How Positive Emotions Build Physical Health: Perceived Positive Social Connections Account for the Upward Spiral Between Positive Emotions and Vagal Tone

Abstract

The mechanisms underlying the association between positive emotions and physical health remain a mystery. We hypothesize that an upward-spiral dynamic continually reinforces the tie between positive emotions and physical health and that this spiral is mediated by people's perceptions of their positive social connections. We tested this overarching hypothesis in a longitudinal field experiment in which participants were randomly assigned to an intervention group that self-generated positive emotions via loving-kindness meditation or to a waiting-list control group. Participants in the intervention group increased in positive emotions relative to those in the control group, an effect moderated by baseline vagal tone, a proxy index of physical health. Increased positive emotions, in turn, produced increases in vagal tone, an effect mediated by increased perceptions of social connections. This experimental evidence identifies one mechanism-perceptions of social connections-through which positive emotions build physical health, indexed as vagal tone. Results suggest that positive emotions, positive social connections, and physical health influence one another in a self-sustaining upward-spiral dynamic.

Full text

http://pss.sagepub.com/
Psychological Science
http://pss.sagepub.com/content/early/2013/05/06/0956797612470827
The online version of this article can be found at:
DOI: 10.1177/0956797612470827
published online 6 May 2013Psychological Science
Brantley and Barbara L. Fredrickson
Bethany E. Kok, Kimberly A. Coffey, Michael A. Cohn, Lahnna I. Catalino, Tanya Vacharkulksemsuk, Sara B. Algoe, Mary
Upward Spiral Between Positive Emotions and Vagal Tone
How Positive Emotions Build Physical Health : Perceived Positive Social Connections Account for the
Published by:
http://www.sagepublications.com
On behalf of:
Association for Psychological Science
can be found at:Psychological ScienceAdditional services and information for
http://pss.sagepub.com/cgi/alertsEmail Alerts:
http://pss.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
What is This?
- May 6, 2013OnlineFirst Version of Record >>
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

Psychological Science
XX(X) 1 –10
© The Author(s) 2013
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0956797612470827
pss.sagepub.com
Research Article
People who experience warmer, more upbeat emotions
live longer and healthier lives. Indeed, prospective evi-
dence connecting positive emotions to physical health
and longevity has steadily grown for a decade (for a meta-
analysis linking positive emotions to objective health out-
comes, see Howell, Kern, & Lyubomirsky, 2007; for a
meta-analysis linking positive emotions to mortality, see
Chida & Steptoe, 2008). Experiencing positive emotions
more frequently, for instance, forecasts having fewer colds
(Cohen, Alper, Doyle, Treanor, & Turner, 2006), reduced
inflammation (Steptoe, O’Donnell, Badrick, Kumari, &
Marmot, 2007), and lower likelihood of cardiovascular
disease (Boehm & Kubzansky, 2012). Comple menting this
prospective correlational evidence, a recent longitudinal
field experiment designed to test Fredrickson’s (1998, in
press) broaden-and-build theory of positive emotions
found that individuals randomly assigned to self-generate
positive emotions reported experiencing fewer headaches
470827PSSXXX10.1177/0956797612470827Kok et al.Positive Emotions, Social Connections, and Health
research-article2013
Corresponding Authors:
Barbara L. Fredrickson, Department of Psychology, University of
North Carolina at Chapel Hill, Davie Hall CB 3270, Chapel Hill, NC
27599-3270
E-mail: blf@unc.edu
Bethany E. Kok, Department of Social Neuroscience, Max Planck
Institute for Human Cognitive and Brain Sciences, Stephanstraße 1A,
04103 Leipzig, Germany
E-mail: bethkok@cbs.mpg.de
How Positive Emotions Build Physical
Health: Perceived Positive Social
Connections Account for the Upward Spiral
Between Positive Emotions and Vagal Tone
Bethany E. Kok1, Kimberly A. Coffey1, Michael A. Cohn2,
Lahnna I. Catalino1, Tanya Vacharkulksemsuk1, Sara B. Algoe1,
Mary Brantley3, and Barbara L. Fredrickson1
1Department of Psychology, University of North Carolina at Chapel Hill; 2Osher Center for Integrative
Medicine, University of California, San Francisco; and 3Duke Integrative Medicine, Duke Center for
Living Campus, Duke University
Abstract
The mechanisms underlying the association between positive emotions and physical health remain a mystery. We
hypothesize that an upward-spiral dynamic continually reinforces the tie between positive emotions and physical health
and that this spiral is mediated by people’s perceptions of their positive social connections. We tested this overarching
hypothesis in a longitudinal field experiment in which participants were randomly assigned to an intervention group
that self-generated positive emotions via loving-kindness meditation or to a waiting-list control group. Participants
in the intervention group increased in positive emotions relative to those in the control group, an effect moderated
by baseline vagal tone, a proxy index of physical health. Increased positive emotions, in turn, produced increases in
vagal tone, an effect mediated by increased perceptions of social connections. This experimental evidence identifies
one mechanism—perceptions of social connections—through which positive emotions build physical health, indexed
as vagal tone. Results suggest that positive emotions, positive social connections, and physical health influence one
another in a self-sustaining upward-spiral dynamic.
