measure-
ments were recorded for the whole of the 5 min whilst
viewing the slideshow.
Eur J Appl Physiol
123
Data analysis: heart rate variability
ECG data were analysed using Kubios HRV software
(Niskanen et al. 2004) (http://www.kubios.uku.fi). Data for
each set of slides were analysed and averaged over the
5 min periods of viewing slides of either natural or built
environments. No aberrant or ectopic beats were identiWed.
RR intervals were then extracted from the ECG signal and
re-sampled at 4 Hz using cubic spline interpolation to pro-
vide equidistant time points. In the time domain, the mean
R–R interval and HR, standard deviation of RR intervals
(SDRR) and root mean square of successive diVerences
(rMSSD) were calculated as recommended (Task Force
1996). Data then underwent Fast Fourier transformation
(non-parametric) using Welch’s periodogram method. Data
were split into windows of a width of 256 s with an overlap
of 50%. The power spectrum was obtained by averaging
the spectra within these windows. Two spectral compo-
nents of the recording were analysed: low frequency (LF,
0.04–0.15 Hz) and high frequency (HF 0.15–0.40 Hz) spec-
tral power, in accordance with international guidelines
(Task Force 1996). HF provides an indication of parasym-
pathetic activity, whereas LF oscillations result from the
combined activity of both autonomic nervous system
branches (Task Force 1996).
Non-linear analysis was performed using Poincaré plot
analysis, a graphical representation of the correlation
between successive RR intervals with SD1 indicating short-
term variability (analogous to rMSSD and HF) and SD2
indicating overall variability (analogous to SDRR) (Bren-
nan et al. 2001).
Data analysis: blood pressure and baroreceptor sensitivity
Combined measurements of HRV and BP variability give
information on both parasympathetic and sympathetic ner-
vous system activity. Systolic, diastolic and mean BP val-
ues were detected from the measured BP signal for each
heart beat. Systolic BP (SBP) and RR time series were used
in baroreceptor sensitivity (BRS) estimation (the BP value
was as compared to the following RR interval). Both were
Wrst interpolated at 4 Hz and de-trended with smoothing
prior method. BRS values were estimated using two meth-
ods: multivariate autoregressive (AR) spectral estimation
Fig. 1 An example of the slides taken for the built slideshow (a, b) and from the natural slideshow (c, d). There were 18 slides in each slide show
with each slide shown for 17 s
Eur J Appl Physiol
123
method and sequence analysis. Power spectra were Wrst cal-
culated by Wtting a multivariate AR model of order 22 to
RR and SBP time series. AR coeYcients were then used to
calculate power, coherence and phase spectra (Di Rienzo
et al. 2001). BRS estimates were calculated from the spec-
tra using the frequency–domain alpha technique. Values
were calculated for LF (0.04–0.15) Hz and HF (0.15–
0.4 Hz). Spectrum values were calculated in 301 points
between 0 and 2 Hz and only where coherence was higher
than 0.5 and phase was below 0 were accepted to the sum
of spectrum power.
The sequence method was also used to estimate BRS
values. Sequences were detected from the original signals
where RR interval and SBP value both ascended or when
both descended at the same time for at least three consecu-
tive intervals. Minimum change that was accepted was
5 ms for RR and 1 mmHg for SBP. A regression line (RR
as a function of SBP) was Wtted to the detected sequences,
and then correlations between these variables were calcu-
lated (Pearson’s correlation coeYcient), and those with
r> 0.85 were accepted. The BRS value was then obtained
as the mean slope of the regression lines Wtted to all
accepted sequence points (Di Rienzo et al. 2001).
Data analysis: respiration
The respiratory trace was analysed oZine in Chart by look-
ing at cyclic variables and obtaining average cycle length,
and average of maximum peak height and average of peak
minimum height and average cycle height were calculated
(Table 2).
Statistical analysis
Paired ttests were used to statistically analyse the data for
the eVect of the two types of view (natural versus built)
with signiWcance set at p·0.05. Paired ttests were also
used to statistically analyse the data for the eVect of slide
show one compared with slide show two (irrespective of
type of view) with signiWcance set at p·0.05. All the data
given are normally distributed except for absolute values
for HF and LF (as assessed by Kolomogorov–Smirnov test
for normality). The eVect size was calculated using Cohen’s
dfor all physiological variables.
Results
Twenty-nine participants were included in the analysis.
Mean HR, systolic BP (SBP) and diastolic BP (DBP) were
similar while viewing natural or built environment images
(Table 1). The eVect sizes for all these were very small
(drange 0.04–0.19) (Table 1).
