The autonomic nervous system in its natural environment: Immersion in nature is associated with changes in heart rate and heart rate variability

Abstract

Stress Recovery Theory (SRT) suggests that time spent in nature reduces stress. While many studies have examined changes in stress physiology after exposure to nature imagery, nature virtual reality, or nature walks, this study is the first to examine changes in heart rate (HR) and vagally mediated HR variability, as assessed by Respiratory Sinus Arrythmia (RSA), after a longer duration of nature exposure. Consistent with SRT, we hypothesized that immersion in nature would promote stress recovery, as indexed by an increase in RSA and a decrease in HR. We also predicted that exposure to nature would improve self-reported mood. We used a within- subjects design (N = 67) to assess changes in peripheral physiology before, during, and after a 5-day nature trip. Results demonstrated a significant decrease in RSA and a significant increase in HR during the trip compared to before or after the trip, suggesting that immersion in nature is associated with a shift toward parasympathetic withdrawal and possible sympathetic activation. These results were contrary to our hypotheses and may suggest increased attentional intake or presence of emotions associated with an increase in sympathetic activation. We also found an improvement in self-reported measures of mood during the trip compared to before or after the trip, confirming our hypotheses and replicating previous research. Implications of this study are discussed in the context of SRT.

Full text

Psychophysiology. 2020;00:e13698. wileyonlinelibrary.com/journal/psyp
|
1 of 14
https://doi.org/10.1111/psyp.13698
© 2020 Society for Psychophysiological Research
1
|
INTRODUCTION
In our modern (Western) world, people are increasingly spend-
ing less time outside (Klepeis etal.,2001). Sedentary indoor
lifestyles can increase risk for negative health outcomes such as
heart disease and obesity (Lichtenstein etal.,2006). Likewise,
living in urban areas has been associated with increased levels
of stress (Lederbogen etal.,2011). Stress Recovery Theory
(SRT; Ulrich et al., 1991) suggests that time spent in natural
environments can promote the recovery of stress. Natural en-
vironments are broadly defined as “areas that contain elements
of living systems that include plants and nonhuman animals
across a range of scales and degrees of human management”
(Bratman etal.,2012; see, also Wilson,1984). SRT draws on
psycho-evolutionary theory (Plutchik,1984) to explain how
nonthreatening nature reduces stress. This theory suggests that
Received: 5 June 2020
|
Revised: 26 August 2020
|
Accepted: 28 August 2020
DOI: 10.1111/psyp.13698
ORIGINAL ARTICLE
The autonomic nervous system in its natural environment:
Immersion in nature is associated with changes in heart rate and
heart rate variability
Emily E.Scott1
|
Sara B.LoTemplio1
|
Amy S.McDonnell1
|
Glen D.McNay2
|
KevinGreenberg3
|
TyMcKinney1
|
Bert N.Uchino1
|
David L.Strayer1
1Department of Psychology, University of
Utah, Salt Lake City, UT, USA
2Department of Health, Kinesiology, and
Recreation, University of Utah, Salt Lake
City, UT, USA
3Department of Educational Psychology,
University of Utah, Salt Lake City, UT,
USA
Correspondence
Emily E. Scott, Department of Psychology,
University of Utah, 380 S 1530 E Beh S
502, Salt Lake City, UT 84112, USA.
Email: Emily.scott@psych.utah.edu
Abstract
Stress Recovery Theory (SRT) suggests that time spent in nature reduces stress.
While many studies have examined changes in stress physiology after exposure to
nature imagery, nature virtual reality, or nature walks, this study is the first to ex-
amine changes in heart rate (HR) and vagally mediated HR variability, as assessed
by Respiratory Sinus Arrythmia (RSA), after a longer duration of nature exposure.
Consistent with SRT, we hypothesized that immersion in nature would promote
stress recovery, as indexed by an increase in RSA and a decrease in HR. We also pre-
dicted that exposure to nature would improve self-reported mood. We used a within-
subjects design (N=67) to assess changes in peripheral physiology before, during,
and after a 5-day nature trip. Results demonstrated a significant decrease in RSA and
a significant increase in HR during the trip compared to before or after the trip, sug-
gesting that immersion in nature is associated with a shift toward parasympathetic
withdrawal and possible sympathetic activation. These results were contrary to our
hypotheses and may suggest increased attentional intake or presence of emotions as-
sociated with an increase in sympathetic activation. We also found an improvement
in self-reported measures of mood during the trip compared to before or after the trip,
confirming our hypotheses and replicating previous research. Implications of this
study are discussed in the context of SRT.
KEYWORDS
ECG, emotion, environmental psychology, heart rate, HRV, nature, stress

2 of 14
|
SCOTT eT al.
changes in emotional tone and arousal served adaptive pur-
poses to enhance our survival and reproductive success. Such
changes would occur quickly and prepare us for fight-or-flight
responses to threatening stimuli, such as a rattlesnake encoun-
ter. In this example, autonomic activation of stress is adaptive
and beneficial to our survival.
Our current environments, however, can promote
chronic stress in response to psychological stressors that do
not quickly pass (Segerstrom & Miller,2004). Activating
the stress response system frequently and over a longer du-
ration can have deleterious effects on immune functioning
and cardiovascular health (Cacioppo etal.,1998; Hawkley
& Cacioppo, 2004; Uchino et al., 2007; Ulrich-Lai &
Herman,2009). Although our physiological stress response
was adaptive from an evolutionary standpoint in response
to a physical threat, our physiological response to modern
day stress continues to reflect the demands of earlier envi-
ronments (Cacioppo etal.,1998). In other words, events or
stimuli that do not require a physical response are evoking
a physical reaction and causing us to stay in elevated states
for longer, threatening our homeostasis (Hobfoll, 1989;
Uchino etal.,2007).
1.1
|
Stress recovery theory
If humans are readily equipped to respond to threatening stim-
uli in natural settings (e.g., snakes), Ulrich et al. (1991) argues
that humans also have a readiness to quickly acquire restorative
responses (e.g., recharge of physical energy) that, by the same
logic, may strongly apply in natural settings. Therefore, Ulrich
proposes that recovery from stress, a restorative response, will
occur more quickly and completely in natural versus urban
settings. Ulrich et al. (1991) suggests two main components
through which this recovery quickly occurs: (1) activation of the
Parasympathetic Nervous System (PNS), our “rest and digest”
system that is activated during recovery from stress, and (2) posi-
tive changes in affect, or emotional states.
In support of SRT, physiological measurements and
self-report measures have shown faster stress recovery and
improved mood with exposure to natural environments ver-
sus urban environments. For example, studies have demon-
strated lower self-reported stress (Beil & Hanes,2013; Roe
etal.,2013), greater positive affect (Beute & de Kort,2014;
Bratman etal., 2015; Lee et al., 2014), improved immune
functioning (Li etal.,2008, 2011), and reductions in stress-re-
lated hormones (i.e., cortisol; Beil & Hanes,2013) following
exposure to natural environments. Some studies have found
decreased HR (Laumann etal.,2003; Lee etal.,2014) and
decreased blood pressure (Hartig et al., 2003; Li,2010) after
viewing nature imagery or taking walks in nature, although
other studies have reported no significant changes in these
measures (Brown etal.,2013; Gladwell etal.,2012).
Additionally, correlational studies have determined
greater well-being (Grinde & Patil,2009; White etal.,2013),
lower levels of health inequality (Mitchell & Popham,2008),
and lower incidences of morbidity, especially depression and
anxiety (Maas et al., 2009), in people who live close to areas
high in green space compared to people who live in areas that
are low in green space. One study even found shorter hospital
stays in rooms with nature views compared to rooms with-
out these views (Ulrich,1984). Taken together, this literature
suggests that both access and proximity to restorative envi-
ronments may have an impact on short-term stress recovery
and long-term cardiovascular health.
1.2
|
Vagal tone reflects central
modulation of the periphery
The vagus nerve links the central and peripheral nervous sys-
tems through neural regulation of the heart (Porges,1995).
Resting vagus nerve activity, or vagal tone, has been used
as a measure of cognitive, emotional, and self-regulation.
Vagal tone can be indirectly captured through indices of va-
gally mediated HR variability (vmHRV), which is thought
to isolate the parasympathetic influence on the heart (Task
Force of the European Society of Cardiology,1996). Indices
of vmHRV such as Respiratory Sinus Arrhythmia (RSA) are
known to be reliable measures that capture variation in heart
rate in synchrony with respiration, by which the beat-to-beat
intervals are shortened during inspiration and expanded dur-
ing expiration. Low resting RSA, or having too “fixed” a HR
interval, is associated with poor regulation of the stress re-
sponse and self-regulatory behaviors. In contrast, high rest-
ing RSA reflects greater variability among HR intervals and
is associated with efficient regulation of stress and behavior
in response to environmental demands (Porges, 1995, 2007;
Smith et al., 2020; Thayer etal.,2009).
The current study examines changes in resting RSA and
HR before, during, and after prolonged immersion in na-
ture. Recent studies have found evidence of greater activity
in measures of vmHRV following exposure to nature imag-
ery (Beute & de Kort, 2014; Brown etal.,2013; Gladwell
etal.,2012) and nature walks (Lee etal.,2014) compared to
their urban equivalents. Findings from these recent studies
are consistent with Ulrich et al. (1991) hypothesis that acti-
vation of the PNS is necessary to facilitate recovery of stress.
However, there is limited research on how longer durations in
nature may affect the autonomic nervous system activity. We
hypothesized that immersion in nature would increase resting
RSA and decrease HR compared to a control testing environ-
ment before and after immersion.
In line with SRT, this study also assesses changes in self-re-
ported mood and self-reported stress with the expectation that
nature exposure would improve mood and reduce stress. Due