Keywords
emotions, social interaction, health, happiness, well-being
Received 7/16/12; Revision accepted 10/11/12
Psychological Science OnlineFirst, published on May 6, 2013 as doi:10.1177/0956797612470827
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

2 Kok et al.
and less chest pain, congestion, and weakness compared
with a control group (Fredrickson, Cohn, Coffey, Pek, &
Finkel, 2008). These first causal data lend support to the
conclusion suggested by prospective correlations: Positive
emotions build physical health. Stronger evidence still
would be to find that an experimental manipulation of
positive emotions influenced an objective marker of phys-
ical health. Providing such evidence was one aim of the
work reported here.
Cardiac vagal tone provided our objective proxy for
physical health. Indexed at rest as variability in heart rate
associated with respiratory patterns, vagal tone reflects
the functioning of the vagus nerve, which is the 10th
cranial nerve and a core component of the parasympa-
thetic nervous system, which regulates heart rate in
response to signals of safety and interest (Porges, 2007).
Low vagal tone has been linked to high inflammation
(Thayer & Sternberg, 2006), and lower vagal tone fore-
casts greater risk for myocardial infarction and lower
odds of survival after heart failure (Bibevski & Dunlap,
2011).
Intriguingly, recent prospective evidence suggests that
the causal link between positive emotions and physical
health may run in the opposite direction as well: Physical
health appears to promote positive emotions. Building
on findings that high vagal tone has been associated with
superior abilities to regulate one’s own emotions (Fabes
& Eisenberg, 1997; Thayer, Hansen, Saus-Rose, & Johnsen,
2009) and with positive emotionality (Oveis et al., 2009),
we found that people with higher vagal tone show
greater gains over time in their positive emotions (Kok &
Fredrickson, 2010). More strikingly, these same data also
revealed that people who show greater gains in positive
emotions show greater improvements over time in their
vagal tone; in short, positive emotions and vagal tone
show the reciprocal influence indicative of an upward-
spiral dynamic (Kok & Fredrickson, 2010). This prospec-
tive evidence not only challenges the view that vagal
tone in adulthood is a largely stable, traitlike attribute
(Bornstein & Suess, 2000), but also raises the possibility
that changes in habitual emotions drive changes in vagal
tone, and thereby constitute one pathway through which
emotional health influences physical health.
Still, the causal mechanisms that tie positive emotions
to vagal tone specifically and to physical health more
generally remain a mystery. We propose that people’s
ability to translate their own positive emotions into posi-
tive social connections with others may be one of the
keys to solving this mystery. Three lines of evidence sup-
port our logic.
First, laboratory experiments provide ample causal
evidence that positive emotions promote positive social
connections. For instance, compared with people in
affectively neutral control conditions, people randomly
assigned to experience positive emotions show greater
social engagement (Isen, 1970), social inclusiveness
(Dovidio, Gaertner, Isen, & Lowrance, 1995), individuat-
ing other-focus (Johnson & Fredrickson, 2005), perspec-
tive taking (Nelson, 2009), self-disclosure (Cunningham,
1988), interpersonal trust (Dunn & Schweitzer, 2005), and
compassion (Nelson, 2009). Complementing these labo-
ratory experiments, the same longitudinal field experi-
ment that established a causal link between people’s
positive emotions and their subsequent self-reported
physical health established a similar causal link between
positive emotions and perceived positive social connec-
tions. Plus, the more time people devoted to generating
positive emotions in themselves, the more pleasant their
interactions with others became (Fredrickson et al.,
2008).
Second, a long-standing corpus of prospective evi-
dence shows that having more diverse and rewarding
social relationships robustly forecasts better physical
health and greater longevity. Indeed, a recent meta-ana-
lytic review of 148 studies (representing more than
300,000 individuals) concluded that the influence of
social integration on mortality risk is comparable in mag-
nitude to that of other well-established risk factors, such
as smoking, excessive alcohol consumption, obesity, and
lack of physical activity (Holt-Lunstad, Smith, & Layton,
2010). For instance, perceiving oneself as enmeshed
within a variety of social relationships prospectively pre-
dicts reduced susceptibility to cardiovascular disease
(e.g., Kaplan et al., 1988), cancer (e.g., Welin, Larsson,
Svärdsudd, Tibblin, & Tibblin, 1992), and various infec-
tions (Cohen, Doyle, Skoner, Rabin, & Gwaltney, 1997).