There were no signiWcant diVerences in breathing depth
or cycle duration between the diVerent views (Table 2).
Time domain (SDRR, rMSSD), and non-linear (SD1)
indices of vagal outXow were all signiWcantly higher when
viewing natural versus built environments (Fig. 2; Table 2).
23/29 participants increased rMSSD whilst viewing nature
as compared to built views (Fig. 2). Following natural log
transformation of HF (lnHF), there was also a signiWcant
diVerence in lnHF between natural and built environment
views with 21/29 participants with increased lnHF whilst
viewing nature as compared to built views (Fig. 2). EVect
sizes for vagal indices were moderate (Table 3).
BRS values using AR analysis in the HF domain (BRS-
HF) were signiWcantly greater during viewing natural
scenes compared to built scenes (Table 4). BRS values
obtained using sequence analysis (BRS-UP) were signiW-
cantly higher while viewing natural environments, whilst
BRS-combined was also close to statistical signiWcance
(p=0.06) (Table4). EVect sizes for BRS variables are
classed as small to moderate (Table 4).
An analysis of order eVect was undertaken to ensure
responses were due to the images on the slides and not
which set of slides was presented Wrst. No signiWcant diVer-
ences were found for any parameters.
Discussion
Previous experimental work suggests exercising in nature
improves mental health, in particular mood and self-esteem
(Pretty et al. 2005, 2007; Barton et al. 2009; Barton and
Pretty 2010; Bowler et al. 2010; Thompson Coon et al.
Table 1 Cardiovascular variables during the two diVerent slide view-
ings: nature and built
Data are shown as mean (SD); n= 29. EVect size is also shown
(Cohen’s d)
HR mean heart rate, SBP systolic blood pressure, DBP diastolic blood
pressure
Nature Built SigniWcance EVect size
HR (bpm) 62.6 (9.2) 62.6 (9.3) 0.9 0.04
SBP (mmHg) 116.4 (10.3) 118.4 (11.0) 0.15 0.19
DBP (mmHg) 59.0 (11.0) 59.8 (9.0) 0.25 0.08
Table 2 Respiratory values during the two diVerent slide viewings:
nature and built
Data are shown as mean (SD), n=29
Nature Built SigniWcance
Duration of 1 breathing cycle (s) 4.4 (0.8) 4.2 (0.8) 0.12
Average minimum peak height ¡1.1 (2.1) ¡1.3 (1.6) 0.27
Average maximum peak height 3.3 (2.9) 3.1 (2.5) 0.23
Average cycle height 4.4 (3.0) 4.4 (3.1) 0.89
Eur J Appl Physiol
123
2011). However, fewer studies report physiological
responses (Bowler et al. 2010) and the underlying physio-
logical mechanisms involved in viewing nature alone are
unclear. Within a laboratory setting, previous studies have
measured HR (Ulrich et al. 1991; Laumann et al. 2003) and
blood pressure volume (Chang et al. 2008) after inducing
stress Wrst to alter baseline values. The nature views
appeared to have a restorative eVect with greater decreases
towards baseline values after the stressor when viewing
nature environments as compared to viewing built environ-
ments. In addition, following an outdoor walk in nature,
urinary noradrenaline levels were found to be lower versus
a built walk (Li et al. 2011). However, all of the previous
papers have inferred that changes in physiological mea-
sures are likely to be induced by the ANS, but have not
measured ANS control using established methods. Only
one previous study has explored the role of the ANS using
HRV, and this was conducted whilst the participants were
exposed to real environments (Park et al. 2010). Although
real environments allow investigations to be ecologically
valid, it makes it more diYcult to undertake a well-con-
trolled study to investigate ANS mechanisms. Our study is
the Wrst to explore ANS control in a controlled environ-
ment, also enabling BP and respiration to be measured
simultaneously.
In the current study, alterations in cardiovascular auto-
nomic control (in particular vagal activity) were measured
by the use of well-established non-invasive measures of
HRV and BRS. HRV and BRS-HF increased signiWcantly
during the viewing of nature compared with built environ-
ment scenes, suggesting increases in vagal activity. This
suggests that the simple act of viewing nature may induce
changes in autonomic control, in particular vagal activity.
In the current study, the increases in vagal activity are pres-
ent without prior exercise or stress inducing components, or
the additional factors that are present in a real environment.