|
3 of 14
SCOTT eT al.
to the positive relationship between RSA and social support
(Kok & Fredrickson,2010; Smith etal.,2011, 2020; Uchino
etal.,2020), a self-report measure of social connectivity was
included to examine as a potential covariate. Similarly, we in-
cluded self-report measures of exercise and sleep disturbance,
as well as an objective measure of blood glucose levels, which
could have a metabolic influence on the physiological mea-
sures of interest (Hall et al., 2004; Sandercock etal.,2005;
Singh et al., 2000). For example, blood glucose levels are in-
versely associated with measures of vmHRV in diabetic pa-
tients (Frattola etal.,1997; Rothberg etal.,2016; Singh et al.,
2000). By administering these measures, we can examine how
these factors influence our results.
2
|
MATERIALS AND METHODS
2.1
|
Participants
Participants were recruited from a preexisting upper level
psychology course (Trip 1) and from flyers around the
University of Utah campus and the greater Salt Lake City
community advertising a paid research trip designed for this
study (Trips 2 and 3). A total of 67 participants (33 male, 31
female, 2 transgender, and 1 other/nonbinary) were recruited
across the three trips (Trip 1: N=19, Trip 2: N=22, Trip 3:
N=26), with an age range of 18–46 (M=25.58, SD=6.27).
A majority of participants (90%) identified as White, Non-
Hispanic, 6% identified as Hispanic/Latino, 1% identified as
Pacific Islander, 3% identified as Black/African American,
1% identified as Native American/Alaska Native, and 1%
identified as Other. There were no significant differences in
demographic variables between the three trips.
2.2
|
Study design
Our experiment involved a quasi-experimental, within-sub-
jects design, with all participants completing a 5-day nature
trip in Bluff, UT. All participants completed three, 2-hr ses-
sions to assess changes in electrocardiography (ECG), blood
glucose, and self-report measures. The three sessions took
place up to 2weeks before (pre-testing), during days 2–4 of
the trip (desert testing), and up to 2weeks after (post-testing)
the 5-day nature trip. Repeated measures from the three ses-
sions was the within-subjects factor. We collected three ad-
ministrative waves of this study via the three trips, as trip
logistics did not permit a large sample within a single trip.
Every effort was made to keep trips identical. When in na-
ture, participants completed the same low to moderate inten-
sity hikes at the group's pace (one hike per day ranging from
2–5 miles roundtrip). Aside from hiking, participants relaxed
at the campground (e.g., reading, journaling, and swimming)
and completed their research testing session at their desig-
nated day and time. As part of the class for the first trip, edu-
cational content was provided at night around the campfire.
2.3
|
Procedure
At each session, researchers attached ECG electrodes while
the participant filled out demographic and self-report ques-
tionnaires. All impedances were below 10 kOhms as deter-
mined using BIOPAC’s EL-CHECK electrode impedance
checker. During each testing session, the participants sat in
an enclosed, clear pod made by Under the Weather that is
designed to protect against wind- and weather-related issues.
All data were collected outdoors to control for potential dif-
ferences in signal-to-noise ratios or any artifacts due simply
to being outside. Temperature, time, and weather information
were recorded at each session.
Participants first completed a 10-min baseline used to as-
sess resting changes in RSA, followed by several cognitive
FIGURE 1 Pre and post-testing location
FIGURE 2 Desert testing location

4 of 14
|
SCOTT eT al.
tasks that were part of a separate study.Participants were
instructed to sit upright in a relaxed state, as well as to min-
imize movement and to keep their eyes open. Participants
were tested at the same time for all three sessions to con-
trol for diurnal influences on the dependent variables. All
sessions were identical in procedure but differed in the
environment and location of testing in the second session
(see, Figures1 and 2). In both the desert and the pre- and
post-testing control locations, efforts were made to ensure
that there were no people nearby and that participants sat
in the pods for the duration of testing. In the control loca-
tion, participants were set up in the lab and brought outside
the psychology building for testing on a concrete terrace,
where they could see the building's exterior (see, Figure1)
but did not have a view of nature (i.e., greenery and the
surrounding mountain range). In the desert location, partic-
ipants were set up under a large tent and brought to a sandy
riverbank for testing (see, Figure 2). Altitude (control:
4,226 ft; desert: 4,324 ft) and climate (arid desert) were
similar across both the control and trip testing locations.
The second session took place during the 2nd, 3rd, and 4th
days of the 5-day nature trip (due to logistics we were un-
able to test all participants on the same day).
2.4
|
Measures
2.4.1
|
Electrocardiography (ECG)
ECG data were recorded using BIOPAC Smart Center
(BIOPAC Systems, Goleta, CA, USA). The wireless
BioNomadix Smart Center amplified the ECG signal with a
2kHz per channel maximum sampling rate. ECG data were
observed through AcqKnowledge (Version 5.0) software.
The BioNomadix Smart Center is a small-form data acquisi-
tion unit and wireless receiver that connects to a computer
USB port and records physiological and data from a wireless
transmitter. Physiological data were collected using the lead
II configuration to place active electrodes diagonally across
the heart (Berntson etal.,2007). Before attaching electrodes,
participants were instructed to clean and lightly abrade areas
of the skin using alcohol wipes and NuPrep gel. Electrodes
were attached at each skin site location using SignaGel.
ECG preprocessing pipeline
ECG data were processed using the AcqKnowledge software
following standard guidelines for ECG artifact detection and
correction (Berntson etal.,1990). ECG recordings were band
pass filtered from 0.5–35Hz. Interbeat interval (IBI) time se-
ries were obtained based on the QRS peak detection algorithm
developed by Pan and Tompkins (1985), and QRS peaks were
marked and kept for analysis. Missing peaks were manually
placed and artifacts were edited through visual inspection using
the detection algorithm of Bernston and colleagues (1990).
Data were epoched into 60s segments and epochs with unus-
able contaminated data (e.g., drop in signal) were removed from
analyses (see, percentage removed below under data loss). If a
file contained over 80% of epochs removed due to issues in the
signal, a file was removed from analysis. The AcqKnowledge
software uses a Fast Fourier Transform with a hamming window
function to convert IBIs to the appropriate frequency band for
spectral analyses of RSA (defined as 0.15–40Hz based on rec-
ommendations from the Task Force of the European Society of
Cardiology,1996). Frequency domain (RSA) and time domain
(HR; beats per minute) parameters extracted from the epochs
were then averaged to create overall RSA and HR indices per
file.
Data loss
Of the final data set across all three trips with 67 participants,
2 participants were excluded from ECG analysis (1 recording
failure and 1 participant did not complete ECG testing). Three
additional recordings, one per session, were removed from anal-
ysis due to issues in the signal (over 80% of epochs removed).
The remaining data set for ECG analysis included 192 record-
ings from 65 participants across the three sessions, with 64 re-
cordings per session. Of these recordings, there was an average
of 0.24 epochs (SD = 0.79) removed from the data collapsed
across sessions, which equates to 2.4% of data loss due to data
contamination. Power analyses using G*Power (Cohen,1988;
Faul etal.,2007) revealed that we were powered at 87% to de-
tect repeated-measures changes based on a medium effect size
(Cohen's d= 0.35) and our final physiological sample of 65
participants.
2.4.2
|
Blood glucose
Blood glucose was measured using the OneTouch Verio IQ glu-
cose testing kit. This measurement was included as a potential
covariate in understanding how changes in physiological activ-
ity might be influenced by blood glucose levels. Researchers
used an alcohol swab to clean the tip of either the fourth or fifth
finger of the participant before inserting a disposable lancet into
the finger. Researchers collected a 1µl blood sample onto a glu-
cose test strip, inserted the strip into the glucose reader, and dis-
posed of the testing strip and lancet into a biohazard container.
2.4.3
|
Self-report questionnaires
Participants completed a battery of self-report questionnaires
to assess changes in exercise, sleep, social connectivity, per-
ceived stress, and mood across the three sessions.