Considerable work has also linked loneliness, or self-per-
ceived lack of social connections, to ill health (particu-
larly cardiovascular disease) and ill-being (for a review,
see Hawkley & Cacioppo, 2010).
Third, our focus on positive social connections is fur-
ther inspired by Porges’s (2007) polyvagal theory, which
identifies the myelinated vagus as central to the mamma-
lian social-engagement system. The vagus nerve, for
instance, is anatomically linked to nerves involved in
coordinating eye gaze, generating facial expressions, and
tuning the ear to the frequency of the human voice
(Porges, 2007)—functions that are critical for social affili-
ative behavior. Supporting this theory, studies have found
that higher vagal tone is associated with more prosocial
behavior (Fabes, Eisenberg, & Eisenbud, 1993) and social
closeness (Kok & Fredrickson, 2010). Moreover, experi-
mental evidence shows that vagal activation is greater in
the presence of supportive, compared with ambivalent,
friendships (Holt-Lunstad, Uchino, Smith, & Hicks, 2007)
and that vagal activation is increased by intranasal expo-
sure to oxytocin, a key neuropeptide that regulates social
engagement (Kemp et al., 2012). Perceiving oneself as
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

Positive Emotions, Social Connections, and Health 3
socially connected is critical to autonomic regulation
more generally (Beckes & Coan, 2011). Moreover, consis-
tent with Porges’s polyvagal theory, the study that dem-
onstrated reciprocal prospective ties between positive
emotions and vagal tone also demonstrated a similar pat-
tern of reciprocal relations between perceived positive
social connections and vagal tone (Kok & Fredrickson,
2010).
Looking across these three separate strands of evi-
dence, one sees that positive emotions build positive
social connections and that positive social connections
are linked reciprocally to both physical health in general
and vagal tone in particular. Integrating this evidence, we
postulate that a self-sustaining upward-spiral dynamic
continually reinforces the tie between positive emotions
and physical health. Specifically, we posit that vagal tone,
by virtue of its association with superior emotion regula-
tion, supports people’s abilities to self-generate positive
emotions. Positive emotions in turn promote perceived
positive social connections, which in turn promote
improved physical health, as indexed by increases in
vagal tone.
The current investigation was motivated by the follow-
ing questions: Can people willfully harness this upward-
spiral dynamic to steer themselves toward greater physical
health? That is, can people’s efforts to self-generate posi-
tive emotions improve their vagal tone? If so, are per-
ceived positive social connections a mechanism through
which this health benefit is achieved? We designed a lon-
gitudinal field experiment, spanning more than 2 months,
to attempt to budge the proposed upward spiral and test
whether and how positive emotions build physical
health, indexed objectively as vagal tone. By also includ-
ing a pretest of vagal tone, we investigated whether a
higher level of this health marker positions people for
greater success in self-generating positive emotions and
securing attendant benefits.
Figure 1 portrays the conceptual model that unites the
three hypotheses we tested. Although past work has
shown reciprocal links between each pair of constructs
represented in this spiral model, the novel contribution
of the present work is that it experimentally tested the
causal link between positive emotions and improved
vagal tone, as mediated by positive social connections.
Our hypotheses were as follows:
• Hypothesis 1: Individuals with higher vagal tone
show greater changes in positive emotions when
randomly assigned to positive-emotions training.
• Hypothesis 2: To the extent that random assignment
to positive-emotions training produces changes in
positive emotions, it also produces increases in per-
ceived positive social connections.
• Hypothesis 3: To the extent that random assign-
ment to positive-emotions training increases per-
ceived positive social connections, it increases vagal
tone. Specifically, changes in positive social con-
nections mediate the impact of assignment to train-
ing and increases in positive emotions on vagal
tone.
We tested these hypotheses by randomly assigning study
volunteers either to learn to self-generate positive emo-
tions through the ancient mind-training practice of loving-
kindness meditation (LKM) or to serve in a waiting-list
control group. LKM teaches individuals how to cultivate
positive emotions toward themselves and others, and
past research has documented that it increases positive
emotions, yielding attendant improvements in perceived
positive social connections and self-reported physical
health (Fredrickson et al., 2008).