The real environment and prior exercise may act synergisti-
cally with the nature views to produce greater physiological
eVects. However, in this study, we wished to explore, in a
Fig. 2 Individual comparisons of responses to built and nature views
for arMSSD, bSDRR, clnHF, dlnLF, eSD1, fSD2 (see text for
explanation of abbreviations). Built is shown on the left of each graph
with nature on the right
Table 3 Measures of HRV during the two diVerent slide viewings:
nature and built
Data are shown as mean (SD), n= 29. The absolute values for HF are
1,039.7 (with range 193.8–3,516.8) ms2Hz¡1 and 868.1 (range 172.9–
2,629.7) ms2Hz¡1 collected during viewing natural and built scenes,
respectively. The absolute values for LF are 709.2 (with range 158.1–
2,268.9) ms2Hz¡1 and 970.0 (range 51.4–7,443.8) ms2Hz¡1 collected
during viewing natural and built scenes, respectively. All the data
shown are normally distributed except for absolute values for HF and
LF (as tested by Kolomogorov–Smirnov test for normality). EVect size
is also shown (Cohen’s d)
SDRR standard deviation of RR interval, lnLF natural log of low fre-
quency spectral power (0.04–0.15 Hz), SD2 long-term variability
Poincare plot, rMSSD root mean squared of successive diVerences,
lnHF natural log of high frequency spectral power (0.15–0.4 Hz), SD1
short-term variability Poincare plot
Nature Built SigniWcance EVect size
Overall variability
SDRR (ms) 54.2 (16.0) 49.0 (13.2) 0.001 0.35
lnLF 6.36 (1.0) 6.32 (0.7) 0.8 0.04
SD2 67.0 (19.1) 60.3 (16.1) 0.008 0.38
Vagal mediated
rMSSD (ms) 50.6 (22.1) 46.4 (18.6) 0.008 0.32
lnHF 6.65 (0.8) 6.45 (0.8) 0.007 0.23
SD1 35.8 (15.6) 32.8 (13.1) 0.008 0.21
Table 4 Measures of BRS during the two diVerent slide viewings:
nature and built
Data are shown as mean (SD), n= 29. These measures generally reXect
vagally mediated changes. EVect size is also shown (Cohen’s d)
BRS baroreceptor sensitivity, BRS-HF derived from cross-spectral
analysis, BRS-UP and BRS-combined derived from sequence analysis
Nature Built SigniWcance EVect
size
BRS-HF (ms mmHg¡1) 23.1 (11.9) 20.5 (8.5) 0.008 0.25
BRS-UP (ms mmHg¡1) 15.9 (9.6) 12.9 (5.1) 0.048 0.39
BRS-combined
(ms mmHg¡1)
14.5 (5.6) 13.7 (5.6) 0.06 0.15
Eur J Appl Physiol
123
controlled environment, the underlying physiological
mechanisms of the eVects of nature views without the eVect
of prior exercise or stress.
It is likely that the views of nature induced relaxation
and indeed Park et al. (2010) suggest the augmented vagal
activity they observed while participants were engaged
with nature was due to relaxation. Relaxation is also pro-
posed to occur via stress reduction (Ulrich 1981; Ulrich
et al. 1991) or alterations in attentional capacity (Kaplan
and Kaplan 1989). In earlier studies, EEG has shown an
increase in alpha waves (suggesting relaxation) when view-
ing nature, but they did not deWne particular areas of the
brain (Ulrich 1981). Anecdotally, our participants reported
a preference for viewing the natural scenes and feeling
more relaxed. Relaxation can cause changes in breathing
rate and depth and these in turn can aVect HRV. In the cur-
rent study, due to the controlled environment, breathing
rate and depth were able to be measured and were not sig-
niWcantly diVerent between the views with a breathing fre-
quency at 0.24 Hz. We did not control breathing frequency
as previous papers show that HRV measures are not signiW-
cantly aVected by controlled or free breathing if HR is
within normal ranges (Patwardhan et al. 1995; BloomWeld
et al. 2001). We do understand the need for breathing rate
not to be signiWcantly diVerent for a participant in the two
conditions. In practical terms, we believe that in this study
asking participants to breathe to a metronome may have
caused a distraction from viewing the scenes and thus
altered the physiological eVects of viewing.
The rigorous application of established autonomic mea-
sures in a controlled environment advances existing physio-
logical research in this area. The controlled environment
allowed the act of viewing natural environments to be iso-
lated and eliminated other factors that may alter physiology
when humans are exposed to built or natural environments.
