Protocol for an experimental investigation of the roles of oxytocin and social support in neuroendocrine, cardiovascular, and subjective responses to stress across age and gender.

Abstract

.
Substantial empirical evidence has demonstrated that individuals who are socially isolated or have few positive social connections seem to age at a faster rate and have more chronic diseases. Oxytocin is a neurohypophyseal hormone hypothesized to coordinate both the causes and effects of positive social interactions, and may be involved in positive physiological adaptations such as buffering the deleterious effects of stress and promoting resilience. The proposed research will examine whether and how oxytocin influences responses to stress in humans and will consider effects in relation to those of social support.
Experimental research will be used to determine whether exogenously administered oxytocin (intranasal) influences psychological and physiological outcomes under conditions of stress across gender and age in adulthood. Hypotheses to be tested are: 1) Oxytocin ameliorates the deleterious neuroendocrine, cardiovascular, and subjective effects of stress; 2) Oxytocin and social support have similar and additive stress-buffering effects; 3) Oxytocin effects are stronger in women versus men; and 4) Oxytocin effects are similar across a range of adult ages. Hypotheses will be tested with a placebo-controlled, double-blind study using a sample of healthy men and women recruited from the community. Participants are randomly assigned to receive either oxytocin or placebo. They undergo a social stress manipulation with and without social support (randomly assigned), and outcome measures are obtained at multiple times during the procedure.
Understanding the determinants of healthy aging is a major public health priority and identifying effective measures to prevent or delay the onset of chronic diseases is an important goal. Experimental research on oxytocin, social relationships, and health in adulthood will contribute to the scientific knowledge base for maximizing active life and health expectancy. At conclusion of the study we will have solid evidence concerning the effects of oxytocin on stress response and whether it has similar effects across age and gender groups. A neurobiological understanding of resilience can inform efforts for both prevention and intervention of diseases or problems common in later life.

Full text

Page 1

BioMed Central
Page 1 of 17
(page number not for citation purposes)
BMC Public Health
Open Access
Study protocol
Protocol for an experimental investigation of the roles of oxytocin
and social support in neuroendocrine, cardiovascular, and
subjective responses to stress across age and gender
Laura D Kubzansky*1, Wendy B Mendes2, Allison Appleton1, Jason Block3
and Gail K Adler4
Address: 1Department of Society, Human Development, and Health, Harvard School of Public Health, Boston, MA USA, 2Department of
Psychology, Harvard University, Cambridge, MA USA, 3Department of Population Medicine, Harvard Medical School/Harvard Pilgrim Health
Care Institute, Boston, MA USA and 4Brigham and Women's Hospital Department of Medicine, Division of Endocrinology, Brigham and Women's
Hospital, Boston, MA USA
Email: Laura D Kubzansky* - lkubzans@hsph.harvard.edu; Wendy B Mendes - wbm@wjh.harvard.edu;
Allison Appleton - appleton@hsph.harvard.edu; Jason Block - Jason_Block@harvardpilgrim.org; Gail K Adler - gadler@partners.org
* Corresponding author
Abstract
Background: Substantial empirical evidence has demonstrated that individuals who are socially isolated or have
few positive social connections seem to age at a faster rate and have more chronic diseases. Oxytocin is a
neurohypophyseal hormone hypothesized to coordinate both the causes and effects of positive social
interactions, and may be involved in positive physiological adaptations such as buffering the deleterious effects of
stress and promoting resilience. The proposed research will examine whether and how oxytocin influences
responses to stress in humans and will consider effects in relation to those of social support.
Methods/Design: Experimental research will be used to determine whether exogenously administered oxytocin
(intranasal) influences psychological and physiological outcomes under conditions of stress across gender and age
in adulthood. Hypotheses to be tested are: 1) Oxytocin ameliorates the deleterious neuroendocrine,
cardiovascular, and subjective effects of stress; 2) Oxytocin and social support have similar and additive stress-
buffering effects; 3) Oxytocin effects are stronger in women versus men; and 4) Oxytocin effects are similar across
a range of adult ages. Hypotheses will be tested with a placebo-controlled, double-blind study using a sample of
healthy men and women recruited from the community. Participants are randomly assigned to receive either
oxytocin or placebo. They undergo a social stress manipulation with and without social support (randomly
assigned), and outcome measures are obtained at multiple times during the procedure.
Discussion: Understanding the determinants of healthy aging is a major public health priority and identifying
effective measures to prevent or delay the onset of chronic diseases is an important goal. Experimental research
on oxytocin, social relationships, and health in adulthood will contribute to the scientific knowledge base for
maximizing active life and health expectancy. At conclusion of the study we will have solid evidence concerning
the effects of oxytocin on stress response and whether it has similar effects across age and gender groups. A
neurobiological understanding of resilience can inform efforts for both prevention and intervention of diseases or
problems common in later life.
Trial registration: Clinical trial identification number is NCT01011465.
Published: 21 December 2009
BMC Public Health 2009, 9:481doi:10.1186/1471-2458-9-481
Received: 30 October 2009
Accepted: 21 December 2009
This article is available from: http://www.biomedcentral.com/1471-2458/9/481
© 2009 Kubzansky et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Page 2

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 2 of 17
(page number not for citation purposes)
Background
The biological underpinnings for how positive social rela-
tionships may promote health are not well understood.
Building on human and animal research that finds early-
life nurturing crucial for a range of social, behavioral,
physiological, and development outcomes across the
lifespan, investigators have hypothesized that social con-
nectedness affects longevity by influencing the rate of
aging of the organism. Support for this idea is provided by
a large body of epidemiologic evidence linking positive
social relationships with greater longevity and delays in
functional decline. Oxytocin is a neurohypophyseal hor-
mone hypothesized to coordinate both the causes and
effects of positive social interactions. Animal studies have
clearly demonstrated a role for oxytocin in the central
nervous system whereby it reduces stress-related activa-
tion; it inhibits sympathetic nervous system activity and
increases parasympathetic activity, and inhibits release of
glucocorticoids. There is growing interest in whether the
oxytocin system is involved in both social behavior and
positive physiologic adaptations related to growth and
restorative processes [1-5].
Research on this topic is limited. Much of the work inves-
tigating the effects of oxytocin on physiology and behav-
ior in relation to social connectedness as well as stress has
been conducted in animals. In addition to their primary
findings, these studies have suggested that oxytocin is
more potent in the presence of higher estrogen levels,
leading investigators to hypothesize that effects of oxy-
tocin are stronger in females than in males. Serious meth-
odological barriers in assessment and measurement of
oxytocin in humans present challenges to working in this
area. As a result, relevant empirical evidence for physio-
logical and behavioral effects of oxytocin in human pop-
ulations is still scarce. Of the studies in humans, many
have been conducted in samples of either all men or all
women with restricted age ranges. Thus, few of these stud-
ies have had the opportunity to consider effects of oxy-
tocin in relation to age, or to compare effects of oxytocin
on stress and positive social interaction across men and
women. This paper describes an experimental protocol in
humans that addresses many of these limitations. The
protocol is designed to determine whether oxytocin
reduces stress-related neuroendocrine and autonomic
activation and subjective distress across age and gender,
and to examine the effects of oxytocin in relation to those
of social support. To make the relevant comparisons pos-
sible, the protocol includes men and women across a
broad age range. This protocol also breaks through prior
methodological barriers that have made it particularly dif-
ficult for U.S. investigators to conduct experimental
research with oxytocin in humans, by identifying tech-
niques for using intranasal (exogenously administered)
oxytocin in a rigorously controlled laboratory setting with
human participants.
Why are Social Relationships Beneficial?
Substantial empirical evidence has demonstrated that
individuals who are socially isolated or have few positive
social connections seem to age at a faster rate and show
evidence of greater physiologic damage over time than
those with more positive social connections [6,7]. Much
research has focused on negative aspects of relationships,
i.e., conflict within or loss of a significant relationship [8-
11]. Other stressful aspects of relationships have also been
examined. For example, in one study, among mothers of
chronically ill children, those who had been caregiving for
longer had shorter telomere length, lower telomerase (a
cellular enzyme that protects against shortening of tel-
omere-DNA protein complexes) activity, and greater oxi-
dative stress, all markers of accelerated aging [12]. Less
work has looked directly at positive aspects of social rela-
tionships. Research in this area has broadly suggested that
positive feelings generated by close social ties promote
health across the life course. For example, in a prospective
study of healthy undergraduate men, those who reported
feeling more warmth and closeness with their parents dur-
ing childhood were less likely to suffer from a variety of
chronic diseases 35 years later [13]. Moreover, psycholog-
ical benefits of positive social relationships appear to
accrue and persist across the life course [14]. Countering
the notion that aging is associated with pervasive loss and
sadness, numerous studies have indicated that as people
age they tend to increase their involvement in emotionally
close relationships (and correspondingly reduce their
investment in peripheral relationships). This results in sta-
ble levels of positive affect and declining levels of negative
affect among older adults [14-17]. Taken together, this
work suggests that biological benefits of positive social
relationships are likely to be evident even (perhaps espe-
cially) at older ages. However, the neurobiological mech-
anisms by which positive aspects of relationships
promote health are somewhat under studied.
Oxytocin, Positive Social Experiences, and Resilience
Animal research on nurturing and social bonding suggests
that oxytocin serves a crucial function in creating a power-
ful bond between a mother and child. Disruptions to this
bond can lead to dysregulation of brain chemistry and
specifically the stress response in the baby [18,19]. Recent
work also has suggested that dysregulated stress responses
early in life are associated with high levels of distress, dys-
regulated immune response, and other processes in adult-
hood, which are related to accelerated aging [20].
Oxytocin is a peptide that is produced in a variety of
hypothalamic neurons and can be released into the brain
or into general circulation in pulsatile fashion by sensory
and other stimuli. It is a hormone but also functions as a