|
5 of 14
SCOTT eT al.
Physical activity scale
The Brief Physical Activity Assessment (Marshall
etal.,2005) scale was used to assess changes in moderate
and vigorous weekly exercise. Scores were totaled to display
an overall physical activity score for each participant, with
greater scores reflecting more physical activity.
Sleep disturbance scale
A sleep quality score was calculated by totaling responses
to the eight items on the Sleep Disturbances Scale (Yu
etal., 2012), in which higher scores reflect greater distur-
bances, and thus poorer quality of sleep.
Social connectivity scale
Participants completed this short, 2-item scale adapted from
Kok and Fredrickson (2010). Items were averaged to obtain a
total social connectivity score, with higher scores indicating
greater perceived social connectivity.
Perceived stress scale (PSS)
A shortened version of the PSS was used (Karam etal.,2012),
consisting of four items that asked participants to rate the ex-
tent to which each item applied to them within the past few
days. Items were averaged to obtain a total stress score, with
higher scores indicating greater perceived stress.
Positive and negative affect schedule (PANAS)
Participants completed the PANAS to assess changes in
mood (Watson & Clark,1999). Participants were presented
with a series of words and asked to indicate how they felt
about each word at the present moment. These words were
categorized and summed into positive and negative affect
categories, with higher scores reflecting greater positive or
negative affect.
2.4.4
|
Quantitative methods
All analyses were conducted in R (version 3.4.2) using the
lme4 package (Bates et al., 2012) to determine within-subject
changes in RSA, HR, and subjective measures. We used linear
mixed-effects models in order to account for repeated measures
and random effects on the intercept. We created a three-level
session variable (pre-testing, desert testing, and post-testing)
as a fixed-effect predictor for the dependent measures of inter-
est. Models were estimated using maximum likelihood and p
values were obtained by likelihood ratio tests comparing the
model with the session variable against the intercept model.
Effect sizes were calculated as Cohen's d. Trip wave was also
entered as a predictor into each of the models to determine if
there were differences by trip. Exploratory Pearson correlation
analyses were conducted to determine the association between
subjective measures and blood glucose with physiological
results. Finally, an exploratory mediation analysis was con-
ducted to examine the relationship between RSA and HR
to decouple changes in parasympathetic activity (RSA) and
changes in parasympathetic/sympathetic activity (HR).
3
|
RESULTS
3.1
|
Physiological results
A moderate amount of variance in RSA was accounted for by
the clustering of subject (ICC = 0.54). There was a signifi-
cant omnibus effect of session (χ2(2) = 22.52, p < .001) on
RSA (see, Figure3). There was a significant decrease in RSA
from pre-testing to desert testing (d = −0.54; See, Table1)
and increase back to baseline from desert testing to post-test-
ing (d=0.35; See, Table1). There was no significant differ-
ence in RSA from pre-testing to post-testing. Average RSA
for pre-testing was 7.23 (SE = 0.15), for desert testing was
6.5 (SE = 0.17), and for post-testing was 6.99 (SE=0.16).
FIGURE 3 Average RSA by testing session. Error bars reflect the
standard error of the mean

6 of 14
|
SCOTT eT al.
Results revealed an opposite pattern for HR, with a similar
amount of variance that could be attributed to within-subject
clustering (ICC = 0.59). There was a significant omnibus ef-
fect of session (χ2(2) = 23.12, p < .001) on HR (see, Figure 4).
There was a significant increase in HR from pre-testing to des-
ert testing (d = 0.51; See, Table 1) and a significant decrease
back to baseline from desert testing to post-testing (d = −0.37;
See, Table 1). There was no significant difference in HR from
pre-testing to post-testing. Average HR for pre-testing was 69.6
(SE = 1.3), for desert testing was 75.07 (SE = 1.4), and for
post-testing was 70.82 (SE = 1.28). Trip wave was not a signif-
icant predictor in these models (p values > .6), suggesting that
results did not vary by trip.
3.2
|
Blood glucose results
A relatively small amount of variance in glucose level was
accounted for by the clustering of subject (ICC = 0.07).
There was a significant omnibus effect of session on blood
glucose (χ2(2) = 17.90, p < .001), such that there was a sig-
nificant increase in blood glucose from pre-testing to desert
testing (d=0.44, See, Table2) and a significant decrease
in blood glucose from desert testing to post-testing (d =
−0.65, See, Table2). Average blood glucose for pre-test-
ing was 106.05 (SE=2.43), for desert testing was 114.81
(SE=2.79), and for post-testing was 101.67 (SE=1.9). Trip
wave did not emerge as a significant predictor when entered
into the model (p = .15), suggesting that results did not vary
by trip.
3.3
|
Subjective results
3.3.1
|
Physical activity
A moderate amount of variance in self-reported physical ac-
tivity was accounted for by the clustering of subject (ICC
= 0.40). There was a significant omnibus effect of session
on physical activity (χ2(2) = 9.19, p = .01). There was no
FIGURE 4 Average HR by testing session. Error bars reflect the
standard error of the mean
ΒSE df t p CI UL
CI
LL
RSA
Intercept (Pre) 7.23 0.16 115.04 44.6 <.001*** 6.91 7.54
Pre-Desert −0.72 0.15 126.26 −4.84 <.001*** −1.01 −0.43
Pre-Post −0.25 0.15 126.26 −1.66 .10 −0.54 0.04
Desert-Post 0.47 0.15 125.95 3.18 .002** 0.18 0.76
HR
Intercept (Pre) 69.52 1.34 107.21 51.89 <.001*** 66.89 72.15
Pre-Desert 5.57 1.15 125.70 4.83 <.001*** 3.31 7.83
Pre-Post 1.50 1.15 125.70 1.30 .20 −0.76 3.76
Desert-Post −4.07 1.15 125.42 −3.53 <.001*** −6.32 −1.81
Abbreviations: CI LL, confidence interval lower limit; CI UL, confidence interval upper limit.
*p < .05, **p < .01, ***p < .001.
TABLE 1 Coefficients from linear
mixed models on physiological measures