These include potentially negative factors, such as noise,
air pollution or potentially positive factors, such as natural
sounds and phytoncides (Li et al. 2011). The authors of pre-
vious studies suggest that the eVects in cardiovascular
markers are caused by walking or being surrounded by the
relaxing forest environments (Park et al. 2010) with the nat-
ural fragrance of trees (phytoncides) contributing to the
reduction in BP and attenuation of biomarkers in the blood
and urine, including noradrenaline (Li et al. 2011). They
suggest BP reductions may be due to a decrease in sympa-
thetic activity and an increase in parasympathetic activity.
The exercise component itself could contribute to some of
the reduction in sympathetic activity and increase in para-
sympathetic activity, although the BP reductions were
greater following nature walks, suggesting an additional
eVect of nature.
Previous experiments that have used the controlled envi-
ronment of a laboratory have generally used prior exposure
to a stressor to increase participants’ HR and BP before
examining recovery and have not measured ANS activity.
Improved recovery of HR (Ulrich et al. 1991; Laumann
et al. 2003) and BP (Chang et al. 2008) while viewing natu-
ral scenes after exposure to a stressor has led to nature
being regarded as ‘restorative’. In contrast to the previous
experiments, we were interested in exploring the direct
eVects of viewing nature on HRV and BRS at rest without
the addition of exercise or a stressor to establish the physio-
logical mechanisms of viewing nature alone. Unfortu-
nately, the physiological changes are less dramatic when
changes are investigated from baseline values, especially in
a well-controlled experiment, where the participant should
be relaxed prior to taking part in the experimental condi-
tions. At rest, the body aims to maintain homeostasis, and
large decreases were not expected in the already low, rest-
ing values in our study for HR (62 beats per minute) or BP
(117/59 mmHg) on exposure to nature. It would be highly
unlikely and unusual for an individual’s HR to drop by
5 beats per minute from this level, which would be needed
to show a signiWcant diVerence. However, it may be
expected that HR would decrease, in conjunction with the
increase in HRV, but this was not the case. This is unsur-
prising since mean HR cannot reXect the oscillating inXu-
ences on cardiac vagal neurones (Gilbey et al. 1984). In a
previous study involving respiratory training, HRV was
measured alongside other markers of vagal activity (includ-
ing measures of HR recovery following exercise), and all
markers indicated that vagal activity increased despite no
signiWcant change in HR (Hepburn et al. 2005). However,
HRV is a sensitive indicator of ANS function and in partic-
ular vagal activity and the utilisation of such measures adds
to existing laboratory data (Ulrich 1981; Ulrich et al. 1991;
Laumann et al. 2003; Berto 2005; Pretty et al. 2005; Ber-
man et al. 2008; Chang et al. 2008) as it provides, for the
Wrst time, information on how viewing nature aVects the
ANS and the interaction of HR and BP in terms of barore-
ceptor activity. However, for comparisons to be made
within participants, respiratory frequency and depth should
not be signiWcantly altered, as was the case in our study.
Increases in HRV measures may have an important
physiological relevance, despite the lack of signiWcant
changes in HR and BP. The alterations that are seen in
lnHF, rMSSD and SD1 give an indication of the degree of
inXuence on cardiac vagal neurones at respiratory frequen-
cies (Task Force 1996), i.e., it does not necessarily imply
that there is an increase in overall vagal tone, but suggests
increased vagal phasic activity. AVerent input from higher
centres and/or from feedback mechanisms (e.g. barorecep-
tors and chemoreceptors) induce changes in vagal neurone
outXow from the nucleus tractus solitarii (NTS). Increases
in vagal neurone outXow can be caused by an increase in
synaptic excitatory input to cardiac vagal neurones; or a
Eur J Appl Physiol
123
decrease in inhibition elicited by inspiratory drive; or both.
At rest, during normal breathing (as was the case in our
study), if there is an increase in vagal neurone excitability
(maybe via baroreceptor inXuence), R–R interval will
increase. This can be followed by a rebound inhibition of
these vagal neurones by inspiratory drive, decreasing R–R
interval to a greater extent, giving augmented HRV values,
but with an unchanged mean HR. This is likely to be the
case in our participants. The baroreceptors may play a role
in maintaining homeostasis as BRS is elevated during
nature views.