Page 3

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 3 of 17
(page number not for citation purposes)
neurotransmitter. Circulating levels of oxytocin and the
number of oxytocinergic neurons in the hypothalamus
are independent of gender [21]. Animal research has dem-
onstrated effects of oxytocin not only on a range of social
and affiliative behaviors, but also on physiological and
developmental outcomes. Oxytocin is clearly involved in
general central nervous arousal processes, with effects that
reduce stress-related activation [22]. It also seems to shift
energy use toward more positive health-promoting inter-
nal activities like storing nutrients and increasing the rate
of wound healing [23]. Estrogen appears to increase oxy-
tocin receptor gene expression and receptor binding, and
animal studies have found that greater oxytocin release
occurs in females versus males in response to threat [24-
26]. Overall, the research suggests that oxytocin could
influence health via two related mechanisms: 1) by pro-
moting social bonding and leading to release of more oxy-
tocin, and 2) by inhibiting stress-related cognitive,
affective, and biological activation. The proposed work
aims to test aspects of this model concerning potential
anti-stress effects of oxytocin and positive social relation-
ships in a controlled laboratory setting.
Methodological and Conceptual Challenges in Studying
Oxytocin in Humans
Human-based studies of oxytocin, social relationships
and health are rare, due in part to a complex relationship
between central and peripheral levels of oxytocin in
humans. Thus, it remains unclear whether mechanisms
and effects of oxytocin on social relationships and stress
identified in animal models translate to humans and are
relevant for human health. Moreover, experimental stud-
ies have found that release of oxytocin in the brain is not
accompanied by increased oxytocin secretion into the sys-
temic blood circulation, suggesting a separation in the
central and peripheral release patterns of oxytocin and
presenting additional challenges to conducting epidemio-
logic research in humans. Some work has examined the
interrelationships between oxytocin, social support, and
stress in humans; however there are only a few such stud-
ies and they are subject to a range of conceptual and meth-
odological limitations. Limitations for this work arise in
relation to our ability to obtain valid measures of oxy-
tocin, to assess whether key effects are centrally mediated
in the brain versus in the periphery, and the possibility
that estrogen or aging alters the potency of oxytocin and
its effects
Measuring Oxytocin
Findings from research using peripheral measures of oxy-
tocin have been mixed in terms of whether oxytocin is
positively associated with better social relationships and
less distress [27-35]. However, the variability in the find-
ings may be due to the use of plasma oxytocin measures,
which do not necessarily correspond with oxytocin activ-
ity in the brain and perhaps should not be interpreted as
functioning similarly [33]. Specific biological features of
oxytocin limit the applicability of these prior studies.
First, effects of oxytocin on general arousal processes are
hypothesized to occur via central nervous rather than
peripheral mechanisms. Secondly, plasma oxytocin does
not easily cross the blood-brain barrier and various factors
that influence the release of oxytocin can differ between
the peripheral and central oxytocin systems. Experimental
studies show limited correlation between oxytocin levels
in the blood with concentrations in the brain or the cere-
brospinal fluid [36,37]. Moreover, because oxytocin is
released into the systemic blood stream in pulsatile form
and has a short half-life, it is difficult to accurately assess
levels of circulating oxytocin in human plasma without
frequent blood sampling. Thus, plasma oxytocin levels
may not be a useful indicator of the oxytocin activity in
which investigators are most interested. Another way to
study oxytocin in humans is to administer it experimen-
tally and directly observe its effects. To circumvent prob-
lems with obtaining valid measures of oxytocin, our study
experimentally manipulates oxytocin rather than attempt-
ing to measure endogenous levels with technology cur-
rently available. Experimental manipulations of oxytocin
may also provide a better understanding of its effects by
providing the opportunity to make stronger causal infer-
ences relative to other study designs.
Manipulating Oxytocin
Manipulating oxytocin presents other challenges. When
administered peripherally neuropeptides like oxytocin do
not readily pass the blood-brain barrier and may also
evoke potent hormone-like side effects. However, recent
research has suggested that intranasal administration of
neuropeptides can bypass the bloodstream and achieve
direct access to the brain in both men and women [38].
Also, intranasal effects are more likely to be centrally
mediated which mitigates concerns about the relevance of
peripheral levels. An intranasal preparation of oxytocin is
not commercially available in the U.S. at present which
has made it difficult for U.S. investigators to pursue this
line of research. Research by other investigators however,
suggests that this is a promising line of work [36]. For
example, Heinrichs and colleagues examined effects of
social support and oxytocin under conditions of stress
among healthy young men. Provision of social support
and oxytocin each reduced cortisol response and distress
in the men, and the combination of both social support
and oxytocin had the strongest effects. Another study
found intranasal administration of oxytocin increased
trust among men [39]. Although baseline endogenous
levels of oxytocin were not assessed, findings indicate that
studies using intranasal administration of oxytocin may
yield insight into the biological processes linking positive
social interactions and health.

Page 4

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 4 of 17
(page number not for citation purposes)
One concern for studies that look at effects of exogenously
administered oxytocin is whether exogenous and endog-
enous effects are likely to be similar. Prior research sug-
gests that endogenous and exogenous oxytocin have
similar peripheral and central effects, including attenua-
tion of behavioral and endocrine responses to stress [40-
45]. Taken together, the research suggests that specific
effects of oxytocin as an underlying biological mechanism
linking socioemotional processes and health can be
appropriately investigated using experimental methods
with intranasal oxytocin administration [46].
To make it possible to pursue this line of research in the
U.S., we have determined that the aqueous form of oxy-
tocin which is commercially available in the U.S. (Oxy-
tocin injection, USP, manufactured by American
Pharmaceutical Partners, Inc.) can be used as a nasal
spray. The FDA has concurred and they deemed our use of
oxytocin injection exempt from requirements for an
Investigational New Drug Application (IND).
Generalizability of Effects
Much of the experimental research in humans with intra-
nasal oxytocin has been conducted with healthy young
men only. Effects of oxytocin may be more difficult to
detect in women because oxytocin effects may vary at dif-
ferent points in the menstrual cycle possibly related to
estradiol levels. Similarly, only a limited amount of work
has considered oxytocin over the life course. One study in
humans suggested that younger men may respond to
physiologic challenge with increases in plasma oxytocin
concentration, but older men do not [47]. Some investi-
gators have speculated that such changes in oxytocin
response to challenge may contribute to age-related
hypothalamic-pituitary-adrenal (HPA) axis changes such
as more prolonged patterns of HPA activation with phys-
iological aging. Empirical evidence however, is not cur-
rently available. To gain more insight into effects of
oxytocin across gender and age, we are including both
men and women ranging in age from 22 to 65 years in our
study. We are directly measuring estradiol to consider
whether effects of oxytocin vary according to estrogen lev-
els in both men and women, and if such an interaction
may account for any observed gender differences in
effects.
Considering Oxytocin, Stress, Social Support, and Health
within an Experimental Paradigm
Strong links have been identified between social support
and cardiovascular outcomes [48-52] though there is
some debate as to whether social support is the critical risk
factor versus other psychosocial factors (e.g., depression)
that tend to cluster with social support [53,54]. Experi-
mental work has been advocated for understanding
observed epidemiologic associations between social rela-
tionships and cardiovascular or other health outcomes
[55]. Thus building on the epidemiologic findings,
numerous laboratory studies have considered the relation
between social support and indices of cardiovascular
function [50,56,57]. Most of these have focused on the
reactivity hypothesis, which posits that exaggerated cardi-
ovascular reactivity (CVR) to stressful experience is one
pathophysiologic mechanism linking social stress with
the development of cardiovascular disease [58]. While the
strongest evidence that CVR is associated with adverse car-
diovascular outcomes is from animal studies [50,59,60],
there is some suggestive evidence in humans as well [61-
63].
Investigators have argued that social support may buffer
the cardiotoxic effects of stress by reducing reactivity to
acute psychological stress [64]. Generally, these studies
have examined whether the cardiovascular and neuroen-
docrine responses to a stressor depend on type of social
support available. This work has consistently found evi-
dence that social support reduces CVR to a stressful expe-
rience, although a number of boundary conditions (e.g.,
friend vs. stranger, passive versus active task, etc) and indi-
vidual differences (e.g., hostility, defensiveness) have
been identified as relevant for whether and when the pres-
ence of a supportive person reduces reactivity in a labora-
tory setting [56,57,65,66]. Consistent with the qualitative
reviews of the literature, a meta-analysis of 22 studies sug-
gested that experimental manipulation of social support
has medium to large attenuating effects on cardiovascular
reactivity measured in the laboratory [56].
Specific Aims
We are conducting research to gain greater insight into the
biological mechanisms underlying epidemiologic evi-
dence linking social relationships, stress, and cardiovascu-
lar health. Specifically, our experimental research
examines the effects of exogenously administered (intra-
nasal) oxytocin on psychological and physiological out-
comes, under conditions of stress. The specific aims of this
project are to test the following hypotheses:
1. Oxytocin ameliorates the deleterious neuroendo-
crine, cardiovascular, and subjective effects of stress.
2. Oxytocin and social support have similar and addi-
tive stress-buffering effects.
3. Effects of oxytocin are stronger in women versus
men.
4. Effects of oxytocin are similar across a range of
younger and older adult ages.

