University of Southern California Dissertations and Theses

Adolescent life stress and the cortisol awakening response: the moderating roles of emotion regulation, attachment, and gender

(USC Thesis Other)

Adolescent life stress and the cortisol awakening response: the moderating roles of emotion regulation, attachment, and gender

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: ADOL STRESS AND CORTISOL AWAKENING_______________________1

Adolescent Life Stress and the Cortisol Awakening Response:
The Moderating Roles of Emotion Regulation, Attachment, and Gender
Kelly F. Miller
University of Southern California

A Thesis Submitted to the Faculty of the Graduate School
In Partial Fulfillment of the Degree of Master of Arts in Psychology
Department of Psychology, Clinical Science
Faculty Advisor: Gayla Margolin, PhD
May 2014

ADOL STRESS AND CORTISOL AWAKENING

2
Table of Contents
Abstract ……………… ……………………………………………………………………………4
Introduction ………………………………………………………………………………………..5
The Cortisol Awakening Response and Stress …… ………………………………………6
Social Buffering of Stress …………………………………………………………………9
Parents …………………………………………………………………………….9
Peers …………………… ………………………………………………………...10
Emotion Regulation as a Buffer ………………… ……………………………………….12
HPA Activity, Social Relationships, and Gender … ……………………………………..12
The Present Study …… …………………………………………………………………..13
Me thod…………………………………………………………………………………………...14
Overview ……………… …………………………………………………………………14
Participants ……………………………………………………………………………….15
Procedures …………… …………………………………………………………………..15
Home saliva collection … ………………………………………………………...16
Measures ……………… …………………………………………………………………17
Diurnal cortisol …………………………………………………………………..17
Covariates for Cortisol Analysis ………… ……… ………………………………17
Life stress ……………… ………………………………………………………...18
Emotion Regulation ……………………………………………………………...19
Attachment to Parents and Peers ………………… ………………………………19
Analytic Plan ………… ………………………………………………..…………………19
Results ……………………………………………………………………………………………21
ADOL STRESS AND CORTISOL AWAKENING

3
Inter-relationships among variables …………… ………………………………………...21
Hypothesis 1: Past vs. current life stress ………… ………………………………………22
Hypothesis 2a: Moderating the Relationship Between Current Stress and CAR ………..23
Hypothesis 2b: Moderating the Relationship B e tw e e n P a st S tre ss and CAR….……..… 23
Emotion Regulation …………………………………………………………...…23
Attachment to Parents … …………………………………………………………24
Attachment to Peers …… ………………………………………………………...24
Discussion ………………………………………………………………………………………..24
References …………… …………………… ……………………………………………………..33
Tables and Figures …… ………………………………………………………………………….46

ADOL STRESS AND CORTISOL AWAKENING

4
Abstract
The cortisol awakening response (CAR) is an index of hypothalamic-pituitary-adrenal (HPA)
activity that is sensitive to the effects of stress among adults (Chida & Steptoe, 2009). However,
less is known about the relationship between stress and CAR in adolescence. The present study
e x a mi ne s whe ther a doles c e nts’ pa st and c urr e nt st re ss have dif fe r e nt assoc i a ti ons wit h C AR a nd
tests whether emotion regulation, attachment to parents, or attachment to peers moderates these
relationships. Ninety-nine adolescents were recruited from a longitudinal study to provide home
saliva samples for cortisol analyses. In the laboratory, adolescents additionally reported on
stressful life experiences, emotion regulation, and attachment to parents and peers. Multi-level
models reveal that current stress is associated with increased CAR. In contrast, past stress is
associated with decreased CAR for males only. No moderation was found for the association
between current stress and CAR. In contrast, several moderators were identified for the
relationship between past stress and CAR: attachment to peers was protective for all adolescents,
while emotion regulation and attachment to parents were protective for males only. Additionally,
adolescents who are low in attachment to peers appear to represent a distinct risk group from
adolescents who are low in attachment to parents or low in emotion regulation. Results highlight
the dist inct r oles pe e rs pl a y in a dolesc e nts’ changing social worlds and underscore the gender
differences in bio-psycho-social development across middle childhood and adolescence.