The cause of the increase in vagal excitability is very
interesting, and it is likely that the nucleus ambiguus con-
taining the vagal neurones is inXuenced by other areas of
the brain. Some of the images that we used in our current
study were examined for how aversive and uncomfortable
they were by examining their spatial frequency properties
(Fernandez and Wilkins 2008). Our nature images were
found to be more pleasant and less aversive. Interestingly,
in a recent study which used MRI during natural and urban
views diVerent areas of the brain were stimulated depend-
ing on which scenes were viewed (Kim et al. 2010). The
visual cortex was stimulated more in the urban views than
the natural views. These Wndings may suggest that in part,
the increase in vagal activity may be because the nature
images have less impact on the visual system, with nature
images being able to be processed more eYciently and
inducing diVerent changes within the visual cortex than
images of built slides. The inXuence of visual cortex and
other areas of the brain (maybe the frontal cortex) are likely
to be mediated via the left insular cortex, a region contain-
ing neurones that excite cardiac vagal neurones selectively
via the amygdala as increases in amygdala excitation have
been associated with inhibition of parasympathetic activity
(Thayer and Lane 2009).
Although the results show alterations in HRV measures
for both overall variability and what is considered to be
vagally mediated variability, there were no signiWcant
changes in HR and BP, which may be considered as impor-
tant functional measures. Without signiWcant alterations in
HR and BP, this may limit the conclusions that can be
drawn from the present study.
A further limitation of the study is that although there
are statistically signiWcant diVerences between the two
views, inter-individual diVerences exist in responses to the
slides. This might be due to a lack of continuous engage-
ment from the participants with the projected scenes
throughout the whole slideshow, i.e., the participants’
minds wandering. Alternatively, participants may have
varying aYnities to nature which results in diVerent
responses towards the scenes viewed in this study.
Future studies are needed to explore the mechanisms and
in particular physiological changes that occur in response to
nature, taking into consideration individual diVerences and
aYnity to nature. This study provides an important basis to
start to derive explanations of changes that are seen in
cohort and epidemiological studies.
Conclusion
This study shows for the Wrst time that the simple act of
viewing natural scenes, without the additional factors found
in a real environment, may induce changes in autonomic
control via increases in vagal modulation. The increases in
vagal activity are evident without prior exercise or stress
inducing components.
Acknowledgments V.F. Gladwell is an ESRC research fellow (pro-
ject number RES-064-27-0019). D.K. Brown is supported by a BHF
studentship Grant FS/10/32/28204. The study was also supported by
Academy of Finland (project number 126873).
ConXict of interest None.
References:
Barton J, Pretty J (2010) What is the best dose of nature and green
exercise for improving mental health? A multi-study analysis.
Environ Sci Technol 44:3947–3955
Barton J, Hine RE, Pretty J (2009) The health beneWts of walking in
greenspaces of high natural and heritage value. J Integrative Envi-
ron Sci 6:261–278
Berman MG, Jonides J, Kaplan S (2008) The cognitive beneWts of
interacting with nature. Psychol Sci 19:1207–1212
Berto R (2005) Exposure to restorative environments helps restore
attentional capacity. J Environ Psychol 25:249–259
BloomWeld DM, Magnano A, Bigger JT Jr, Rivadeneira H, Parides
M, Steinman RC (2001) Comparison of spontaneous vs. metro-
nome-guided breathing on assessment of vagal modulation
using RR variability. Am J Physiol Heart Circ Physiol
280:H1145–H1150
Bowler DE, Buyung-Ali LM, Knight TM, Pullin AS (2010) A system-
atic review of evidence for the added beneWts to health of expo-
sure to natural environments. BMC Public Health 10:456
Brennan M, Palaniswami M, Kamen P (2001) Do existing measures of
Poincare plot geom etry reXect nonlinear features of heart rate var-
iability? IEEE Trans Biomed Eng 48:1342–1347
Chang C, Hammitt WE, Chen P, Machnik L, Wei-Chia S (2008)
Psychophysiological responses and restorative values of natural
environments in Taiwan. Landsc Urban Plan 85:79–84