Page 5

BMC Public Health 2009, 9:481 http://www.biomedcentral.com/1471-2458/9/481
Page 5 of 17
(page number not for citation purposes)
To test these hypotheses, we are conducting a placebo-
controlled double blind study using a sample of healthy
men and women recruited from the community. We are
considering oxytocin effects on a range of outcomes.
These include autonomic reactivity as measured by blood
pressure responses and high frequency heart rate variabil-
ity (measure of cardiac vagal tone). Stress-related cardio-
vascular phenotypes as characterized by the patterning of
ventricle contractility, vascular resistance, and cardiac out-
put will also be assessed. Other outcomes include meas-
ures of neuroendocrine effects as measured by levels of
cortisol and dehydroepiandrosterone (DHEA), subjective
distress and positive affect. Participants are randomly
assigned to receive either exogenous oxytocin or placebo.
They undergo a social stress manipulation with or without
social support (randomly assigned), and outcome meas-
ures are obtained at multiple times during the experimen-
tal procedure. The experiment will test whether effects of
oxytocin and social support are similar and additive, and
will also compare effects of oxytocin and social support
across adult men and women of varying ages.
Methods/Design
A placebo-controlled, double-blind design is employed.
The key experimental manipulation involves the receipt of
intranasal oxytocin spray versus placebo (saline) nasal
spray. Social support (presence or absence) is also manip-
ulated and age is measured as a continuous factor. Meas-
ures of estradiol levels are obtained. Primary outcomes are
autonomic and neuroendocrine regulation and affective
responses. Effects of oxytocin will be compared with
effects of social support, as well as across gender and age.
Within-gender group variation in effects of oxytocin will
also be considered. This protocol has been approved by
the Institutional Review Boards at the two institutions
overseeing the research: Harvard School of Public Health
and Brigham and Women's Hospital.
Sample
Participants are recruited from the community via adver-
tising in local newspapers, flyers posted in local areas, and
Craig's List. Individuals between the ages of 25 and 65
years are recruited with careful attention to including indi-
viduals from across this age spectrum for assignment to
oxytocin versus placebo groups. The age range was chosen
to capture the spectrum from younger to older adulthood.
The older age limit was selected to include those who have
entered older adulthood, but among whom there is likely
to be a substantial pool of healthy individuals. Interested
participants complete an initial telephone screen and a
face-to-face screening (see Figure 1).
We conduct extensive screening to determine individuals'
health status. Initial screening is conducted on the tele-
phone. For this, participants are informed of the nature of
the questions to be asked and the anticipated duration of
the screening process (up to 20 minutes). Participants are
asked a series of questions about age, medication use,
health status, health behaviors, and whether they have a
same-sex close friend that could come with them to the
laboratory, and will be given basic information about the
study. Individuals who have any known medical condi-
tion (including mental disorders) or are taking any type
of medication (including birth control pills) are excluded.
Because overweight individuals may be at excess risk of
undetected diabetes or hypertension, individuals with
body mass index (BMI) ≥ 30 kg/m2 are excluded. Further
exclusions include pregnancy, breastfeeding, smoking,
heavy alcohol use, drug use [as assessed by validated
instruments, the CAGE and the RAGS, [67,68], or lack of
a close friend who could join the participant for the exper-
iment.
If they are still menstruating, all eligible women are asked
to participate during the follicular phase of their men-
strual cycle. Study outcomes of interest may be influenced
by hormonal fluctuations related to the menstrual cycle.
Sample Recruitment Protocol
Figure 1
Sample Recruitment Protocol.
????????????????????????????????????
????????????????????????????????????????
????????????????????????????????????
???????????????????
????????
????????????????
??????????????????
????????????????????
??????????
????????
????????????????
???????????????????
??????????????????????
???????????
??????????????
????????????????????????????
????????????????????????????????
????????????????????????????????????
??????????????????
????????
????????????????
????????
?????????
??????????????????
?????????
??????
????????????
??????????
????????
????
???????????????
??????????????????????????????
???????????????
????????
????????????????
??????????????????
????????????????????
??????????

Page 6

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 6 of 17
(page number not for citation purposes)
Due to variability in the length of menstrual cycles, it is
difficult to determine which phase women are in (luteal,
ovulation) and their hormone levels. However, when the
menstrual cycle begins, it essentially resets the clock for
these fluctuations. Thus we can be confident women are
in the follicular phase if we examine them within the first
7 days of starting their cycle. Moreover, at this time hor-
mone levels are relatively low and stable; studies that are
concerned about effects of these hormones on other phys-
iological processes are often done during this time [69].
Eligible individuals who agree to participate undergo a
brief final screen after arrival at the laboratory and con-
senting procedures. This is conducted by a physician who
obtains a medical history, performs a physical exam, and
does a pregnancy test for women. At this time, anyone
who is pregnant, has blood pressure values above the nor-
mal range (140/90 mm Hg), or meets other medical
exclusion criteria is not invited to continue. Eligible par-
ticipants continue with the protocol. Each participant
receives monetary compensation for participation as well
as transportation costs. If participant is assigned to bring
a friend, the friend also receives monetary compensation
for participation as well as transportation costs. Partici-
pants are asked to abstain from food and drink (except
water) for 2 hours prior to participation in the experiment
and from exercise, caffeine, and alcohol during the 12
hours prior to participation. Based on power analyses, the
target number of participants per condition is 40.
Data Collection
This placebo-controlled,
employs a between subjects factorial design, using a 2
(male vs. female) × 2 (oxytocin vs. placebo) × 2 (social
support vs. no social support) design. Blinding is utilized
for the oxytocin versus placebo groups, but not in relation
to who has social support. Participants are naïve as to the
experimental hypotheses. Participants are scheduled indi-
vidually and all experimental sessions are initiated within
the same time period (afternoon) to control for diurnal
changes in cortisol secretion. Figure 2 shows the timeline
for the experimental procedures.
double-blind experiment
Consent Procedures
Primary informed consent is obtained by the research
assistant and study physician directly after participants
General Study Protocol
Figure 2
General Study Protocol.
a. Demographics & covariates
b. Individual difference
measures
a. Pre-task qu’aires
b. Saliva #2
a. Secondary Informed Consent
b. Oxytocin/Placebo
c. Task instructions
a. Post-task qu’aires
b. Saliva #3
a. Primary Informed Consent
b. Review of eligibility
c. Attach sensors for CV measures
(obtained throughout procedure)
Trier Task
Debrief
MEASURES
PROCEDURES
TIME (min)
0308090110130
45
Vanilla baseline
Saliva #1Saliva #4
Recovery period
Task prep Pre-task prep

Page 7

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 7 of 17
(page number not for citation purposes)
arrive at the Society and Health Psychophysiology Labora-
tory. This includes consent for the final screening proce-
dures, psychophysiological monitoring, and potential use
of oxytocin. If the participant has been instructed to bring
a friend, consent for the friend is also obtained at this
time. However, consent for public speaking and math per-
formance tasks is delayed to avoid anticipatory stress in
the baseline resting period. After the brief final screening,
eligible participants are invited to continue. Prior to oxy-
tocin administration written consent is obtained for par-
ticipation in the public speaking and math performance
tasks and for the use of the videotape that will be obtained
during task preparation and performance. Participants
who refuse to complete the stress tasks are dismissed from
further study.
Basic protocol
For eligible participants, sensors and transducers are
attached for continuous monitoring of autonomic nerv-
ous system activity. Participants are asked to complete
baseline questionnaire measures assessing relevant demo-
graphics and individual differences. After completing
these questionnaires participants have a quiet period dur-
ing which time they read neutral magazines. Following
this "vanilla baseline" resting period (found to be effective
in reducing effects of extraneous factors), a baseline meas-
ure of autonomic indices and the first saliva sample is
obtained [70,71]. Consent for the experimental tasks and
videotaping is then obtained, and the study physician
subsequently administers a single dose of either 24 IU
oxytocin or placebo intranasally (approximately 40 min
before stress exposure). Following previous work in this
area, a dose of 12 IU is sprayed into each nostril [39,72].
In the highly unlikely event of an adverse reaction during
the conduct of the experiment, clear procedures have been
set up to protect participants.
Task instructions for the stress tasks are given at this time.
The stress tasks are initiated 40 minutes after oxytocin or
placebo administration, the amount of time other studies
have demonstrated it takes for oxytocin to act in the cen-
tral nervous system [72]. Immediately prior to performing
the task, a state mood measure and a second saliva sample
are obtained. State mood and post-task measures and
additional saliva samples are obtained after completion of
the stress tasks. Autonomic function is monitored
throughout the procedure. Prior research has strongly
indicated that social-evaluative threat is strongest when an
evaluative audience is present and the performance is
being captured for permanent record [73]. Thus, video
cameras mounted in the ceiling record participants' face
and upper body during task preparation and perform-
ance. In addition to increasing levels of social-evaluative
threat, the videotaping procedure also provides an addi-
tional method for assessing whether and how partici-
pants' responses to social stress varies across conditions;
videotapes may be coded later for emotional responses
during task conditions by raters blind to study hypotheses
or experimental conditions.
At the completion of the protocol, the experimenter
assesses whether the subject was suspicious about any
aspect of the experiment, explains the nature of the study
and debriefs the subject (and friend if present). Partici-
pants (and friends, if relevant) are then paid for participa-
tion. In summary, the experimental session follows
standard procedure for social stress protocols, including a
vanilla baseline, task preparation and performance
period, and a recovery period. Components of the proto-
col that are unique to this study are uncomplicated - the
administration of oxytocin or placebo prior to task prepa-
ration (~1 minute + quiet wait time for it to take effect),
the presence of a supportive friend, and the brief screen at
the start of the session (~20 minutes in most cases).
Oxytocin Manipulation
Building on research suggesting that intranasal adminis-
tration of neuropeptides bypasses the bloodstream and
achieves direct access to the brain [38], we administer
intranasal oxytocin, using the biologically active form of
the neuropeptide. Intranasal oxytocin has been widely
prescribed in pregnant women to help induce labor and
delivery, and in non-pregnant women to promote lacta-
tion. In non-pregnant women, intranasal oxytocin is typi-
cally self-administered by patients multiple times over the
course of the day for several weeks at a time in a non-med-
ical setting (e.g., in the patient's home prior to breast feed-
ing) and is well tolerated. Experimental studies of effects
of oxytocin have been conducted in men as well as
women, with intranasal doses between 20 and 264 IU,
and no adverse side effects have been reported [74-79].
Moreover, studies in humans on memory, affect, and bio-
logical parameters have clearly demonstrated behavioral
and psychoneuroendocrine effects using a dose of 24 IU
intranasal oxytocin (the dose we are using) [39,72,80-82].
No adverse or unanticipated side effects have been
reported in these studies. While studies of intranasal oxy-
tocin have largely focused on younger adults (ages 18 to
25), a number of studies have included older adults with
and without mental health problems and reported no
concerns around the use of intranasal oxytocin [79,83].
In this study, participants are randomly assigned to
receive either oxytocin or placebo intranasally with both
investigators and participants blind to condition. Partici-
pants assigned to the oxytocin condition receive a dose of
24 IU oxytocin and participants in the placebo condition
receive saline. The Brigham and Women's Hospital Inves-
tigational Drug Pharmacy is responsible for obtaining the
oxytocin and placebo and preparing them for administra-
tion to participants. For participants receiving oxytocin,
aqueous oxytocin is inserted into a spray bottle. The spray