Method
Participants
Participants were faculty and staff of the University of
North Carolina at Chapel Hill. Recruitment materials
referred to the benefits of meditation for relieving stress
and pain, but did not specifically mention LKM or antici-
pated effects on positive emotions or social experiences.
Seventy-one university employees gave their consent
to participate; 6 were ultimately excluded, 5 for failure to
attend the meditation workshops and 1 for previous
meditation experience, which violated the exclusion cri-
terion. The Supplemental Material available online reports
an intent-to-treat analysis incorporating all participants; it
yielded a pattern of results identical to those described
here.
Of our final 65 participants, 66% were female, and 34%
were male; the majority (83%) were White. The median
Positive
Emotions
Perceived Social
Connections
Autonomic
Regulation
(i.e., vagal tone)
Experimental
Intervention
Fig. 1. Conceptual model describing the relationships among the
experimental intervention, vagal tone, positive emotions, and social
connections.
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

4 Kok et al.
age of this sample was 37.5 years. Forty-eight percent
were married or otherwise in a committed relationship.
Chi-square and t tests did not reveal any significant or
marginally significant differences between the experi-
mental conditions in gender, race, age, education,
income, or marital status.
Loving-kindness meditation
LKM is a contemplative practice that focuses on self-gen-
erating feelings of love, compassion, and goodwill toward
oneself and others. The LKM workshop offered in this
study followed the same training as the workshops in our
past work (Fredrickson et al., 2008). It was taught by one
of the authors (M. B.), a licensed therapist with training
in meditation instruction. Participants in the intervention
condition attended one hour-long class per week for
6 weeks. Each class involved guided meditation practice,
as well as discussion of how to maintain a meditative
practice and how LKM can be helpful in dealing with
everyday situations. Participants were asked to practice
meditation at home, ideally daily, but it was made clear
that the frequency was up to them. Practice could be self-
guided or guided by a recording provided by the instruc-
tor.1 Participants in the control condition were told that
they were on a waiting list; after the study ended, they
were given the opportunity to receive meditation training
from M. B.
Daily assessment of meditation
practice, emotions, and social
connection
Each day, for 61 consecutive days, participants in both
conditions reported the amount of time (in minutes)
that they had engaged in “meditation, prayer, or solo
spiritual activity” since the last time they had provided a
daily report. They then rated their most powerful expe-
riences of 20 different emotions within the past day,
using a 5-point scale (1 = not at all, 5 = extremely).
These included 9 positive emotions (i.e., amusement,
awe, gratitude, hope, interest, joy, love, pride, and
serenity) and 11 negative emotions (i.e., anger, bore-
dom, contempt, disgust, embarrassment, fear, guilt,
hatred, sadness, shame, and stress). Daily Cronbach’s α
coefficients for the 9 positive-emotion items ranged
from .89 to .97 over the 61 days (M = .93, SD = .015);
thus, as in past work (Fredrickson et al., 2008), these
items cohered into one factor. Accordingly, we averaged
ratings for these 9 items to create a daily positive-
emotions score (M = 2.87, SE = 0.09). Likewise, daily
Cronbach’s α coefficients for the 11 negative-emotion
items ranged from .75 to .93 (M = .86, SD = .038), so we
averaged ratings for these 11 items to create a daily
negative-emotions score (M = 1.74, SE = 0.06).
Next, participants considered the three social interac-
tions in which they had spent the most time that day.
They then rated these three interactions in aggregate
using two items adapted from Russell’s (1996) UCLA
Loneliness Scale: “During these social interactions, I felt
‘in tune’ with the person/s around me” and “During these
social interactions, I felt close to the person/s.” Responses
were made on a 7-point scale (1 = not at all true, 7 = very
true). The daily Cronbach’s α for these two items ranged
from .80 to .98 (M = .94, SD = .03). Accordingly, we aver-
aged responses to the items to create a daily social-
connections score (M = 4.89, SE = 0.15)
Vagal tone
Vagal tone was assessed using spectral frequency analy-
sis of heart rate (HR) data to obtain high-frequency heart
rate variability (HF-HRV). Data were collected for 2 min
at rest, with continuous recording at 1000 Hz, using dis-
posable snap electrodes in a bipolar configuration on
opposite sides of the chest. The raw HR recordings were
preprocessed and manually edited to correct for artifacts.