Di Rienzo M, Castiglioni P, Mancia G, Pedotti A, Parati G (2001)
Advancements in estimating baroreXex function. IEEE Eng Med
Biol Mag 20:25–32
Diette GB, Lechtzin N, Haponik E, Devrotes A, Rubin HR (2003)
Distraction therapy with nature sights and sounds reduces pain
during Xexible bronchoscopy: a complementary approach to
routine analgesia. Chest 123:941–948
Fernandez D, Wilkins AJ (2008) Uncomfortable images in art and
nature. Perception 37:1098–1113
Gilbey MP, Jordan D, Richter DW, Spyer KM (1984) Synaptic mech-
anisms involved in the inspiratory modulation of vagal cardio-
inhibitory neurones in the cat. J Physiol 356:65–78
Eur J Appl Physiol
123
Hepburn H, Fletcher J, Rosengarten TH, Coote JH (2005) Cardiac
vagal tone, exercise performance and the eVect of respiratory
training. Eur J Appl Physiol 94:681–689
Kaplan R, Kaplan S (1989) The experience of nature: a psychological
perspective. Cambridge University Press, Cambridge
Kim GW, Jeong GW, Kim TH, Baek HS, Oh SK, Kang HK, Lee SG,
Kim YS, Song JK (2010) Functional neuroanatomy associated
with natural and urban scenic views in the human brain: 3.0T
functional MR imaging. Korean J Radiol 11:507–513
Laumann K, Garling T, Stormark K (2003) Selective attention and
heart rate responses to natural and urban environments. J Environ
Psychol 23:125–134
Lechtzin N, Busse AM, Smith MT, Grossman S, Nesbit S, Diette GB
(2010) A randomized trial of nature scenery and sounds versus
urban scenery and sounds to reduce pain in adults undergoing
bone marrow aspirate and biopsy. J Alternative Complementary
Med 16:965–972
Li Q, Otsuka T, Kobayashi M, Wakayama Y, Inagaki H, Katsumata M,
Hirata Y, Li Y, Hirata K, Shimizu T, Suzuki H, Kawada T,
Kagawa T (2011) Acute eVects of walking in forest environments
on cardiovascular and metabolic parameters. Eur J Appl Physiol
111(11):2845–2853
Moore EO (1982) A prison environment’s eVect on health care
demands. J Environ Syst 11:17–24
Niskanen JP, Tarvainen MP, Ranta-Aho PO, Karjalainen PA (2004)
Software for advanced HRV analysis. Comput Methods Programs
Biomed 76:73–81
Park BJ, Tsunetsugu Y, Kasetani T, Kagawa T, Miyazaki Y (2010) The
physiological eVects of Shinrin-yoku (taking in forest atmosphere
or forest bathing): evidence from Weld experiments in 24 forests
across Japan. Environ Health Prev Med 15:18–26
Patwardhan AR, Evans JM, Bruce EN, Eckberg DL, Knapp CF (1995)
Voluntary control of breathing does not alter vagal modulation of
heart rate. J Appl Physiol 78:2087–2094
Pretty J, Peacock J, Sellens M, GriYn M (2005) The mental and phys-
ical health outcomes of green exercise. Int J Environ Health Res
15:319–337
Pretty J, Peacock J, Hine R, Sellens M, South N, GriYn M (2007)
Green exercise in the UK countryside: eVects on health and psy-
chological well-being. J Environ Plan Manage 50:211–231
Sidery MB, Macdonald IA (1994) The eVect of meal size on the car-
diovascular responses to food ingestion. Br J Nutr 71:835–848
Stubbs TA, Macdonald IA (1995) Systemic and regional haemody-
namic eVects of caVeine and alcohol in fasting subjects [cor-
rected]. Clin Auton Res 5:123–127
Task Force (1996) Heart rate variability: standards of measurement,
physiological interpretation and clinical use. Task Force of the
European Society of Cardiology and the North American Society
of Pacing and Electrophysiology. Circulation 93:1043–1065
Thayer JF, Lane RD (2009) Claude Bernard and the heart–brain con-
nection: further elaboration of a model of neurovisceral integra-
tion. Neurosci Biobehav Rev 33:81–88
Thompson Coon J, Boddy K, Whear R, Barton J, Depledge M, Stein K
(2011) Does participating in physical activity in the outdoor envi-
ronment have a greater eVect on physical and mental wellbeing
than participating in physical activity indoors? A systematic
review. Environ Sci Technol 45(5):1761–1772
Tukek T, Yildiz P, Atilgan D, Tuzcu V, Eren M, Erk O, Demirel S,
Akkaya V, Dilmener M, Korkut F (2003) EVect of diurnal vari-
ability of heart rate on development of arrhythmia in patients with
chronic obstructive pulmonary disease. Int J Cardiol 88:199–206
Ulrich RS (1981) Natural versus urban scenes some psychophysiolog-
ical eVects. Environ Behav 13:523–556
Ulrich RS (1984) View through a window may inXuence recovery
from surgery. Science 224:420–421
Ulrich RS, Simons RF, Fiorito E, Miles MA, Zelson M (1991) Stress
recovery during exposure to natural and urban environments.
J Environ Psychol 11:201–230