Page 8

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 8 of 17
(page number not for citation purposes)
bottle is calibrated so that emptying the spray bottle
results in delivering 24 IU of oxytocin, with about half the
bottle sprayed into each nostril. No dilution of the origi-
nal aqueous form of the oxytocin is necessary. The nasal
spray (oxytocin or placebo) is prepared within 24 hours of
administration. Duration of drug administration is
approximately 2-5 minutes.
Social Support Manipulation
Support manipulations have varied across studies, with
some studies requiring support providers to be in physical
contact with participants [84,85] vs. activating support
schemas [86], others asking support providers to give ver-
bal support [66,87], use of confederates vs. friends to
increase standardization of support provided [87-89] and
various strategies employed to reduce potential for evalu-
ation [85,90]. Findings generally suggest that reduced
potential for evaluation, having a genuine relationship
with the support provider, and verbal support are key
components [56,57]. Building on prior work and our own
pilot research, the social support condition is designed to
standardize the support available to participants across
individuals, as well as to maximize the effectiveness of the
support provided. Prior to coming to the laboratory, par-
ticipants are randomly assigned to one of two conditions,
requiring them to appear alone or accompanied by a
friend. They are then instructed to either bring a same-sex
close friend (non-romantic, excluding spouses) with
them, or to come alone. Same-sex friends are recruited to
reduce variability associated with the meaning of their
interactions. Upon arrival at the laboratory, individuals
and their friends/support providers are introduced to the
study and consented together. Friends then stay in a wait-
ing room, and join participants at the start of the stress
exposure when task instructions are given. All participants
in the friend-present condition are told that the friend's
role is to be "a support partner," and "to silently cheer
him/her on." Friends sit where they can see the participant
but not directly facing him or her. To control interaction
between the participant and friend all remarks by the
friend are scripted. Friends are asked to make an encour-
aging remark when they join participants and when par-
ticipants finish each task. To minimize possible
evaluation effects, friends are given their own tasks to
complete (fill out questionnaires) and asked to wear a
headset playing white noise during task performance.
Friends are instructed not to distract participants at this
time, but are given a stop watch and asked to nod and
smile briefly at the subject at 1-minute intervals during
each task. Friends leave participants at the start of the
recovery.
Social Stress Manipulation
Social stress is induced using the Trier Social Stress Test
(TSST), which prior work has suggested effectively creates
a stressful situation that might be encountered in the real
world [91]. The TSST generally requires participants to
perform two separate tasks in front of an audience: a 5-
minute public speaking task and a mental arithmetic exer-
cise. Modifications have been made to ensure the effec-
tiveness of the manipulation for all ages. For the speech
task, participants are asked to prepare a speech about
important self-relevant topics. All participants (regardless
of age) are given the same 4 topics (Social Security reform,
cost of prescription drugs, education reform, or the cost of
gasoline) and asked to be prepared to present their
thoughts about all the topics and potentially to answer
questions about the topics from the evaluator. They are
asked to try to talk for the full 5 minutes with as few
pauses as possible. They are informed that the talk will be
videotaped for later evaluation and that evaluators will be
coding verbal and non-verbal behavior to assess their abil-
ity to successfully present their thoughts. They are further
told that the evaluators have had extensive training on
speech evaluation, are very skilled in assessing non-verbal
behavior and body language, as well the persuasiveness
and coherence of the speech. After the speech task, partic-
ipants are asked to perform an arithmetic task, which is
described as being related to mental ability and general
intelligence. Participants are instructed to count backward
from 996 by 7. Each time they make a mistake the evalu-
ator informs them they have made an error and must start
again. Participants are informed that they will be evalu-
ated based on how many errors they make, how quickly
they perform the task, and their overall presentation. As
social support has been shown to be more beneficial
when social threat is high [56], evaluators are trained to
act in an impersonal fashion and to remind participants
that they are being evaluated, to increase the social threat
inherent in the task.
Biological Stress Responses
Multiple regulatory systems interact via a non-linear net-
work to enable individuals to adapt to challenges and
stressors in their environment. Exposure to acute stress
elicits a cascade of cognitive, affective, and biological
responses. Moreover, according to the biopsychosocial
model cardiovascular responses may be either adaptive or
maladaptive [92,93]. When situations are construed as
threatening, cardiovascular responses show more vascular
increases, as evidenced by increased ventricular contractil-
ity, heart rate, and total peripheral resistance with little
change in cardiac output. When situations are construed
more positively, cardiovascular patterns are also marked
by increased ventricular contractility, heart rate, and car-
diac output, but with decreases in total peripheral resist-
ance. Cardiovascular responses consistent with more
positive versus threatening appraisals have been linked
with better outcomes -- lower anxiety, improved hormone
functioning and higher telomerase activity -- and are con-
sidered more benign [94-96].

Page 9

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 9 of 17
(page number not for citation purposes)
In a healthy or resilient system, a variety of regulatory
mechanisms are maintained to counteract or speed recov-
ery from stress-related activation. In a less resilient system,
repeated and frequent activation of processes that are
adaptive in the short-term may lead to cumulative sys-
temic damage and ultimately initiate disease-related proc-
esses [97]. Vagal tone is hypothesized to be critically
important in psychological and physical health. Porges
(2003) argues that greater vagal control is associated with
emotion regulatory capacity and possibly more positive
social emotions and awareness of the social environment.
Similarly, lower vagal tone has been related to depression,
hostility, and anxiety [98-101]. Greater vagal control has
also been associated with reduced risk of heart disease,
relapse after myocardial infarctions, and reduced cellular
aging [102-104]. If oxytocin and social support either
directly increase vagal tone and physiologic resilience, or
buffer the effects of stress, we might expect to see signifi-
cantly more benign cardiovascular reactivity and higher
levels of heart rate variability in response to a stressor and/
or increased basal levels of parasympathetic tone. Each of
these cardiovascular responses is measured over the
course of the experiment.
Another index of regulatory capacity considers the neu-
roendocrine system. Previous work has found that the
hormone cortisol (released by HPA axis activation) is
highly reactive to stress, and chronically high levels of cor-
tisol have been linked with risk for a variety of poor health
outcomes including cardiovascular
decreased immune functioning (although hypocortiso-
lism has also been identified as problematic). Less studied
but of growing interest, is the class of anabolic hormones
which counter-regulate catabolic hormones and provide
indications of physical and psychological thriving and
greater adaptive coping [105]. One such anabolic hor-
mone is DHEA and its active metabolite DHEA sulfate
(DHEA-S), which is excreted by the zona reticularis of the
adrenal cortex in response to adrenocorticotropic hor-
mone (ACTH, part of the HPA axis cascade that directly
precedes cortisol production). Epidemiological data has
revealed relationships between low levels of DHEA and
DHEA-S and cancer, age-related disorders, immune func-
tioning, and also mental health outcomes, especially
depression [106]. The ratio of anabolic to catabolic hor-
mones provides additional information regarding the net
anabolic vs. catabolic effect on the body, and is a more
sensitive indicator of well-being than either measure
alone. For example, androgens like DHEA can counter cat-
abolic effects of cortisol on immune function, neurons,
and on protein synthesis. The net ratio provides a measure
called 'anabolic balance' (operationalized by DHEA/corti-
sol) [103,105,107]. This has been related to better per-
formance, fewer depressive symptoms, and greater
positive well-being [95,108]. We collect saliva at several
diseases and
points during the course of the experiment so that we can
assess the effect of the experiment on hormonal response.
Measures
Autonomic Nervous System (ANS) Activation
Measures of ANS activation include parameters that are
either directly measured or that can be derived from pri-
mary signals. Thus, measures of systolic and diastolic
blood pressure and mean arterial pressure are obtained as
well as heart rate, cardiac output, total peripheral resist-
ance, pre-ejection period, and heart rate variability. Car-
diac and hemodynamic
noninvasively according to established guidelines for
impedance cardiography and blood pressure measure-
ment [109]. Condition averages of the cardiac and hemo-
dynamic measures are derived from the electrocardiogram
(ECG) and blood pressure monitor. For each measure,
scores from the baseline condition will be compared with
task condition scores to indicate reactivity [110]. Vagal
control is assessed via respiratory sinus arrhythmia (RSA)
which will be derived from both ECG recordings and res-
piration estimated from impedance waveforms (ZKG). A
computer program (Mindware) is used to identify R
spikes and R-R intervals in the ECG waveform with inac-
curately identified R spikes manually corrected. The data
are detrended and submitted to Fast Fourier Transforma-
tion, and the high-frequency power spectrum (0.15-
0.40Hz) is used as an index of RSA [111]. Because we are
studying a healthy population and reactivity is not a clin-
ical endpoint, we will consider whether patterned
responses during and after the task are consistent with a
maladaptive vs. an adaptive response. Following Porges'
work we will interpret greater vagal control as a more
adaptive regulatory response. While statistical and clinical
significance may not be the same, statistically significant
effects in these experiments would provide important evi-
dence of an underlying biological mechanism for the ben-
eficial effects of social interactions and the role of
oxytocin in this process.
measures are recorded
Hormone Measures
Salivary hormone concentrations reflect the unbound
serum hormone fraction and are considered a reliable and
valid measure of the free fraction concentration of the
hormone in the blood [112,113]. Saliva samples are col-
lected using IBL sampling devices and the methodology
described by Kirschbaum and Hellhammer [114]. Four
saliva samples are collected from each participant over the
course of the experimental session. After participants
expectorate 1 ml of saliva into the Sali-caps, each saliva
sample is stored (in -80 degrees C) until required for bio-
chemical analysis, when they are sent to a laboratory to be
assayed. A random sample of 20% of the samples will be
assayed in duplicate. Anabolic balance will be derived
from assay results by creating a ratio of DHEA/cortisol