|
7 of 14
SCOTT eT al.
significant difference in physical activity from pre-testing to
desert testing, but there was a significant decrease in self-
reported physical activity from desert testing to post-testing
(d = −0.41, See, Table 2). Average physical activity for
pre-testing was 2.68 (SE=0.17), for desert testing was 2.95
(SE=0.15), and for post-testing was 2.41 (SE=0.17). Trip
wave did not emerge as a significant predictor when entered
into the model (p = .18), suggesting that results did not vary
by trip.
3.3.2
|
Sleep disturbance
A large amount of variance in self-reported sleep was ac-
counted for by the clustering of subject (ICC = 0.47). There
was no significant omnibus effect of session on sleep dis-
turbance. Average sleep disturbance for pre-testing was 19.2
(SE=0.73), for desert testing was 19.07 (SE=0.68), and for
post-testing was 17.82 (SE=0.69).
βSE df t p CI LL CI UL
Blood glucose
Intercept (Pre) 106.06 2.39 180.53 44.41 <.001*** 101.39 110.73
Pre-Desert 8.70 3.24 122.01 2.68 .01** 2.33 15.03
Pre-Post −4.25 3.25 122.69 −1.21 .19 −10.64 2.12
Desert-Post −12.92 3.27 121.96 −3.96 <.001*** −19.35 −6.53
Physical activity
Intercept (Pre) 2.68 0.17 145.29 16.21 <.001*** 2.35 3.00
Pre-Desert 0.26 0.18 129.39 1.48 .14 −0.09 0.61
Pre-Post −0.28 0.17 129.39 −1.59 .12 −0.63 0.07
Desert-Post −0.54 0.18 129.26 −3.06 .003** −0.90 −0.20
Sleep disturbance
Intercept (Pre) 19.24 0.70 138.38 27.44 <.001*** 17.87 20.62
Pre-Desert −0.19 0.72 128.74 −0.27 .79 −1.60 1.22
Pre-Post −1.42 0.72 129.07 −1.97 .05 −2.83 −0.01
Desert-Post −1.23 0.71 128.00 −1.73 .09 −2.63 0.17
Social connection
Intercept (Pre) 4.73 0.17 100.88 27.95 <.001*** 4.40 5.06
Pre-Desert 0.23 0.13 127.77 1.8 .08 −0.02 0.49
Pre-Post 0.01 0.13 128.28 0.11 .91 −0.25 0.27
Desert-Post −0.22 0.13 127.72 −1.68 .10 −0.48 0.04
Perceived stress
Intercept (Pre) 6.10 0.33 125.17 18.57 <.001 *** 5.46 6.47
Pre-Desert −1.29 0.32 128.45 −4.15 <.001 *** −1.9 −0.68
Pre-Post −0.66 0.32 129.27 −2.09 .04*−1.28 −0.04
Desert-Post 0.63 0.31 128.36 2.02 .05 0.02 1.24
Negative affect
Intercept (Pre) 12.57 0.34 107.42 36.97 <.001*** 11.90 13.23
Pre-Desert −1.09 0.29 128.02 −3.77 .002** −1.66 −0.52
Pre-Post −0.32 0.29 127.81 −1.13 .26 −0.89 0.24
Desert-Post 0.77 0.29 127.75 2.66 .009*0.20 1.33
Positive affect
Intercept (Pre) 27.95 0.92 102.23 30.34 <.001*** 24.24 29.15
Pre-Desert 0.55 0.72 128.70 0.76 .45 −0.86 1.96
Pre-Post −1.94 0.72 129.06 −2.67 .009*−3.36 −0.51
Desert-Post −2.48 0.72 128.49 −3.46 .001** −3.89 −1.08
Abbreviations: CI LL, confidence interval lower limit; CI UL, confidence interval upper limit.
*p < .05, **p < .01, ***p < .001.
TABLE 2 Coefficients from linear
mixed models on blood glucose and
subjective measures

8 of 14
|
SCOTT eT al.
3.3.3
|
Social connection
A large amount of variance in self-reported social connection
was accounted for by the clustering of subject (ICC = 0.70).
There was no significant omnibus effect of session on social
connectivity. Average social connection for pre-testing was
4.73 (SE=0.18), for desert testing was 4.96 (SE=0.16), and
for post-testing was 4.73 (SE=0.17).
3.3.4
|
Perceived stress
A moderate amount of variance in self-reported perceived
stress was accounted for by the clustering of subject (ICC
= 0.52). There was a significant omnibus effect of session
on perceived stress (χ2(2) = 16.45, p < .001). There was a
significant decrease in perceived stress from pre-testing to
desert testing (d = −0.47, See, Table2), and from pre-testing
to post-testing (d = −0.24, See, Table2). There was also a
marginally significant increase in perceived stress from de-
sert testing to post-testing (d=0.23, See, Table2). Average
perceived stress for pre-testing was 6.12 (SE=0.35), for de-
sert testing was 4.81 (SE=0.32), and for post-testing was
5.47 (SE=0.32). Trip wave did not emerge as a significant
predictor when entered into the model (p = .42), suggesting
that results did not vary by trip.
3.3.5
|
Negative affect
A moderate amount of variance in self-reported negative affect
was accounted for by the clustering of subject (ICC = 0.62).
There was a significant omnibus effect of session on negative
affect (χ2(2) = 14.32, p < .001). There was a significant de-
crease in self-reported negative affect from pre-testing to desert
testing (d = −0.39, See, Table2) and a significant increase from
desert testing to post-testing (d=0.27, See, Table2) but there
was no significant difference in negative affect from pre-test-
ing to post-testing. Average negative affect for pre-testing was
12.45 (SE=0.34), for desert testing was 11.46 (SE=0.27),
and for post-testing was 12.24 (SE=0.39). Trip wave did not
emerge as a significant predictor when entered into the model
(p = .71), suggesting that results did not vary by trip.
3.3.6
|
Positive affect
A large amount of variance in self-reported positive affect
was accounted for by the clustering of subject (ICC = 0.67).
There was a significant omnibus effect of session on positive
affect (χ2(2) = 12.69, p = .002). There was no difference in
self-reported positive affect from pre-testing to desert test-
ing, but there was a significant decrease in positive affect
from pre-testing to post-testing (d = −0.25, See, Table 2)
and from desert testing to post-testing (d = −0.33, See,
Table2). Average positive affect for pre-testing was 27.78
(SE=0.94), for desert testing was 28.49 (SE=0.94), and for
post-testing was 25.94 (SE=0.89). Trip wave did not emerge
as a significant predictor when entered into the model (p =
.21), suggesting that results did not vary by trip.
3.4
|
Exploratory analyses
As expected, RSA and HR were strongly and negatively
correlated (r = −0.58, p < .001) such that increases in RSA
were associated with decreases in HR. Because we did not
have a direct measure of sympathetic activation, mediational
analyses were performed to determine the degree to which
parasympathetic withdraw accounted for changes in HR to
indirectly decouple parasympathetic and sympathetic activ-
ity. Notably, the mediation was conducted on coefficients
from the pre-testing to desert testing contrast. First, RSA was
entered as a predictor in the linear mixed model for HR. RSA
emerged as a significant predictor (B = −5.08, SE=0.44,
df=179.06, t = −11.53, p < .001 [CI = −5.94, −4.22] in
accounting for changes in HR over time. There was still a
significant increase in HR from pre-testing to desert testing
after controlling for RSA (B=1.89, SE=0.89, df=131.70,
t=2.12, p = .04 [CI=0.15, 3.64); however, this variance
was reduced from the original model. Multilevel mediation
analysis using Bayesian estimation (simulations = 5,000)
was performed in R using the mediate package (Tofighi &
MacKinnon,2011) with session (pre-testing to desert testing)
as the predictor variable, RSA as the mediator variable, and
HR as the outcome variable (see, Figure5). The indirect ef-
fect of RSA on HR was found to be statistically significant
(effect=3.65, p < .001 [CI=2.09, 5.32]), confirming that
RSA mediated the relationship between session and HR.
RSA was also positively associated with self-reported ex-
ercise, such that increases in exercise were associated with
increases in RSA (r = 0.14, p < .05). However, when en-
tered as a predictor into the model, self-reported exercise did
not predict changes in RSA (p = .22). HR was negatively
associated with self-reported exercise, such that increases in
FIGURE 5 Coefficients for the relationship between session and
HR as mediated by RSA. The coefficient between session and HR,
controlling for RSA, is in parentheses