Customized software by James Long Company (Caroga
Lake, NY) employed discrete Fourier transforms to extract
the high-frequency components of the HR signal (0.12–
0.4 Hz) that primarily reflect vagal influences on the
heart. To assess stability of measurement, we calculated
the correlations between the first and last 60 s of the
recording (start of study: r = .77, p < .0001; end of study:
r = .90, p < .0001). Analyses were also conducted using a
different measure of vagal tone, one based on respiratory
sinus arrhythmia (RSA; Porges, 2007), a measure that
combines HR with respiration data. Although the pattern
of results was largely similar for the two measures, we
report results for HF-HRV here because overall fit for the
hypothesized model was better for HF-HRV than for RSA.
Detailed results for models using RSA are available in the
Supplemental Material.
Procedure
Participants visited our laboratory for a baseline measure
of vagal tone. They were then given access to a secure
Web site for daily reporting of time spent meditating,
positive emotions, and social connections. Daily reports
were made for 9 weeks, spanning a 2-week baseline
period, random assignment to conditions, the 6-week
meditation workshop, and 1 week after the end of the
workshop. At the end of the 9 weeks, participants made
a second and final laboratory visit so that we could obtain
a second measure of vagal tone.
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

Positive Emotions, Social Connections, and Health 5
Results
Hypothesized model
Preliminary tests, described in the Supplemental Material
available online, demonstrated that the experimental
intervention (i.e., the LKM workshop) produced increases
in positive emotions, perceived social connections, and
vagal tone relative to the control condition. These analy-
ses, however, did not test the mediational relationships
hypothesized to underlie the causal chain among these
constructs (Fig. 1). A variant of a mediational, parallel-
process, latent-curve model (Cheong, MacKinnon, &
Khoo, 2003) was used to test the full hypothesized model.
The model was estimated using maximum likelihood
estimation and all available data.
In this model (Fig. 2), both positive emotions and
social connections were modeled as latent curves, with
their intercept and slope factors loading on the corre-
sponding weekly composite scores. All intercept factor
loadings were fixed to 1, and slope factor loadings were
fixed to week in the study, beginning with 0 for the first
baseline week. In addition, the residual error for each
weekly rating of positive emotions was allowed to cor-
relate with the residual error for that week’s rating of
social connections, to reflect the within-week relation-
ship between these constructs.
The model produced a root-mean-square error of
approximation of 0.078 (95% confidence interval = [0.056–
0.098]) and a comparative-fit index of .95, which placed
model fit in the acceptable range. Results supported all
PE w1 PE w3
PE w2 PE w8
PE w6PE w4 PE w5 PE w7 PE w9
SC w1 SC w2 SC w3 SC w4 SC w5 SC w6 SC w7 SC w8 SC w9
Positive-
Emotions
Intercept
Positive-
Emotions
Slope
Social-
Connections
Intercept
Social-
Connections
Slope
Experimental Condition Baseline Vagal Tone
Change in Vagal Tone
a
b
c
g h
jk
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
δ
d
e
f
Fig. 2. Parallel-process mediational model depicting the impact of experimental condition (loving-kindness meditation vs. control condition),
baseline vagal tone, and their interaction on positive emotions (PE); of PE on social connections (SC); and of SC on vagal tone. Black lines repre-
sent hypothesized relationships, solid gray lines represent anticipated significant replications of the literature, and dotted gray lines represent paths
expected to be nonsignificant. The labels “w1” through “w9” refer to the 9 weeks of the study. For convenience, paths referred to in this article and
in the Supplemental Material are labeled by italicized letters a through j.
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

6 Kok et al.
three of our hypotheses. (We provide results for nonhy-
pothesized paths in the Supplemental Material.) As pre-
dicted by Hypothesis 1, the interaction of experimental
condition and baseline vagal tone significantly predicted
slope of change in positive emotions (path e in Fig. 2; b =
0.043, z = 2.63, p = .009); in the LKM group, participants
who entered the study with higher vagal tone exhibited
steeper increases in positive emotions over the course of
the study (Fig. 3). As predicted by Hypothesis 2, slope of
change in positive emotions significantly and positively
predicted slope of change in social connections (path h in
Fig. 2; b = 1.04, z = 4.12, p < .001; see Fig. 4). Finally, as
predicted by Hypothesis 3, slope of change in social con-
nections in turn positively predicted change in vagal tone
(path k in Fig. 2; b = 4.90, z = 2.14, p = .03; see Fig. 5).
Thus, participants who reported greater increases in posi-
tive emotions over the course of the study, who were
mostly the ones randomly assigned to the LKM group, also
exhibited greater increases in social connections, which
were in turn associated with larger increases in vagal tone.