Page 10

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 10 of 17
(page number not for citation purposes)
after molar transformation to equate unit values [107].
Estrogen is measured using estradiol obtained from the
first saliva sample taken at the laboratory session
[113,115].
Psychological Measures
Negative emotions are assessed with the Spielberger State-
Trait Personality Inventory (STPI), assessing symptoms of
anxiety, anger, and depression [116]. Trait measures
assess stable individual differences in the frequency and
intensity with which individuals experience these negative
emotions. Subscales all have high a coefficients (all < .75)
and demonstrated validity. State measures assess how
anxious, angry or depressed individuals feel while in the
laboratory, and will be given just prior to performing the
speech task and again after completion of the math task.
State and trait positive affect are measured using the Posi-
tive and Negative Affect Schedule (positive affect sub-
scale) which prior studies have indicated is both valid and
reliable [117]. Because some studies have indicated that
hostility, defensiveness, or attachment style may affect the
relationship between social support and CVR, we include
measures of these constructs [66,118,119]. Hostility is
measured with the Cook-Medley hostility scale, which has
demonstrated reliability and validity, and has been found
to predict coronary heart disease [120,121]. Defensive-
ness is measured using the Marlowe-Crown Social Desira-
bility scale, following other work in this area [66,122].
Attachment is measured using the Experiences in Close
Relationships measure [123].
Because basal levels of oxytocin may be influenced by gen-
eral availability of social support, prior to the experimen-
tal tasks we also evaluate perceived social support with the
Interpersonal Support Evaluation List, which measures
perceived availability of social support resources [124].
This permits assessment of the adequacy of existing social
support networks, which could conceivably affect peo-
ple's response to the manipulations. If a support provider
is present, prior to the experimental tasks we also ask par-
ticipants about the length and nature of their relationship
with the friend. After participants complete the experi-
mental tasks we ask about their perceptions of the extent
to which the friend was supportive during the stress tasks
[89]. All participants are also asked to report how
involved they felt, and how much difficulty they experi-
enced while performing the tasks.
Equipment
The psychophysiology laboratory at HSPH includes
equipment for assessing cardiovascular function within a
four-room suite that includes a control room and rooms
for collecting physiological and other types of data. It is
equipped with a Biopac MP150 system (BIOPAC Systems
Inc, Goleta, CA) for the measurement of ECG, impedance
cardiography, and blood pressure. It is further equipped
with a stadiometer, stimulus presentation equipment, and
a complete audio-visual system for monitoring and data
collection (including multiple video cameras, speakers,
etc). The laboratory also has capacity for extraction and
storage of biological samples (e.g., blood, urine, saliva).
Blood pressure is recorded with a Medwave blood pres-
sure monitor that provides continual readings of blood
pressure. Cardiac responses are obtained with ECG system
using a Standard lead II configuration (ECG100C) and an
impedance cardiograph system (NICO100C). The imped-
ance cardiograph uses a four mylar band configuration
placed around the neck and torso and was developed
according to published guidelines [125]. All signals are
conditioned, amplified, and recorded onto a computer for
subsequent editing and ensembling. The editing program
(Mindware Suite, Mindware Technologies, Westerville,
OH), is a specialized scoring program that allows for arti-
fact editing and ensembling averages of impedance wave-
forms.
Data Analyses
Analysis will begin with thorough descriptive and graphi-
cal examinations of variables in both studies, including
comparisons across randomization groups. Analyses will
include adjustment for covariates if not covered by rand-
omization. Distributional assumptions of dependent var-
iables will be assessed, and measures will be transformed,
if necessary. Analyses will use applications of the general
linear model as appropriate for each hypothesis under
study. For example, when examining relations between
the dichotomous factors of social support, oxytocin, and
gender, we will use analyses of variance (ANOVA) with
repeated measures. Responses are measured over time
both before and after the stress induction. Cardiovascular
outcomes will be considered both in separate models and
using multivariate analysis of variance, as they are likely
correlated within each individual. When considering the
relation between age (a continuous factor) and oxytocin,
we will use multiple linear regression techniques incorpo-
rating an interaction term between age and oxytocin as
well as relevant covariates, but we will also consider using
linear mixed models to account for the repeated measures
on the outcomes and the resulting intra-individual corre-
lation in these measures. Statistical tests will be two-tailed
and conducted at the .05 level of significance. Regarding
the specific hypotheses, this experimental study provides
data for analyses that will examine whether:
1) Under conditions of stress, higher levels of oxytocin
lead to higher vagal control, more benign stress reac-
tivity (reduced blood pressures, cardiovascular
response patterns characterized by increased ventricu-
lar contractility, heart rate, and cardiac output, but
decreased total peripheral resistance), higher anabolic

Page 11

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 11 of 17
(page number not for citation purposes)
balance, and reduced subjective distress, by compar-
ing the change in responses before and after stress
across the oxytocin and placebo control groups.
2) Oxytocin combined with social support reduces the
effects of stress more than the individual effects of oxy-
tocin and social support by comparing the change in
responses before and after stress across all possible
combinations of the two treatment groups (oxytocin
vs. placebo; social support vs. no social support). Oxy-
tocin and social support combined is expected to be
associated with highest levels of vagal control, the
most benign pattern of autonomic reactivity, and the
lowest levels of subjective distress.
3) Oxytocin reduces effects of stress more for women
than for men by comparing changes in responses
before and after stress in the presence of oxytocin or
placebo across men and women. Exploratory analyses
will consider whether the strength of the effects of oxy-
tocin may be attributable to differences in estrogen
levels by comparing effects of oxytocin across pre- and
post-menopausal women (estrogen levels are expected
to be higher among pre-menopausal women) and also
considering interactions between estrogen (measured
directly via levels of estradiol) and oxytocin in both
men and women.
4) Effects of oxytocin are similar across younger and
older aged adults by comparing the change in
responses before and after stress in the presence of
oxytocin or placebo across age.
The general analytic strategy will start by fitting a model
that includes all interactions and then move to a statisti-
cally equivalent simpler model using likelihood ratio
tests. In cases where there are more than two measure-
ments over time, several covariance structures will be con-
sidered. The possibility of multiplicative effects (between
oxytocin and social support, for instance) will be explored
graphically. Other research has suggested that the effects
of oxytocin and/or social support may not emerge imme-
diately. Thus, elapsed time will be an important consider-
ation, and separate analyses will be conducted to
determine when over the course of the experimental tasks,
effects of oxytocin and/or social support are evident on
cardiovascular, neuroendocrine, and subjective parame-
ters. These analyses will involve multiple single compari-
sons between baseline and later levels of the dependent
variables, and will be conducted with adjusted levels of
significance as appropriate (e.g., t-tests with Bonferroni
corrections).
Additional analyses will use separate models testing inter-
action terms to consider whether stable characteristics of
the individual (i.e., hostility or defensiveness, general lev-
els of perceived social support, nature of relationship with
the support provider) may alter effects of either oxytocin
or social support on responses to stress. Population based
studies that have looked at the distribution of hostility or
defensiveness suggest that there is a broad range of scores
that are skewed toward lower levels, even in populations
with somewhat higher mean scores [126,127]. Moreover,
individuals with high levels of these traits tend to have
lower levels of social support [66,126], making it less
likely that they would have a close friend and would
therefore be excluded from the study initially. As we are
randomizing individuals to support or no support, and
cell sizes are moderate (n = 40), it is not unreasonable to
expect a relatively equal distribution across the conditions
of individuals with specific characteristics that might
affect responses to the social support manipulation. We
will however, control for these individual differences in
the primary analyses. Additional analyses will consider
effects of individual differences by stratifying across indi-
viduals who are high and low in a particular trait and
examine for trends that might suggest possible interac-
tions. We will also examine if perceptions of support pro-
vided during the experiment are influenced by oxytocin.
As the support provided is highly standardized, percep-
tions of greater support may suggest that oxytocin
enhances receptivity to support.
Power and Sample Size
Power is calculated to accommodate both the effects of
the dichotomous experimental factors, as well as age
(measured as a continuous variable). Since power will be
limited by hypotheses where 3-way interactions are being
tested, power is first computed for representative out-
comes for those hypotheses (including vagal tone meas-
ured by RSA), based on information available in the
existing literature. Power for hypothesis 1 will be higher.
Power is calculated for a two-sided test of a three-way
interaction over ranges of reliabilities and differences to
be detected. For all outcomes, 40 participants per gender/
treatment group will result in power of 80-90% to detect
effect sizes similar to those seen in the literature. Since
participants are used as their own controls, power varies
by measurement reliability, and this sensitivity was con-
sidered. Power was computed using a normal test that
assumes independence across participants, compound
symmetric covariance within each subject, and approxi-
mately constant variance before and after the treatment.
Values for the parameters used here were obtained from
the literature [72,88,89,98,128]. In addition, a sample of
this size will provide 80% power to detect a medium sized
effect using multiple linear regression to consider age and
oxytocin adjusting for other covariates as necessary [129].
Discussion
Understanding the determinants of healthy aging is a
major public health priority and identifying effective