|
9 of 14
SCOTT eT al.
exercise were associated with decreases in HR (r = −0.19,
p < .01). When entered as a predictor into the linear mixed
model, self-reported exercise emerged as a marginally signif-
icant predictor of changes in HR (p = .06). RSA and HR were
not significantly associated with any of the other subjective
measures or blood glucose levels (p values > .2).
Temperature (recorded in Fahrenheit) and type of weather
(e.g., sunny/partially sunny, windy, cloudy/overcast, etc.)
were recorded at each testing session to examine as possible
external factors that could be influencing dependent mea-
sures of interest. Weather was not significantly correlated
with RSA or HR (p values > .24). Temperature was margin-
ally correlated with RSA (r = −0.13, p = .07) and was sig-
nificantly and positively correlated with HR (r = 0.16, p <
.05), such that increases in temperature were associated with
increases in HR. When entered as a covariate in the linear
mixed models, temperature emerged as a significant predictor
of changes in HR (B=0.12, SE=0.06, df=173.41, t=2.13,
p = .03 [CI=0.01, 0.24]) and a marginally significant pre-
dictor of changes in RSA (p = .06). However, there was no
significant change in average temperature from pre-testing to
desert testing (t=1.06, p = .29), thus it is unlikely that in-
creases in temperature are driving the decrease in RSA and
increase in HR observed in the desert.
Finally, day of testing session (2, 3, or 4 of the 5-day trip)
in the desert was examined to determine whether there were
differences in average RSA or HR by testing day. Results
demonstrated that there was no significant change in aver-
age RSA (F(1, 61) = 0.266, p = .645), or HR (F(1, 61) =
0.214, p = .608) between the three testing days (RSA: Day
2 – M=6.61, SE=0.19; Day 3 – M=6.15, SE=0.34; Day
4 – M=7.08, SE=0.32; HR: Day 2 – M=74.28, SE=2.52;
Day 3 – M=75.61; SE=2.48; Day 4 – M=76, SE=1.83).
Furthermore, testing day did not emerge as a significant pre-
dictor when entered into the linear mixed models for HR or
RSA (p values > .33), suggesting that results did not vary by
testing day in the desert.
4
|
DISCUSSION
This study was one of the first to examine changes in vmHRV
during prolonged immersion in nature. While some studies
have found increases in vmHRV after nature walks or viewing
nature imagery compared to their urban equivalents (Beute
& de Kort,2014; Brown etal.,2013; Gladwell etal.,2012;
Lee etal.,2014), our study found a decrease in vmHRV and
an increase in HR while immersed in nature. These findings
were contradictory to our hypotheses of a shift toward para-
sympathetic activity in nature. Although we did not have a
direct measure of sympathetic activity, our exploratory me-
diation analysis suggests that the decrease in RSA from pre-
testing to desert testing accounts for most, but not all of the
variance involving changes in HR. These findings suggest a
shift toward parasympathetic withdrawal and possibly some
sympathetic excitation during immersion in nature compared
to a control testing environment. Replicating previous re-
search, subjective results showed an improvement in mood
and decreased stress in the desert. None of our results varied
by trip wave, suggesting these effects replicated across three
independent samples. Results from this study are discussed in
the context of SRT.
A decrease in RSA has been associated with poor cog-
nitive, emotional, and self-regulation via increased auto-
nomic activation and downregulation of the prefrontal cortex
(Porges, 1995, 2007; Thayer et al., 2009, 2012). On our
extended trip, participants were removed from their every-
day environment and could be experiencing dysregulation
while being away from friends, family, and responsibili-
ties as lower RSA is also linked to a lack of social support
(Kok & Fredrickson,2010; Smith etal.,2011, 2020; Uchino
et al., 2020). While there was no change in self-reported
social connectivity on the trip, future studies could design
conditions in an attempt to isolate social influence on nature
exposure. However, participants consistently give anecdotal
reports of positive experiences on these trips as well as our
findings showing improvements in self-reported mood on the
trip. As increases in vmHRV are also associated with increases
in cognitive demands and recruitment of the prefrontal cortex
(Honey etal.,2002; Nikolin etal.,2017; Thayer etal.,2009),
a decrease in RSA in nature could reflect downregulation of
the prefrontal cortex while participants experience decreased
cognitive demands in nature. The increase in blood glucose
levels observed in the desert, while unrelated to the physio-
logical measures of interest, may also reflect this downregu-
lation, as studies have found decreased blood glucose levels
after participants experience depletion from cognitive and
self-regulatory (effortful) tasks (Gailliot & Baumeister,2007;
Hagger etal.,2010; Heatherton & Wagner,2011). However,
this interpretation is subject to mixed findings and appears
to be sensitive to task demands, as some studies have found
decreases in vmHRV with increasing cognitive workload
(Gianaros etal.,2004) and have failed to replicate the link
between depletion, glucose, and effort (Hagger etal.,2016).
A shift in sympathetic activity could also be due to posi-
tive changes in mood; for example, several studies have found
increases in sympathetic activity after participants were pre-
sented with film clips or imagery to induce various aspects
of positive mood (Christie & Friedman, 2003; Demaree
etal.,2004; Giuliani etal.,2008; Mauss etal.,2005; Shiota
et al., 2011). However, this literature is mixed, with other
studies showing little or even reduced physiological arousal
in response to positive mood manipulations compared to
negative ones (Cacioppo et al., 2000; Levenson, 1992).
Some studies also show increased sympathetic activity due
to negative emotions or feelings of threat-based (as opposed

10 of 14
|
SCOTT eT al.
to positive) feelings of awe (Cacioppo etal.,2000; Gordon
et al., 2017). However, because we observed a significant
decrease in self-reported stress and negative affect in nature
compared to before or after the trip, it is unlikely that the
shift in sympathetic activity found in the desert could be due
to an increased presence of negative emotions. Future stud-
ies could incorporate a more diverse set of emotion-related
measures (e.g., awe and excitement) to further determine
how emotion may influence the autonomic activity in na-
ture. Research could also incorporate physiological measures
of sympathetic activity (e.g., pre-ejection period; Cacioppo
etal.,1994) to corroborate these findings. The dissociation
between self-report and physiological measures of stress that
we observed in our study is not consistent with SRT, and
more research is needed to confirm these effects.
Changes in autonomic activity could also be due to
changes in physical activity (Perini & Veicsteinas, 2003;
Tulppo etal.,1996). While RSA and HR were correlated with
self-reported exercise, exercise did not predict changes in
RSA and marginally predicted changes in HR. Furthermore,
self-reported exercise did not increase in nature, as there was
no difference in self-reported exercise between pre-testing and
testing in the desert. It is possible we did not find differences
in self-reported physical activity because Salt Lake City is an
active city with easy access to hiking and other outdoor activ-
ities, and thus participants may have maintained their current
level of exercise on the nature trip. Future research should use
objective measures of physical activity (e.g., reliable pedom-
eters) to determine the degree to which exercise influences
changes in physiology in nature. Experimental designs could
also tease apart the relative influence of exercise and nature
exposure by having a condition that exercises and a condition
that does not exercise in nature, or recruit participants with
low versus high levels of physical activity.
Autonomic activity has been linked to attentional intake
(Berntson & Boysen, 1987). For example, research sug-
gests that there is an inverted u-shaped relationship between
arousal and attention, such that optimal levels of arousal pro-
mote attention and increases in task performance, whereas
suboptimal and supraoptimal levels are associated with lower
levels of attention and worse task performance (Yerkes &
Dodson, 1908). Nature contains many stimuli that could
capture attention and thereby increase arousal. For example,
Ulrich et al. (1991) found increases in HR while viewing a
nature video, attributing these results to increases in involun-
tary (autonomic) attention. Our results may be due to greater
levels of involuntary attention in the desert compared to the
control testing environment.
The mixed findings on vmHRV in the nature literature
could be explained by differences in study design, measure-
ment, and duration of nature exposure. For example, in one
study, nature or urban imagery slides were presented after a
series of cognitive depletion tasks (Beute & de Kort,2014);
in another study nature imagery or urban imagery slides were
presented before a cognitive task (Brown etal.,2013) which
raises the differences in stress recovery versus reactivity. In
other studies, vmHRV was assessed during nature imagery or
a walk through the forest without the presence of a depletion
task (Gladwell etal.,2012; Lee etal.,2014). Likewise, these
studies each used different indices of vmHRV, including Root
Mean Square of Successive Differences (RMSSD), low fre-
quency (LF) to high frequency (HF) ratio (LF/HF), and HF,
which could have different implications for interpretations of
vmHRV (Smith etal.,2020). RSA is highly correlated with
other time domain (e.g., RMSSD) and HF measures that also
isolate the vagal activity (Cacioppo etal.,1994; Grossman
etal.,1990), but the interpretation of LF and LF/HF mea-
sures as an index of sympathetic activation or the ratio of
sympathetic to parasympathetic activation have been refuted,
as LF bands are thought to contain both sympathetic and
parasympathetic influences (Cacioppo et al., 1994; Moak
etal.,2007). Likewise, each of the studies included an urban
equivalent, suggesting that differences in vmHRV could also
be attributed to specific nature-urban differences. Future re-
search could look at the length, as well as the “level” of na-
ture exposure (nature imagery vs. nature virtual reality vs.
real-world nature) on measures of vmHRV. For example,
there has yet to be a study examining length of time spent in
nature (e.g., short nature walks vs. several days in nature) on
physiological measures.
4.1
|
Limitations and future directions
Although we attempted to measure potential covariates,
such as glucose, social connectivity, exercise, mood, and
sleep, there are many dynamic factors that may be chang-
ing as a result of nature exposure. From our within-subjects
design, we established that changes in RSA and HR ap-
pear to be modulated by immersion in nature and are not
explained by the other variables we measured. However,
nature contains many different types of stimuli (e.g., vis-
ual and auditory) and we were unable to tease apart these
stimuli in the current study to examine the respective influ-
ence of different environmental features in understanding
our results. Likewise, it is possible that natural environ-
ments may be more cognitively stimulating, rather than
something about nature per se that is causing the observed
change in autonomic activity. Future research could exam-
ine other manipulations that are cognitively stimulating,
including artistic or novel scenery, to understand how this
aspect may be contributing to these results. Future research
should continue to administer a comprehensive battery
of measures to determine the degree to which changes in
physiology during nature exposure could covary with other
factors. Studies could also include an urban trip equivalent