Alternative models
To rule out alternative hypotheses, we estimated five
additional models, the statistical details of which are pro-
vided in the Supplemental Material. First, we explored
whether the findings could be explained by a reduction
in negative emotions rather than an increase in positive
emotions. Although assignment to the intervention con-
dition significantly reduced negative emotions, this effect
was neither predicted by baseline vagal tone nor predic-
tive of change in vagal tone. Thus, the negative-emotions
model failed to explain the impact of physical health on
participants’ responsiveness to LKM and LKM’s subse-
quent impact on physical health.
Second, we explored whether positive emotions and
social connections were interchangeable in the model.
Although experimental condition significantly predicted
change in social connections and change in social con-
nections predicted change in positive emotions, critical
paths from and to vagal tone were not significant. As was
the case with the negative-emotions model, this trans-
posed model failed to link physical health to other critical
variables and did not exhibit the hypothesized media-
tional paths.
Third, we explored whether change in vagal tone
might have driven the changes in positive emotions and,
in turn, social connections. Although experimental condi-
tion significantly predicted change in vagal tone, change
in vagal tone did not predict change in positive emotions.
Thus, the third model also failed to effectively link all of
the constructs or to demonstrate the hypothesized
upward spiral.
Fourth, we explored whether change in positive emo-
tions was a necessary mediator. We discovered that when
we prevented positive emotions from mediating, model
fit was marginally significantly worse. Similarly, with our
fifth model, we explored whether change in social con-
nections was a necessary mediator. We discovered that
when we prevented social connections from mediating,
model fit was significantly worse. Thus, the fourth and
fifth models support the importance of positive emotions
and social connections as mediators in the model.
Taken cumulatively, the alternative models exhibited
isolated significant paths. However, none of these models
accounted for the full sequence of relationships as com-
prehensively as our hypothesized upward-spiral model,
as represented conceptually in Figure 1 and statistically
in Figure 2.
Discussion
These findings document not only that positive emotions
build physical health, as indexed objectively by cardiac
vagal tone, but also how they do so: We found that peo-
ple’s perceptions of their positive social connections with
others accounted for the causal link between positive
emotions and improved vagal tone. Supporting the con-
ceptual model depicted in Figure 1, the data suggest that
positive emotions, positive social connections, and physi-
cal health forge an upward-spiral dynamic. Baseline
vagal tone interacted with experimental condition to pre-
dict the degree of success people had in self-generating
positive emotions. Greater positive emotions in turn
prompted individuals to see themselves as more socially
connected. Over time, as moments of positive emotions
and positive social connections increased and accrued,
–1 0 1 2 3
Baseline Vagal Tone (Standardized)
–0.05
0.00
0.05
0.10
0.15
0.20
Slope of Change in Positive Emotions
Intervention
Control
Fig. 3. Relationship between baseline vagal tone and change in positive
emotions over the course of the study, for participants in the control and
intervention conditions. The shaded areas represent the 95% confidence
limits for mean predicted values.
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

Positive Emotions, Social Connections, and Health 7
vagal tone also improved, building a biological resource
that has been linked to numerous health benefits. This
upward-spiral dynamic has the potential to set individu-
als on self-sustaining pathways toward growth that can
explain the strong empirical associations between posi-
tive social and emotional experiences and physical
health. Indeed, these findings suggest that habitually
experienced positive emotions may be an essential psy-
chological nutrient for autonomic health.
Vagal theorists have represented vagal tone as a stable,
traitlike measure of autonomic regulation associated with
various downstream indicators, such as cardiovascular
health, social acuity, and regulation of cognition, emo-
tions, and physiological systems (Porges, 2007; Thayer &
Sternberg, 2006). To our knowledge, the present findings
are the first to show that although vagal tone is largely
stable, it can also be improved through sustained enhance-
ments in an individual’s emotions and social perceptions.
Quartile for Change in Positive Emotions Quartile for Change in Positive Emotions
Control Condition Intervention Condition
Number of Participants
Quartile for Change in Social Connections
Fig. 4. Distribution of participants in the control and intervention conditions according to their
quartiles for change in positive emotions and in social connections over the course of the study.
Quartiles represent model-derived slopes and are numbered such that 1 represents the lowest
quartile and 4 represents the highest quartile across the sample as a whole.
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

8 Kok et al.
Control Condition Intervention Condition
Quartile for Change in Social Connections Quartile for Change in Social Connections
Quartile for Change in Vagal Tone
Number of Participants
Fig. 5. Distribution of participants in the control and intervention conditions according to their quartiles
for change in social connections and in vagal tone over the course of the study. Quartiles represent
model-derived slopes and are numbered such that 1 represents the lowest quartile and 4 represents the
highest quartile across the sample as a whole.