Page 12

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 12 of 17
(page number not for citation purposes)
measures to prevent or delay the onset of chronic diseases
is an important goal. Limited work in human populations
has focused on protective factors that promote resilience
in the face of challenge over the life course, or has
included consideration of these processes at the molecular
level. Taken together with recent theories of biobehavioral
mechanisms underlying both stress and health benefits of
positive social relationships [2,5,130], molecular work
implicates oxytocin and endogenous opioids in regulating
stress-responsive systems or stimulating other internal
regulatory systems involved in social interactions and
health. However, empirical evidence for these ideas in
human populations is still exceedingly limited, and
insight at the molecular level has largely been provided by
brain research in animals, with active investigation of the
effects of oxytocin on physiological and behavioral stress
responses.
Experimental research on oxytocin, social relationships,
and health in adulthood will contribute to the scientific
knowledge base for maximizing active life and health
expectancy. The proposed research on oxytocin and social
relationships integrates concepts of life-course develop-
ment, and represents a critical domain that may contrib-
ute importantly to improving health and well-being at
older ages. This work represents the earliest stages of
investigation, but begins to address questions about oxy-
tocin and how it may help to explain the links between
social relationships and aging. This study will be one of
the first to include older individuals in the study and con-
sider combined effects of oxytocin and social support on
stress response in relation to age. Moreover, despite
hypothesized differences in effects of oxytocin related to
estrogen, this will be one of the first studies to compare
the effects of oxytocin on stress and positive social interac-
tion across men and women. Most experimental studies
have been conducted in samples of either all men or all
women. Additional insight will be gained by considering
effects across multiple domains, including a detailed con-
sideration of cardiovascular stress response with a partic-
ular focus on parameters that have previously been linked
with risk of developing heart disease. Identification of
how a specific biological substrate is involved in the rela-
tionship between positive social and emotional factors
and physical health in older men and women will provide
a more definitive understanding of healthy aging proc-
esses.
Potential Limitations
There are several potential limitations to this study. First,
the animal literature suggests that effects of interest seem
to be a function of brain levels of oxytocin rather than lev-
els in the periphery. However, oxytocin released in the
brain cannot easily be measured, and measures of oxy-
tocin in blood plasma that can be obtained may not be
functionally related to the key biological effects of inter-
est. Relatedly, with intravenous administration neuropep-
tides cross the blood/brain barrier, but at such low rates
that a pharmacological dose would be needed to achieve
behavioral effects [131]. Such high doses would create
confounding peripheral influences that would make
mechanistic interpretations difficult. Making use of recent
work indicating that intranasal administration of neu-
ropeptides can gain access to the brain and produce cen-
tral effects [38,132], we have broken the barrier for
conducting this work in the U.S. by establishing that the
aqueous form of oxytocin (commercially available in the
U.S.) can be used with a nebulizer as a nasal spray.
Second, we cannot directly measure levels of oxytocin
either at baseline or after intranasal oxytocin administra-
tion. Thus, we cannot fully account for basal oxytocin lev-
els nor confirm levels of oxytocin in the brain after
administering the hormone. Due to the invasive nature of
the procedure, few experimental human studies of oxy-
tocin have measured cerebrospinal fluid (CSF) for oxy-
tocin levels either prior or subsequent to intranasal
administration of the peptide. However, studies that have
administered oxytocin intranasally have reliably found
the predicted effects despite the inability to confirm levels
of oxytocin in the brain after administering the hormone
and regardless of basal oxytocin levels [39,72,133]. A sal-
ivary assay for oxytocin is under development [134] but is
not yet available; we hope to be able to incorporate use of
this biomarker with subsequent studies in this line of
work. However, based on the existing literature, we antic-
ipate that our current methods will provide valid evidence
on which to build future studies.
Third, we cannot determine whether exogenous adminis-
tration of oxytocin is equivalent to endogenous oxytocin.
Human studies using exogenous intranasal administra-
tion of oxytocin have produced findings highly consistent
with those in the animal literature examining effects of
both exogenous and endogenous oxytocin [46]. We are
administering the biologically active form of the neu-
ropeptide. Endogenous and exogenous oxytocin have
similar effects on outcomes including with regard to
potential anxiolytic effects, attenuation of behavioral and
endocrine responses to stress, and social and reproductive
competency [41-45]. For example, endogenous central
and peripheral oxytocin are elevated during lactation, and
physiological responses to stressors are attenuated
[42,43,135-137]. In rats, exogenous oxytocin reduces cor-
ticosterone secretion in response to stress and decreases
anxiety [45,138]. There is also evidence in rats that early
exposure to either exogenous or endogenous oxytocin has
long-lasting organizational effects in relation to expres-
sion of adult behavior [139-141]. Studies of effects of
endogenous oxytocin during lactation in humans have

Page 13

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 13 of 17
(page number not for citation purposes)
been used [27,135], but it is difficult to control confound-
ing factors in endogenous stimulation paradigms, partic-
ularly the release of other hormones [72]. Moreover, the
inhibitory effect of oxytocin on HPA axis responsiveness
appears to be a function of central modulation. Taken
together, the research suggests that specific effects of oxy-
tocin as an underlying biological mechanism linking
socioemotional processes and health may be appropri-
ately investigated using experimental methods with intra-
nasal oxytocin administration [46].
Fourth, generalizability is limited because we include only
healthy community volunteers. Our sample includes indi-
viduals of all races and ethnicities who are eligible. Ideally
this will result in a racial/ethnic distribution in the sample
that approximately mirrors the locales from which we will
be recruiting. However we recognize that in fact, it may be
more difficult to recruit minority participants which could
result in a sample that is not representative of the local
community. Future work will be needed to determine
whether effects are similar across diverse populations and
in the more general population.
Fifth, we will not be able to determine how oxytocin spe-
cifically influences cognitive and affective processes
related to fear and anxiety or if effects of intranasal oxy-
tocin are mediated by central or peripheral increases. It
may be possible to gain greater comprehension of the
mechanisms underlying these processes as new tech-
niques for obtaining peripheral measures of oxytocin are
developed. Additional work may also develop animal
models to help address these questions, with the animal
studies explicitly building on the finding from and gaps
noted in the human studies.
Future Directions
This research uses a multidisciplinary approach to exam-
ine neurobiological underpinnings for the observed epi-
demiological associations between positive social
interactions and emotion experiences and physical health
outcomes in older adulthood. Our study will systemati-
cally test a biobehavioral model of social relationships
and health, and examine a key peptide that may underlie
the observed beneficial effects of positive social interac-
tions and psychological states on health and longevity. At
conclusion of the study we will have solid evidence con-
cerning the effects of oxytocin on stress response, whether
it has similar effects regardless of age or gender, and
whether its potency depends in part on circulating levels
of estradiol. Further work can build on this knowledge
and consider potential effects of oxytocin on other biolog-
ical parameters such as levels of inflammation or catecho-
lamines.
Future research should examine how effects of oxytocin
may relate to effects of different types of social support
(friend, same sex, stranger, etc), and whether oxytocin
may actually promote provision of more support or
greater receptiveness to all kinds of support. A greater
understanding of how oxytocin functions in human
health may also be obtained by identification of gene by
environment interactions in relation to oxytocin, and by
considering the relation between oxytocin, the opioid
pathway and endogenous opioids.
This novel area of investigation has the potential to have
major impact on understanding how positive social and
emotional experiences influence adult stress response,
health, and longevity. This research will provide a solid
platform from which to launch a larger interdisciplinary
research program aimed at identifying neurobiological
underpinnings for the observed epidemiological associa-
tions between positive social interactions and emotion
experiences and physical health outcomes in older adult-
hood. A neurobiological understanding of resilience can
inform efforts for both prevention and intervention of dis-
eases or problems common in later life.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
All authors contributed to the design and coordination of
the study and have read, commented on, and approved
the manuscript.
Acknowledgements
This work is supported by a grant from NIH 1R21AG030632-01A2. Funds
for developing this project were provided by the Robert Wood Johnson
Foundation Health and Society Scholars Seed Grant Program. We are
grateful to Dr. Markus Heinrichs for conceptual and methodological advice
on this work.
References
1.Russek LG, King SH, Russek SJ, Russek HI: The Harvard Mastery
of Stress Study 35-year follow-up: Prognostic significance of
patterns of psychophysiological arousal and adaptation. Psy-
chosomatic Medicine 1990, 52:271-285.
2.Taylor SE, Klein LC, Lewis BP, Gruenewald TL, Gurung RA, Updegraff
JA: Biobehavioral responses to stress in females: tend-and-
befriend, not fight-or-flight. Psychological Review 2000,
107(3):411-429.
3.Taylor SE, Lerner JS, Sage RM, Lehman BJ, Seeman TE: Early envi-
ronment emotions, responses to stress and health. Journal of
Personality 2004, 72(6):1365-1393.
4.Uvnas-Moberg K: Oxytocin linked antistress effects - the relax-
ation and growth response. Acta Psychologica Scandinavica 1997,
640(Supplement):38-42.
5.Knox SS, Uvnas-Moberg K: Social isolation and cardiovascular
disease: an atherosclerotic pathway? Psychoneuroendocrinology
1998, 23(8):877-890.
6.Berkman LF, Syme SL: Social networks host resistance and mor-
tality: A nine-year follow-up study of Alameda County resi-
dents. American Journal of Epidemiology 1979, 109:186-204.