|
11 of 14
SCOTT eT al.
to further understand differences between nature and urban
environments on vmHRV.
These novel findings imply that exposure to nature may
boost mood and enhance arousal. The current research dif-
fers from past findings in that it is one of the first to exam-
ine the autonomic activity during prolonged immersion in
nature and to replicate the results across multiple samples.
This research is a critical first step in examining SRT over
longer durations of nature exposure and raises important
methodological and theoretical considerations for future
research.
ACKNOWLEDGMENTS
The authors thank Bret Bradshaw, Kaedyn Crabtree, Tyler
Jette, Trent Simmons, InJeong Song, Jess Thayne, Sean
Cook, Devon Jecmen, Vicky Weaver, Robert Kennedy,
Mason Stephens, Ben Sky, Logan Call, Vicki Gilbert, and
Lauren Ziegelmayer for their help during data collection.
Special thanks to Lauren Ziegelmayer, Daniel Anderson,
Dallen Garner, and Vicki Gilbert for their assistance in data
processing. The authors also thank the town of Bluff, UT,
and Sand Island Campground for letting us use their facili-
ties. Without this collective effort, this research would not
be possible.
AUTHOR CONTRIBUTIONS
Emily E Scott: Conceptualization; Data curation;
Formal analysis; Methodology; Project administra-
tion; Writing-original draft; Writing-review & editing.
Sara B Lotemplio: Conceptualization; Data curation;
Project administration; Writing-review & editing. Amy S
McDonnell: Conceptualization; Data curation; Project ad-
ministration; Writing-review & editing. Glen D McNay:
Data curation; Project administration; Writing-review
& editing. Kevin Greenberg: Data curation; Project ad-
ministration; Writing-review & editing. Ty McKinney:
Data curation; Project administration; Writing-review &
editing. Bert Uchino: Conceptualization; Methodology;
Supervision; Writing-review & editing. David L Strayer:
Conceptualization; Resources; Supervision; Writing-
review & editing.
ORCID
Emily E. Scott https://orcid.org/0000-0002-9750-5827
REFERENCES
Bates, D., Maechler, M., & Bolker, B. (2012). lme4: Linear mixed-ef-
fects models using S4 classes (R package version 0.999999-0).
Beil, K., & Hanes, D. (2013). The influence of urban natural and
built environments on physiological and psychological measures
of stress—A pilot study. International Journal of Environmental
Research and Public Health, 10(4), 1250–1267. https://doi.
org/10.3390/ijerp h1004 1250
Berntson, G. G., & Boysen, S. T. (1987). Cardiac reflections of at-
tention and preparatory set in a chimpanzee (Pan troglodytes).
Psychobiology, 15(1), 87–92.
Berntson, G. G., Quigley, K. S., Jang, J. F., & Boysen, S. T. (1990).
An approach to artifact identification: Application to heart pe-
riod data. Psychophysiology, 27(5), 586–598. https://doi.
org/10.1111/j.1469-8986.1990.tb019 82.x
Berntson, G. G., Quigley, K. S., & Lozano, D. (2007). Cardiovascular
psychophysiology. In J. T. Cacioppo, L. G. Tassinary, & G. G.
Berntson (Eds.), Handbook of psychophysiology (3rd ed., pp. 182–
210). Cambridge University Press.
Beute, F., & de Kort, Y. A. W. (2014). Natural resistance: Exposure
to nature and self-regulation, mood, and physiology after ego-de-
pletion. Journal of Environmental Psychology, 40, 167–178. https://
doi.org/10.1016/j.jenvp.2014.06.004
Bratman, G. N., Daily, G. C., Levy, B. J., & Gross, J. J. (2015). The bene-
fits of nature experience: Improved affect and cognition. Landscape
and Urban Planning, 138, 41–50. https://doi.org/10.1016/j.landu
rbplan.2015.02.005
Bratman, G. N., Hamilton, J. P., & Daily, G. C. (2012). The impacts of
nature experience on human cognitive function and mental health.
Annals of the New York Academy of Sciences, 1249(1), 118–136.
https://doi.org/10.1111/j.1749-6632.2011.06400.x
Brown, D. K., Barton, J. L., & Gladwell, V. F. (2013). Viewing nature
scenes positively affects recovery of autonomic function following
acute-mental stress. Environmental Science & Technology, 47(11),
5562–5569. https://doi.org/10.1021/es305 019p
Cacioppo, J. T., Berntson, G. G., Binkley, P. F., Quigley, K. S.,
Uchino, B. N., & Fieldstone, A. (1994). Autonomic cardiac con-
trol. II. Noninvasive indices and basal response as revealed by au-
tonomic blockades. Psychophysiology, 31(6), 586–598. https://doi.
org/10.1111/j.1469-8986.1994.tb023 51.x
Cacioppo, J. T., Berntson, G. G., Larsen, J. T., Pohlmann, K. M., &
Ito, T. A. (2000). The psychophysiology of emotion. In M. Lewis
& J. M. Haviland-Jones (Eds.), Handbook of emotions (2nd ed., pp.
173–191). Guilford Press.
Cacioppo, J. T., Berntson, G. G., Malarkey, W. B., Kiecolt-Glaser, J.
K., Sheridan, J. F., Poehlmann, K. M., Burleson, M. H., Ernst, J.
M., Hawkley, L. C., & Glaser, R. (1998). Autonomic, neuroendo-
crine, and immune responses to psychological stress: The reactiv-
ity hypothesis. Annals of the New York Academy of Sciences, 840,
664–673. https://doi.org/10.1111/j.1749-6632.1998.tb096 05.x
Christie, I. C., & Friedman, B. H. (2003). Autonomic specificity of
discrete emotion and dimensions of affective space: A multivariate
approach. International Journal of Psychophysiology, 51, 143–153.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Lawrence Erlbaum.
Demaree, H. A., Schmeichel, B. J., Robinson, J. L., & Everhart, D. E.
(2004). Behavioural, affective, and physiological effects of negative
and positive emotional exaggeration. Cognition and Emotion, 18,
1079–1097. https://doi.org/10.1080/02699 93044 1000085
Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G* Power 3:
A flexible statistical power analysis program for the social, behav-
ioral, and biomedical sciences. Behavior Research Methods, 39(2),
175–191. https://doi.org/10.3758/BF031 93146
Frattola, A., Parati, G., Gamba, P., Paleari, F., Mauri, G., Di Rienzo, M.,
Castiglioni, P., & Mancia, G. (1997). Time and frequency domain
estimates of spontaneous baroreflex sensitivity provide early detec-
tion of autonomic dysfunction in diabetes mellitus. Diabetologia,
40(12), 1470–1475. https://doi.org/10.1007/s0012 50050851