Strengths of this work include its experimental design,
which included repeated measures to assess change in
targeted constructs, as well as the use of an objective
marker of physical health. Limitations include the unique
sample of participants motivated for self-improvement
and the reliance on one technique, LKM, for self-genera-
tion of positive emotions. Beyond testing how well these
findings generalize across other samples, other emotion-
change techniques, and other comparison groups, future
work could include other objective markers of physical
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

Positive Emotions, Social Connections, and Health 9
health, other plausible psychological mediators (e.g.,
broadened awareness, optimism), or objective measures
of change in social or health behaviors (Kok, Waugh, &
Fredrickson, in press).
Most dispensed advice about how people might
improve their physical health calls for increased physical
activity, improved nutritional intake, and reductions in
tobacco and alcohol use. This good advice can now be
expanded to include self-generating positive emotions.
Recurrent momentary experiences of positive emotions
appear to serve as nutrients for the human body, increas-
ing feelings of social belonging and giving a needed
boost to parasympathetic health, which in turn opens
people up to more and more rewarding positive emotional
and social experiences. Over time, this self-sustaining
upward spiral of growth appears to improve physical
health.
Acknowledgments
Bethany E. Kok is now at the Department of Social Neuroscience,
Max Planck Institute for Human Cognitive and Brain Sciences.
We thank the chairs and directors of multiple units within the
University of North Carolina at Chapel Hill, who provided
release time for their faculty and staff to participate in this
study. Special thanks go to the participants, who devoted con-
siderable time and effort to this project. We also thank Aly Light
for overseeing data collection and Ann Firestine for providing
editorial assistance.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with
respect to their authorship or the publication of this article.
Funding
This work was supported by a National Institute of Mental
Health Grant (MH59615) awarded to Barbara L. Fredrickson.
Supplemental Material
Additional supporting information may be found at http://pss
.sagepub.com/content/by/supplemental-data
Note
1. Detailed information on the meditations and workshops is
available by request from Mary Brantley (mmbrantley04@gmail
.com).
References
Beckes, L., & Coan, J. A. (2011). Social Baseline Theory: The
role of social proximity in emotion and economy of action.
Social & Personality Psychology Compass, 5, 976–988.
Bibevski, S., & Dunlap, M. E. (2011). Evidence for impaired
vagus nerve activity in heart failure. Heart Failure Reviews,
16, 129–135.
Boehm, J. K., & Kubzansky, L. D. (2012). The heart’s content:
The association between positive psychological well-being
and cardiovascular health. Psychological Bulletin, 138,
655–691.
Bornstein, M. H., & Suess, P. E. (2000). Child and mother
cardiac vagal tone: Continuity, stability, and concordance
across the first 5 years. Developmental Psychology, 36,
54–65.
Cheong, J., MacKinnon, D. P., & Khoo, S. T. (2003). Investigation
of mediational processes using parallel process latent
growth curve modeling. Structural Equation Modeling: A
Multidisciplinary Journal, 10, 238–262.
Chida, Y., & Steptoe, A. (2008). Positive psychological well-
being and mortality: A quantitative review of prospective
observational studies. Psychosomatic Medicine, 70, 741–
756.
Cohen, S., Alper, C. M., Doyle, W. J., Treanor, J. J., & Turner,
R. B. (2006). Positive emotional style predicts resistance
to illness after experimental exposure to rhinovirus or
Influenza A virus. Psychosomatic Medicine, 68, 809–815.
Cohen, S., Doyle, W. J., Skoner, D. P., Rabin, B. S., & Gwaltney,
J. J. M. (1997). Social ties and susceptibility to the common
cold. Journal of the American Medical Association, 277,
1940–1944.
Cunningham, M. R. (1988). Does happiness mean friendliness?
Induced mood and heterosexual self-disclosure. Personality
and Social Psychology Bulletin, 14, 283–297.
Dovidio, J. F., Gaertner, S. L., Isen, A. M., & Lowrance, R.
(1995). Group representations and intergroup bias: Positive
affect, similarity, and group size. Personality and Social
Psychology Bulletin, 21, 856–865.
Dunn, J. R., & Schweitzer, M. E. (2005). Feeling and believing:
The influence of emotion on trust. Journal of Personality
and Social Psychology, 88, 736–748.
Fabes, R. A., & Eisenberg, N. (1997). Regulatory control and
adults’ stress-related responses to daily life events. Journal
of Personality and Social Psychology, 73, 1107–1117.
Fabes, R. A., Eisenberg, N., & Eisenbud, L. (1993). Behavioral
and physiological correlates of children’s reactions to oth-
ers in distress. Developmental Psychology, 29, 655–663.