Page 14

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 14 of 17
(page number not for citation purposes)
7.Seeman TE: Health promoting effects of friends and family on
health outcomes in older adults. Am J Health Promot 2000,
14(6):362-370.
Ryff CD, Singer B: Interpersonal flourishing: A positive health
agenda for the new millennium. Personality and Social Psychology
Review 2000, 4(1):30-44.
Pressman SD, Cohen S, Miller GE, Barkin A, Rabin BS, Treanor JJ:
Loneliness, social network size and immune response to
influenza vaccination in college freshmen. Health Psychol 2005,
24(3):297-306.
Cacioppo JT, Hawkley LC, Crawford LE, Ernst JM, Burleson MH,
Kowalewski RB, Malarkey WB, Van Cauter E, Berntson GG: Loneli-
ness and health: potential mechanisms. Psychosom Med 2002,
64(3):407-417.
Kiecolt-Glaser JK, Loving TJ, Stowell JR, Malarkey WB, Lemeshow S,
Dickinson SL, Glaser R: Hostile marital interactions proinflam-
matory cytokine production and wound healing. Arch Gen Psy-
chiatry 2005, 62(12):1377-1384.
Epel ES, Blackburn EH, Lin J, Dhabhar FS, Adler NE, Morrow JD, Caw-
thon RM: Accelerated telomere shortening in response to life
stress. Proc Natl Acad Sci USA 2004, 101(49):17312-17315.
Russek LG, Schwartz GE: Perceptions of parental caring predict
health status in midlife: a 35-year follow-up of the Harvard
Mastery of Stress Study. Psychosom Med 1997, 59(2):144-149.
Carstensen LL, Charles ST: Emotion in the second half of life.
Current Directions in Psychological Science 1998, 7(5):144-149.
Carstensen LL, Isaacowitz DM, Charles ST: Taking time seriously.
A theory of socioemotional selectivity. Am Psychol 1999,
54(3):165-181.
Carstensen LL, Pasupathi M, Mayr U, Nesselroade JR: Emotional
experience in everyday life across the adult life span. J Pers Soc
Psychol 2000, 79(4):644-655.
Lang FR, Carstensen LL: Close emotional relationships in late
life: further support for proactive aging in the social domain.
Psychol Aging 1994, 9(2):315-324.
National Research Council and Institute of Medicine: From neurons to
neighborhoods: the science of early child development. Committee on inte-
grating the science of early childhood development Washington, D.C.:
National Academy Press; 2000.
McCarthy MM: Oxytocin inhibits infanticide in female house
mice (Mus domesticus). Hormones and Behavior 1990, 24:365-375.
Taylor SE, Lehman BJ, Kiefe CI, Seeman TE: Relationship of early
life stress and psychological functioning to adult C-reactive
protein in the coronary artery risk development in young
adults study. Biol Psychiatry 2006, 60(8):819-824.
Gutkowska J, Jankowski M, Mukaddam-Daher S, McCann SM: Oxy-
tocin is a cardiovascular hormone. Brazilian Journal of Medical
and Biological Research 2000, 33:625-633.
Fehm-Wolfsdorf G, Born J: Behavioral effects of neurohypophy-
seal peptides in healthy volunteers: 10 years of research. Pep-
tides 1991, 12:1399-1406.
Petersson M, Uvnas-Moberg K, Erhardt S, Engberg G: Oxytocin
increases locus coeruleus alpha 2-adrenoreceptor respon-
siveness in rats. Neurosci Lett 1998, 255(2):115-118.
De Kloet ER, Voorhuis TAM, Elands J: Estradiol induces oxytocin
binding sites in rat hypothalamic ventromedial nucleus. Eur J
Pharmacol 1986, 118:185-186.
Young LJ, Muns S, Wang Z, Insel TR: Changes in oxytocin recep-
tor mRNA in rat brain during pregnancy and the effects of
estrogen and interleukin-6.
9(11):859-865.
McCarthy MM: Estrogen modulation of oxytocin and its rela-
tion to behavior. Adv Exp Med Biol 1995, 395:235-245.
Heinrichs M, Meinlschmidt G, Neumann I, Wagner S, Kirschbaum C,
Ehlert U, Hellhammer DH: Effects of suckling on hypothalamic-
pituitary-adrenal axis responses to psychosocial stress in
postpartum lactating women. J Clin Endocrinol Metab 2001,
86(10):4798-4804.
Heinrichs M, Neumann I, Ehlert U: Lactation and stress: protec-
tive effects of breast-feeding in humans. Stress 2002,
5(3):195-203.
Light KC, Grewen KM, Amico JA: More frequent partner hugs
and higher oxytocin levels are linked to lower blood pressure
and heart rate in premenopausal women. Biol Psychol 2005,
69(1):5-21.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.

J Neuroendocrinol 1997,
26.
27.
28.
29.
30.Light KC, Smith TE, Johns JM, Brownley KA, Hofheimer JA, Amico JA:
Oxytocin responsivity in mothers of infants a preliminary
study of relationships with blood pressure during laboratory
stress and normal ambulatory activity. Health Psychol 2000,
19(6):560-567.
Sanders G, Freilicher J, Lightman SL: Psychological stress of expo-
sure to uncontrollable noise increases plasma oxytocin in
high emotionality women. Psychoneuroendocrinology 1990,
15:47-58.
Turner RA, Altemus M, Enos T, Cooper B, McGuinness T: Prelimi-
nary research on plasma oxytocin in normal cycling women:
investigating emotion and interpersonal distress. Psychiatry
1999, 62(2):97-113.
Turner RA, Altemus M, Yip DN, Kupferman E, Fletcher D, Bostrom
A, Lyons DM, Amico JA: Effects of emotion on oxytocin prolac-
tin, and ACTH in women. Stress 2002, 5(4):269-276.
Grewen KM, Girdler SS, Amico J, Light KC: Effects of partner sup-
port on resting oxytocin cortisol, norepinephrine, and blood
pressure before and after warm partner contact. Psychosom
Med 2005, 67(4):531-538.
Taylor SE, Gonzaga GC, Klein LC, Hu P, Greendale GA, Seeman TE:
Relation of oxytocin to psychological stress responses and
hypothalamic-pituitary-adrenocortical axis activity in older
women. Psychosom Med 2006, 68(2):238-245.
Bartz JA, Hollander E: The neuroscience of affiliation: forging
links between basic and clinical research on neuropeptides
and social behavior. Horm Behav 2006, 50(4):518-528.
Carter S, Lederhendler I, Kirkpatrick B: The integrative neurobi-
ology of affiliation: Introduction. Annals of the New York Academy
of Sciences 1997, 807:xiii-xviii.
Born J, Lange T, Kern W, McGregor GP, Bickel U, Fehm HL: Sniffing
neuropeptides: a transnasal approach to the human brain.
Nature Neuroscience 2002, 5(6):514-516.
Kosfeld M, Heinrichs M, Zak PJ, Fischbacher U, Fehr E: Oxytocin
increases trust in humans. Nature 2005, 435(7042):673-676.
Lippert TH, Mueck AO, Seeger H, Pfaff A: Effects of oxytocin out-
side pregnancy. Horm Res 2003, 60(6):262-271.
Insel TR, Winslow JT: Central administration of oxytocin mod-
ulates the infant rat's response to social isolation. Eur J Phar-
macol 1991, 203(1):149-152.
Lightman SL, Windle RJ, Wood SA, Kershaw YM, Shanks N, Ingram
CD: Peripartum plasticity within the hypothalamo-pituitary-
adrenal axis. Prog Brain Res 2001, 133:111-129.
Neumann ID, Torner L, Wigger A: Brain oxytocin: differential
inhibition of neuroendocrine stress responses and anxiety-
related behaviour in virgin pregnant and lactating rats. Neu-
roscience 2000, 95(2):567-575.
Neumann ID, Wigger A, Torner L, Holsboer F, Landgraf R: Brain
oxytocin inhibits basal and stress-induced activity of the
hypothalamo-pituitary-adrenal axis in male and female rats:
partial action within the paraventricular nucleus. J Neuroendo-
crinol 2000, 12(3):235-243.
Windle RJ, Shanks N, Lightman SL, Ingram CD: Central oxytocin
administration reduces stress-induced
release and anxiety behavior in rats. Endocrinology 1997,
138(7):2829-2834.
Heinrichs M, Domes G: Neuropeptides and social behavior:
Effects of oxytocin and vasopressin in humans. Progress in Brain
Research 2008, 170:337-350.
Chiodera P, Volpi R, Capretti L, Caiazza A, Marchesi M, Caffari G,
Rossi G, Coiro V: Oxytocin response to challenging stimuli in
elderly men. Regul Pept 1994, 51(2):169-176.
Berkman LF, Czajkowski SM, Hill DR, Clarkson TB, Rockville MD:
Social support and cardiovascular disease morbidity and
mortality in women. In Women, behavior, and cardiovascular disease
Edited by: Czajkowski SM, Hill DR, Clarkson TB. Volume NIH Publi-
cation No. 94-3309. National Institutes of Health; 1994:159-166.
Berkman LF, Leo-Summers L, Horwitz RI: Emotional support and
myocardial infarction: findings from a prospective popula-
tion-based study of the elderly. Annals of Internal Med 1992,
117:1003-1009.
Christenfeld N, Gerin W: Social support and cardiovascular
reactivity. Biomed & Pharmaocother 2000, 54:251-257.
Eng PM, Rimm EB, Fitzmaurice G, Kawachi I: Social ties and change
in social ties in relation to subsequent total and cause-spe-
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
corticosterone
46.
47.
48.
49.
50.
51.