12 of 14
|
SCOTT eT al.
Gailliot, M. T., & Baumeister, R. F. (2007). The physiology of will-
power: Linking blood glucose to self-control. Personality and Social
Psychology Review, 11(4), 303–327. https://doi.org/10.1177/10888
68307 303030
Gianaros, P. J., Van der Veen, F. M., & Jennings, J. R. (2004). Regional
cerebral blood flow correlates with heart period and high-fre-
quency heart period variability during working-memory tasks:
Implications for the cortical and subcortical regulation of cardiac
autonomic activity. Psychophysiology, 41(4), 521–530. https://doi.
org/10.1111/1469-8986.2004.00179.x
Giuliani, N. R., McRae, K., & Gross, J. J. (2008). The up-and down-reg-
ulation of amusement: Experiential, behavioral, and autonomic
consequences. Emotion, 8(5), 714–719. https://doi.org/10.1037/
a0013236
Gladwell, V. F., Brown, D. K., Barton, J. L., Tarvainen, M. P., Kuoppa,
P., Pretty, J., Suddaby, J. M., & Sandercock, G. R. H. (2012). The
effects of views of nature on autonomic control. European Journal
of Applied Physiology, 112(9), 3379–3386. https://doi.org/10.1007/
s0042 1-012-2318-8
Gordon, A. M., Stellar, J. E., Anderson, C. L., McNeil, G. D., Loew, D.,
& Keltner, D. (2017). The dark side of the sublime: Distinguishing
a threat-based variant of awe. Journal of Personality and Social
Psychology, 113(2), 310. https://doi.org/10.1037/pspp0 000120
Grinde, B., & Patil, G. G. (2009). Biophilia: Does visual contact with
nature impact on health and well-being? International Journal of
Environmental Research and Public Health, 6(9), 2332–2343.
https://doi.org/10.3390/ijerp h6092332
Grossman, P., Van Beek, J., & Wientjes, C. (1990). A compari-
son of three quantification methods for estimation of respiratory
sinus arrhythmia. Psychophysiology, 27(6), 702–714. https://doi.
org/10.1111/j.1469-8986.1990.tb031 98.x
Hagger, M. S., Chatzisarantis, N. L. D., Alberts, H., Anggono, C.
O., Batailler, C., Birt, A. R., Brand, R., Brandt, M. J., Brewer,
G., Bruyneel, S., Calvillo, D. P., Campbell, W. K., Cannon, P.
R., Carlucci, M., Carruth, N. P., Cheung, T., Crowell, A., De
Ridder, D. T. D., Dewitte, S., … Zwienenberg, M. (2016). A
multilab preregistered replication of the ego-depletion effect.
Perspectives on Psychological Science, 11(4), 546–573. https://doi.
org/10.1177/17456 91616 652873
Hagger, M. S., Wood, C., Stiff, C., & Chatzisarantis, N. L. (2010). Ego
depletion and the strength model of self-control: A meta-analy-
sis. Psychological Bulletin, 136(4), 495. https://doi.org/10.1037/
a0019486
Hall, M., Vasko, R., Buysse, D., Ombao, H., Chen, Q., Cashmere, J.
D., & Thayer, J. F. (2004). Acute stress affects HR variability
during sleep. Psychosomatic Medicine, 66(1), 56–62. https://doi.
org/10.1097/01.PSY.00001 06884.58744.09
Hartig, T., Evans, G. W., Jamner, L. D., Davis, D. S., & Gärling, T.
(2003). Tracking restoration in natural and urban field settings.
Journal of Environmental Psychology, 23(2), 109–123. https://doi.
org/10.1016/S0272 -4944(02)00109 -3
Hawkley, L. C., & Cacioppo, J. T. (2004). Stress and the aging immune
system. Brain, Behavior, and Immunity, 18(2), 114–119. https://doi.
org/10.1016/j.bbi.2003.09.005
Heatherton, T. F., & Wagner, D. D. (2011). Cognitive neuroscience of
self-regulation failure. Trends in Cognitive Sciences, 15(3), 132–
139. https://doi.org/10.1016/j.tics.2010.12.005
Hobfoll, S. E. (1989). Conservation of resources: A new attempt at con-
ceptualizing stress. American Psychologist, 44, 513–524. https://
doi.org/10.1037/0003-066X.44.3.513
Honey, G. D., Fu, C., Kim, J., Brammer, M. J., Croudace, T. J.,
Suckling, J., Pich, E. M., Williams, S., & Bullmore, E. T. (2002).
Effects of verbal working memory load on corticocortical connec-
tivity modeled by path analysis of functional magnetic resonance
imaging data. NeuroImage, 17(2), 573–582. https://doi.org/10.1006/
nimg.2002.1193
Karam, F., Bérard, A., Sheehy, O., Huneau, M.-C., Briggs, G.,
Chambers, C., Einarson, A., Johnson, D., Kao, K., Koren, G.,
Martin, B., Polifka, J. E., Riordan, S. H., Roth, M., Lavigne, S. V.,
& Wolfe, L. (2012). Reliability and validity of the 4-item Perceived
Stress Scale among pregnant women: Results from the OTIS anti-
depressants study. Research in Nursing & Health, 35(4), 363–375.
https://doi.org/10.1002/nur.21482
Klepeis, N. E., Nelson, W. C., Ott, W. R., Robinson, J. P., Tsang, A. M.,
Switzer, P., Behar, J. V., Hern, S. C., & Engelmann, W. H. (2001).
The National Human Activity Pattern Survey (NHAPS): A re-
source for assessing exposure to environmental pollutants. Journal
of Exposure Science and Environmental Epidemiology, 11(3), 231.
https://doi.org/10.1038/sj.jea.7500165
Kok, B. E., & Fredrickson, B. L. (2010). Upward spirals of the heart:
Autonomic flexibility, as indexed by vagal tone, reciprocally and
prospectively predicts positive emotions and social connectedness.
Biological Psychology, 85(3), 432–436. https://doi.org/10.1016/j.
biops ycho.2010.09.005
Laumann, K., Gärling, T., & Stormark, K. M. (2003). Selective attention
and HR responses to natural and urban environments. Journal of
Environmental Psychology, 23(2), 125–134.
Lederbogen, F., Kirsch, P., Haddad, L., Streit, F., Tost, H., Schuch, P.,
Wüst, S., Pruessner, J. C., Rietschel, M., Deuschle, M., & Meyer-
Lindenberg, A. (2011). City living and urban upbringing affect
neural social stress processing in humans. Nature, 474(7352), 498.
https://doi.org/10.1038/natur e10190
Lee, J., Tsunetsugu, Y., Takayama, N., Park, B.-J., Li, Q., Song, C.,
Komatsu, M., Ikei, H., Tyrväinen, L., Kagawa, T., & Miyazaki,
Y. (2014). Influence of forest therapy on cardiovascular relaxation
in young adults. Evidence-Based Complementary and Alternative
Medicine, 2014, https://doi.org/10.1155/2014/834360
Levenson, R. W. (1992). Autonomic nervous system differences
among emotions. Psychological Science, 3, 23–27. https://doi.
org/10.1111/j.1467-9280.1992.tb002 51.x
Li, Q. (2010). Effect of forest bathing trips on human immune function.
Environmental Health and Preventive Medicine, 15(1), 9. https://
doi.org/10.1007/s1219 9-008-0068-3
Li, Q., Morimoto, K., Kobayashi, M., Inagaki, H., Katsumata,
M., Hirata, Y., Hirata, K., Suzuki, H., Li, Y. J., Wakayama,
Y., Kawada, T., Park, B. J., Ohira, T., Matsui, N., Kagawa, T.,
Miyazaki, Y., & Krensky, A. M. (2008). Visiting a forest, but not a
city, increases human natural killer activity and expression of an-
ti-cancer proteins. International Journal of Immunopathology and
Pharmacology, 21(1), 117–127. https://doi.org/10.1177/03946
32008 02100113
Li, Q., Otsuka, T., Kobayashi, M., Wakayama, Y., Inagaki, H.,
Katsumata, M., Hirata, Y., Li, Y. J., Hirata, K., Shimizu, T., Suzuki,
H., Kawada, T., & Kagawa, T. (2011). Acute effects of walking in
forest environments on cardiovascular and metabolic parameters.
European Journal of Applied Physiology, 111(11), 2845–2853.
https://doi.org/10.1007/s0042 1-011-1918-z
Lichtenstein, A. H., Appel, L. J., Brands, M., Carnethon, M., Daniels,
S., Franch, H. A., Franklin, B., Kris-Etherton, P., Harris, W. S.,
Howard, B., Karanja, N., Lefevre, M., Rudel, L., Sacks, F., Van