Fredrickson, B. L. (1998). What good are positive emotions?
Review of General Psychology, 2, 300–319.
Fredrickson, B. L. (in press). Positive emotions broaden and
build. In E. A. Plant & P. G. Devine (Eds.), Advances in
experimental social psychology (Vol. 47). San Diego, CA:
Academic Press.
Fredrickson, B. L., Cohn, M. A., Coffey, K. A., Pek, J., & Finkel,
S. M. (2008). Open hearts build lives: Positive emotions,
induced through loving-kindness meditation, build con-
sequential personal resources. Journal of Personality and
Social Psychology, 95, 1045–1062.
Hawkley, L. C., & Cacioppo, J. T. (2010). Loneliness mat-
ters: A theoretical and empirical review of consequences
and mechanisms. Annals of Behavioral Medicine, 40,
218–227.
Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social
relationships and mortality risk: A meta-analytic review.
PLoS Medicine, 7, e1000316. Retrieved from http://www
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from

10 Kok et al.
.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal
.pmed.1000316
Holt-Lunstad, J., Uchino, B. N., Smith, T. W., & Hicks, A. (2007).
On the importance of relationship quality: The impact of
ambivalence in friendships on cardiovascular functioning.
Annals of Behavioral Medicine, 33, 278–290.
Howell, R. T., Kern, M. L., & Lyubomirsky, S. (2007). Health
benefits: Meta-analytically determining the impact of well-
being on objective health outcomes. Health Psychology
Review, 1, 83–136.
Isen, A. M. (1970). Success, failure, attention and reaction to
others: The warm glow of success. Journal of Personality
and Social Psychology, 15, 294–301.
Johnson, K. J., & Fredrickson, B. L. (2005). “We all look
the same to me”: Positive emotions eliminate the own-race
bias in face recognition. Psychological Science, 16, 875–
881.
Kaplan, G. A., Salonen, J. T., Cohen, R. D., Brand, R. J., Syme,
S. L., & Puska, P. (1988). Social connections and mor-
tality from all causes and from cardiovascular disease:
Prospective evidence from Eastern Finland. American
Journal of Epidemiology, 128, 370–380.
Kemp, A. H., Quintana, D. S., Kuhnert, R., Griffiths, K., Hickie,
I. B., & Guastella, A. J. (2012). Oxytocin increases heart
rate variability in humans at rest: Implications for social
approach-related motivation and capacity for social engage-
ment. PLoS ONE, 7, e44014. Retrieved from http://www
.plosone.org/article/info:doi/10.1371/journal.pone.0044014
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, 432–
436.
Kok, B. E., Waugh, C. E., & Fredrickson, B. L. (in press).
Meditation and health: The search for mechanisms of
action. Social & Personality Psychology Compass.
Nelson, D. W. (2009). Feeling good and open-minded: The
impact of positive affect on cross cultural empathic
responding. Journal of Positive Psychology, 4, 53–63.
Oveis, C., Cohen, A. B., Gruber, J., Shiota, M. N., Haidt, J.,
& Keltner, D. (2009). Resting respiratory sinus arrhythmia
is associated with tonic positive emotionality. Emotion, 9,
265–270.
Porges, S. W. (2007). The polyvagal perspective. Biological
Psychology, 74, 116–143.
Russell, D. W. (1996). UCLA Loneliness Scale (Version 3):
Reliability, validity, and factor structure. Journal of Personality
Assessment, 66, 20–40.
Steptoe, A., O’Donnell, K., Badrick, E., Kumari, M., & Marmot,
M. (2007). Neuroendocrine and inflammatory factors asso-
ciated with positive affect in healthy men and women: The
Whitehall II Study. American Journal of Epidemiology, 167,
96–102.
Thayer, J. F., Hansen, A. L., Saus-Rose, E., & Johnsen, B. H.
(2009). Heart rate variability, prefrontal neural function, and
cognitive performance: The neurovisceral integration per-
spective on self-regulation, adaptation, and health. Annals of
Behavioral Medicine, 37, 141–153.
Thayer, J. F., & Sternberg, E. (2006). Beyond heart rate variabil-
ity: Vagal regulation of allostatic systems. Annals of the New
York Academy of Sciences, 1088, 361–372.
Welin, L., Larsson, B., Svärdsudd, K., Tibblin, B., & Tibblin, G.
(1992). Social network and activities in relation to mortality
from cardiovascular diseases, cancer and other causes: A 12
year follow up of the study of men born in 1913 and 1923.
Journal of Epidemiology & Community Health, 46, 127–132.
at Max Planck Society on May 8, 2013pss.sagepub.comDownloaded from