Page 15

BMC Public Health 2009, 9:481http://www.biomedcentral.com/1471-2458/9/481
Page 15 of 17
(page number not for citation purposes)
cific mortality and coronary heart disease incidence. Am J Epi-
demiol 2002, 155:700-709.
Rozanski A, Blumenthal JA, Davidson KW, Saab P, Kubzansky LD:
The epidemiology pathophysiology, and management of psy-
chosocial risk factors in cardiac practice: The emerging field
of behavioral cardiology. Journal of American College of Cardiology
2005, 45(5):637-651.
Lett HS, Blumenthal JA, Babyak MA, Strauman TJ, Robins C, Sher-
wood A: Social support and coronary heart disease: epidemi-
ologic evidence and implications for treatment. Psychosom
Med 2005, 67(6):869-878.
Williams RB, Barefoot JC, Schneiderman N: Psychosocial risk fac-
tors for cardiovascular disease: more than one culprit at
work. Jama 2003, 290(16):2190-2192.
Berkman LF, Glass TA, Brissett I, Seeman TE: From social integra-
tion to health: Durkheim in the new millennium. Social Science
and Medicine 2000, 51:843-857.
Thorsteinsson EB, James JE: A meta-analysis of the effects of
experimental manipulations of social support during labora-
tory stress. Psychology and Health 1999, 14:869-886.
Uchino BN: Social support and health: a review of physiologi-
cal processes potentially underlying links to disease out-
comes. J Behav Med 2006, 29(4):377-387.
Krantz DS, Manuck SB: Acute psychophysiologic reactivity and
risk of cardiovascular disease: A review and methodologic
critique. Psychological Bulletin 1984, 96:435-564.
Manuck SB, Kaplan JR, Adams MR, Clarkson TB: Behaviorally elic-
ited heart rate reactivity and atherosclerosis in female
cynomolgus monkeys (Macaca fascicularis). Psychosom Med
1989, 51:306.
Manuck SB, Kaplan JR, Clarkson TB: Behaviorally induced heart
rate reactivity and atherosclerosis in cynomolgus monkeys.
Psychosom Med 1983, 45:95.
Kamarck TW, Everson SA, Kaplan GA, Manuck SB, Jennings JR, Salo-
nen R, Salonen JT: Exaggerated blood pressure responses dur-
ing mental stress are associated with enhanced carotid
atherosclerosis in middle-aged Finnish men: Findings from
the Kuopio Ischemic Heart Disease study. Circulation 1997,
96(11):3842-3848.
Everson SA, Kaplan GA, Goldberg DE, Salonen JT: Anticipatory
blood pressure response to exercise predicts future high
blood pressure in middle-aged men. Hypertension 1996,
27(5):1059-1064.
Everson SA, Lynch JW, Kaplan GA, Lakka TA, Sivenius J, Salonen JT:
Stress-induced blood pressure reactivity and incident stroke
in middle-aged men. Stroke 2001, 32(6):1263-1270.
Uchino BN, Cacioppo JT, Kiecolt-Glaser JK: The relationship
between social support and physiological processes: a review
with emphasis on underlying mechanisms and implications
for health. Psychological Bulletin 1996, 119(3):488-531.
Kamarck TW, Annunziato B, Amateau LM: Affiliation moderates
the effects of social threat on stress-related cardiovascular
responses: boundary conditions for a laboratory model of
social support. Psychosom Med 1995, 57(2):183-194.
Westmaas JL, Jamner LD: Paradoxical effects of social support
on blood pressure reactivity among defensive individuals.
Ann Behav Med 2006, 31(3):238-247.
Ewing JA: Detecting alcoholism: The CAGE Questionnaire.
JAMA 1984, 252:1905-1907.
Sobell LC, Sobell MB, Levin C, Cleland P, Ellingstad T, Toll B: RAGS:
A new brief drug abuse screening instrument. Association for
Advancement of Behavior Therapy 1999.
Symonds CS, Gallagher P, Thompson JM, Young AH: Effects of the
menstrual cycle on mood neurocognitive and neuroendo-
crine function in healthy premenopausal women. Psychol Med
2004, 34(1):93-102.
Jennings JR, Kamarck T, Stewart C, Eddy M, Johnson P: Alternate
cardiovascular baseline assessment techniques: vanilla or
resting baseline. Psychophysiology 1992, 29(6):742-750.
Piferi RL, Kline KA, Younger J, Lawler KA: An alternative
approach for achieving cardiovascular baseline: viewing an
aquatic video. Int J Psychophysiol 2000, 37(2):207-217.
Heinrichs M, Baumgartner T, Kirschbaum C, Ehlert U: : Social sup-
port and oxytocin interact to suppress cortisol and subjec-
tive responses to psychosocial stress. Biol Psychiatry 2003,
54(12):1389-1398.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.Dickerson SS, Kemeny ME: Acute stressors and cortisol
responses: a theoretical integration and synthesis of labora-
tory research. Psychol Bull 2004, 130(3):355-391.
Bruins J, Hijman R, Van Ree JM: Effect of a single dose of des-gly-
cinamide-[Arg8]vasopressin or oxytocin on cognitive proc-
esses in young healthy subjects. Peptides 1992, 13:461-468.
Hoover RT: Intranasal oxytocin in eighteen hundred patients.
A study on its safety as used in a community hospital. Am J
Obstet Gynecol 1971, 110(6):788-794.
Landgraf R: Plasma oxytocin concentrations in man after dif-
ferent routes of administration of synthetic oxytocin. Exp Clin
Endocrinol 1985, 85(2):245-248.
Lane D, Carroll D, Lip GY: Psychology in coronary care. Quarterly
Journal of Medicine 1999, 92(8):425-431.
Sandholm L: The effect of intravenous and intranasal oxytocin
on intramamary pressure during early lactation. Acta Obstet
Gynecol Scand 1968, 47(2):145-154.
Pitman RK, Orr SP, Lasko NB: Effects of intranasal vasopressin
and oxytocin on physiologic responding during personal
combat imagery in Vietnam Veterans with posttraumatic
stress disorder. Psychiatry Research 1993, 48:107-117.
Heinrichs M, Meinlschmidt G, Wippich W, Ehlert U, Hellhammer DH:
Selective amnesic effects of oxytocin on human memory.
Physiol Behav 2004, 83(1):31-38.
Kirsch P, Esslinger C, Chen Q, Mier D, Lis S, Siddhanti S, Gruppe H,
Mattay VS, Gallhofer B, Meyer-Lindenberg A: Oxytocin modulates
neural circuitry for social cognition and fear in humans. J
Neurosci 2005, 25(49):11489-11493.
Domes G, Heinrichs M, Michel A, Berger C, Herpertz SC: Oxytocin
improves "mind-reading" in humans. Biol Psychiatry 2007,
61(6):731-733.
Epperson CN, McDougle CJ, Price LH: Intranasal oxytocin in
obsessive-compulsive disorder.
40(6):547-549.
Coan JA, Schaefer HS, Davidson RJ: Lending a hand: social regu-
lation of the neural response to threat. Psychol Sci 2006,
17(12):1032-1039.
Kamarck T, Manuck SB, Jennings JR: Social support reduces car-
diovascular reactivity to psychological challenge: A labora-
tory model. Psychosomatic Medicine 1990, 52:42-58.
Ratnasingam P, Bishop GD: Social support schemas trait anger
and cardiovascular responses. Int J Psychophysiol 2007,
63(3):308-316.
Anthony JL, O'Brien WH: An evaluation of the impact of social
support manipulations on cardiovascular reactivity to labo-
ratory stressors. Behav Med 1999, 25(2):78-87.
Glynn LM, Christenfeld N, Gerin W: Gender, social support and
cardiovascular responses to stress. Psychosom Med 1999,
61(2):234-242.
Christenfeld N, Gerin W, Linden W, Sanders M, Mathur J, Deich JD,
Pickering TG: Social support effects on cardiovascular reactiv-
ity: is a stranger as effective as a friend? Psychosom Med 1997,
59(4):388-398.
Kirschbaum C, Klauer T, Filipp SH, Hellhammer DH: Sex-specific
effects of social support on cortisol and subjective responses
to acute psychological stress. Psychosom Med 1995, 57(1):23-31.
Kirschbaum C, Pirke KM, Hellhammer DH: The 'Trier Social
Stress Test' - A tool for investigating psychobiological stress
responses in a laboratory setting. Neuropsychobiology 1993,
28:76-81.
Blascovich J, Mendes WB, Forgas JP: Challenge and threat
appraisals: The role of affective cues. In Feeling and thinking: The
role of affect in social cognition Edited by: Forgas JP. New York NY:
Cambridge University Press; 2000:59-82.
Mendes WB, Blascovich J, Hunter SB, Lickel B, Jost JT: Threatened
by the unexpected: physiological responses during social
interactions with expectancy-violating partners. Journal of Per-
sonality and Social Psychology 2007, 92:698-716.
Epel ES, Lin J, Wilhelm FH, Wolkowitz OM, Cawthon R, Adler NE,
Dolbier C, Mendes WB, Blackburn EH: Cell aging in relation to
stress arousal and cardiovascular disease risk factors. Psy-
choneuroendocrinology 2006, 31(3):277-287.
Mendes WB, Epel ES: Threat reactivity predicts accelerated
cellular aging. Psychological Science Under review .
Tomaka J, Blascovich J, Kelsey RM, Leitten CL: Subjective, physio-
logical, and behavioral effects of threat and challenge
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.

Biol Psychiatry 1996,
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.