|
13 of 14
SCOTT eT al.
Horn, L., Winston, M., Wylie-Rosett, J., & Karanja, N. (2006). Diet
and lifestyle recommendations revision 2006: A scientific state-
ment from the American Heart Association Nutrition Committee.
Circulation, 114(1), 82–96. https://doi.org/10.1161/CIRCU LATIO
NAHA.106.176158
Maas, J., Verheij, R. A., de Vries, S., Spreeuwenberg, P., Schellevis,
F. G., & Groenewegen, P. P. (2009). Morbidity is related
to a green living environment. Journal of Epidemiology &
Community Health, 63(12), 967–973. https://doi.org/10.1136/
jech.2008.079038
Marshall, A. L., Smith, B. J., Bauman, A. E., & Kaur, S. (2005).
Reliability and validity of a brief physical activity assessment for
use by family doctors. British Journal of Sports Medicine, 39(5),
294–297. https://doi.org/10.1136/bjsm.2004.013771
Mauss, I. B., Levenson, R. W., McCarter, L., Wilhelm, F. H., & Gross,
J. J. (2005). The tie that binds? Coherence among emotion expe-
rience, behavior, and physiology. Emotion, 5, 175–190. https://doi.
org/10.1037/1528-3542.5.2.175
Mitchell, R., & Popham, F. (2008). Effect of exposure to natural envi-
ronment on health inequalities: An observational population study.
The Lancet, 372(9650), 1655–1660. https://doi.org/10.1016/S0140
-6736(08)61689 -X
Moak, J. P., Goldstein, D. S., Eldadah, B. A., Saleem, A., Holmes, C.,
Pechnik, S., & Sharabi, Y. (2007). Supine low-frequency power of HR
variability reflects baroreflex function, not cardiac sympathetic innerva-
tion. Heart Rhythm: The Official Journal of the Heart Rhythm Society,
4(12), 1523–1529. https://doi.org/10.1016/j.hrthm.2007.07.019
Nikolin, S., Boonstra, T. W., Loo, C. K., & Martin, D. (2017).
Combined effect of prefrontal transcranial direct current stimulation
and a working memory task on HR variability. PLoS One, 12(8),
e0181833. https://doi.org/10.1371/journ al.pone.0181833
Pan, J., & Tompkins, W. J. (1985). A real-time QRS detection algo-
rithm. IEEE Transactions on Biomedical Engineering, 3, 230–236.
https://doi.org/10.1109/TBME.1985.325532
Perini, R., & Veicsteinas, A. (2003). HR variability and autonomic ac-
tivity at rest and during exercise in various physiological conditions.
European Journal of Applied Physiology, 90(3–4), 317–325. https://
doi.org/10.1007/s0042 1-003-0953-9
Plutchik, R. (1984). Emotions: A general psychoevolutionary theory.
Approaches to Emotion, 1984, 197–219. https://doi.org/10.1042/
bj219 0197
Porges, S. W. (1995). Cardiac vagal tone: A physiological index of
stress. Neuroscience & Biobehavioral Reviews, 19(2), 225–233.
https://doi.org/10.1016/0149-7634(94)00066 -A
Porges, S. W. (2007). The polyvagal perspective. Biological Psychology,
74(2), 116–143. https://doi.org/10.1016/j.biops ycho.2006.06.009
Roe, J., Thompson, C., Aspinall, P., Brewer, M., Duff, E., Miller, D.,
Mitchell, R., & Clow, A. (2013). Green space and stress: Evidence
from cortisol measures in deprived urban communities. International
Journal of Environmental Research and Public Health, 10(9),
4086–4103. https://doi.org/10.3390/ijerp h1009 4086
Rothberg, L. J., Lees, T., Clifton-Bligh, R., & Lal, S. (2016). Association
between heart rate variability measures and blood glucose lev-
els: Implications for noninvasive glucose monitoring for diabetes.
Diabetes Technology & Therapeutics, 18(6), 366–376. https://doi.
org/10.1089/dia.2016.0010
Sandercock, G. R., Bromley, P. D., & Brodie, D. A. (2005). Effects of
exercise on HR variability: Inferences from meta-analysis. Medicine
& Science in Sports & Exercise, 37(3), 433–439.
Segerstrom, S. C., & Miller, G. E. (2004). Psychological stress and
the human immune system: A meta-analytic study of 30 years
of inquiry. Psychological Bulletin, 130, 601–630. https://doi.
org/10.1037/0033-2909.130.4.601
Shiota, M. N., Neufeld, S. L., Yeung, W. H., Moser, S. E., & Perea, E. F.
(2011). Feeling good: Autonomic nervous system responding in five
positive emotions. Emotion, 11(6), 1368. https://doi.org/10.1037/
a0024278
Singh, J. P., Larson, M. G., O’Donnell, C. J., Wilson, P. F., Tsuji, H.,
Lloyd-Jones, D. M., & Levy, D. (2000). Association of hypergly-
cemia with reduced HR variability (The Framingham Heart Study).
The American Journal of Cardiology, 86(3), 309–312. https://doi.
org/10.1016/S0002 -9149(00)00920 -6
Smith, T. W., Cribbet, M. R., Nealey-Moore, J. B., Uchino, B. N.,
Williams, P. G., MacKenzie, J., & Thayer, J. F. (2011). Matters
of the variable heart: Respiratory sinus arrhythmia response to
marital interaction and associations with marital quality. Journal
of Personality and Social Psychology, 100(1), 103. https://doi.
org/10.1037/a0021136
Smith, T. W., Deits-Lebehn, C., Williams, P. G., Baucom, B. R., &
Uchino, B. N. (2020). Toward a social psychophysiology of vagally
mediated HR variability: Concepts and methods in self-regula-
tion, emotion, and interpersonal processes. Social and Personality
Psychology Compass, 14(3), e12516.
Task Force of the European Society of Cardiology. (1996). HR vari-
ability: Standards of measurement, physiological interpretation and
clinical use. Circulation, 93, 1043–1065. https://doi.org/10.1161/01.
CIR.93.5.1043
Thayer, J. F., Åhs, F., Fredrikson, M., Sollers, J. J. III, & Wager, T. D.
(2012). A meta-analysis of HR variability and neuroimaging stud-
ies: Implications for HR variability as a marker of stress and health.
Neuroscience & Biobehavioral Reviews, 36(2), 747–756. https://doi.
org/10.1016/j.neubi orev.2011.11.009
Thayer, J. F., Hansen, A. L., Saus-Rose, E., & Johnsen, B. H. (2009).
Heart rate variability, prefrontal neural function and cognitive per-
formance: The neurovisceral integration perspective on self-regu-
lation, adaptation, and health. Annals of Behavioral Medicine, 37,
141–153. https://doi.org/10.1007/s1216 0-009-9101-z
Tofighi, D., & MacKinnon, D. P. (2011). RMediation: An R pack-
age for mediation analysis confidence intervals. Behavior
Research Methods, 43(3), 692–700. https://doi.org/10.3758/s1342
8-011-0076-x
Tulppo, M. P., Makikallio, T. H., Takala, T. E., Seppanen, T. H. H. V.,
& Huikuri, H. V. (1996). Quantitative beat-to-beat analysis of HR
dynamics during exercise. American Journal of Physiology, 271(1),
H244–H252. https://doi.org/10.1152/ajphe art.1996.271.1.H244
Uchino, B. N., Cronan, S., Scott, E., Landvatter, J., & Papadakis, M.
(2020). Social support and stress, depression, and cardiovascular
disease. In Cardiovascular implications of stress and depression
(pp. 211–223). Academic Press. https://doi.org/10.1016/B978-0-12-
81501 5-3.00009 -X
Uchino, B. N., Smith, T. W., Holt-Lunstad, J., Campo, R., & Reblin, M.
(2007). Stress and illness. In J. T. Cacioppo, L. G. Tassinary, & G.
G. Berntson (Eds.), Handbook of psychophysiology (pp. 608–632).
Cambridge University Press. https://doi.org/10.1017/CBO97 80511
546396.026
Ulrich, R. S. (1984). View through a window may influence recovery
from surgery. Science, 224(4647), 420–421. https://doi.org/10.1126/
scien ce.6143402

14 of 14
|
SCOTT eT al.
Ulrich, R. S., Simons, R. F., Losito, B. D., Fiorito, E., Miles, M. A., &
Zelson, M. (1991). Stress recovery during exposure to natural and
urban environments. Journal of Environmental Psychology, 11(3),
201–230. https://doi.org/10.1016/S0272 -4944(05)80184 -7
Ulrich-Lai, Y. M., & Herman, J. P. (2009). Neural regulation of endo-
crine and autonomic stress responses. Nature Reviews Neuroscience,
10(6), 397. https://doi.org/10.1038/nrn2647
Watson, D., & Clark, L. A. (1999). The PANAS-X: Manual for the
positive and negative affect schedule-expanded form. https://doi.
org/10.1007/978-3-319-56782 -2_9007-1
White, M. P., Alcock, I., Wheeler, B. W., & Depledge, M. H. (2013).
Would you be happier living in a greener urban area? A fixed-ef-
fects analysis of panel data. Psychological Science, 24(6), 920–928.
https://doi.org/10.1177/09567 97612 464659
Wilson, E. O. (1984). Biophilia: The human bond with other species.
Harvard University Press.
Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of
stimulus to rapidity of habit-formation. Journal of Comparative
Neurology and Psychology, 18(5), 459–482. https://doi.org/10.1002/
cne.92018 0503
Yu, L., Buysse, D. J., Germain, A., Moul, D. E., Stover, A., Dodds,
N. E., Johnston, K. L., & Pilkonis, P. A. (2012). Development of
short forms from the PROMIS™ sleep disturbance and sleep-related
impairment item banks. Behavioral Sleep Medicine, 10(1), 6–24.
https://doi.org/10.1080/15402 002.2012.636266
How to cite this article: Scott EE, LoTemplio SB,
McDonnell AS, et al. The autonomic nervous system
in its natural environment: Immersion in nature is
associated with changes in heart rate and heart rate
variability. Psychophysiology. 2020;00:e13698.
https://doi.org/10.1111/psyp.13698