Keywords: HPA axis, cortisol awakening response, stress, adolescence

ADOL STRESS AND CORTISOL AWAKENING

5
Adolescent Life Stress and the Cortisol Awakening Response:
The Moderating Roles of Emotion Regulation, Attachment, and Gender
Stress is an unavoidable part of development. Although mild, manageable stressors are
unlikely to cause harm and may, in contrast, even confer some benefits, (e.g., Lyons, Parker,
Katz, & Schatzberg, 2009), more significant stress takes a toll on the body. The hypothalamic-
pituitary-adrenal (HPA) axis of the neuroendocrine system has been identified as one pathway by
which stressful experiences become embedded in physiology. The HPA axis is activated in
response to acute experiences of stress, a process that mobilizes the body to face challenge by
enhancing cognition, directing energy to muscles, and inhibiting digestive and reproductive
processes (Sapolsky, Romero, & Munck, 2000). Although HPA activation confers short-term
benefits, repeated activations increase allostatic load (McEwen, 1998; Seeman, Singer, Rowe,
Horwitz, & McEwen, 1997). Allostasis is a homeostatic process by which the body maintains
“ stabili t y throu g h c h a n g e ,” (McEwen & Wingfield, 2003, p. 3); that is, the body modifies its
basal functioning in response to the demands of the environment. According to allostatic models,
repeated activation should lead to a recalibration of the HPA axis such that a higher degree of
threat is required in order for the body to mount a stress response. This theory is consistent with
research demonstrating that individuals with a history of severe, chronic, and early stress show
attenuated diurnal HPA activity (Miller, Chen, & Zhou, 2007) and reduced responsiveness to in-
lab stressors (Lovallo, Farag, Sorocco, Cohoon, & Vincent, 2012). Dysregulated HPA activity, in
turn, is associated with the later development of disorders, such as depression and heart disease
(Adam et al., 2010; Kumari, Shipley, Stafford, & Kivimaki, 2011).
Although the effects of early, chronic, and severe stress (e.g., trauma, early
institutionalization, parental psychopathology, Chida & Steptoe, 2009; Quevedo, Johnson,
ADOL STRESS AND CORTISOL AWAKENING

6
Loman, Lafavor, & Gunnar, 2012; Shea et al., 2007) have been extensively researched, less is
known about how more developmentally-typical stressors affect allostasis. However, examining
the role of stress in community samples is crucial for understanding how experiences shape
physiology in the larger population. Additionally, adolescence is a relatively understudied period.
Although research has focused on the cascading effect of stress experienced early in
development (Repetti, Robles, & Reynolds, 2011), adversity continues to shape regulation
throughout the lifespan. Adolescence is a period of transition, both in terms of experiences and
physiology. Therefore, this period of development is well suited for studies of allostatic
processes. Finally, few studies have directly compared the effects of current and past stress.
Allostatic theory suggests that distal and proximal stressors should have opposite effects on HPA
functioning, with past stress reducing HPA activity and current stress increasing HPA activity;
however this process has not yet been directly observed in adolescents.
In addition to identifying how stressors put adolescents at risk for HPA dysregulation,
several recent reviews have called for an increased focus on identifying factors that protect
against this association (Hostinar & Gunnar, 2013; Hostinar, Sullivan, & Gunnar, 2014; Murray-
Close, 2012). Identifying protective factors during adolescence is critical because adolescence
typically precedes the onset of disorders associated with HPA dysregulation; therefore,
developing interventions to buffer adolescents from the effects of stress may prevent later
disability.
The Cortisol Awakening Response and Stress
HPA activity is typically measured via the secretion of cortisol, the end product of the
HPA axis in humans (Sapolsky, 1994). Cortisol follows a diurnal pattern, with levels rising
steeply after awakening and declining steadily through the remainder of the day (Edwards, Evans,
ADOL STRESS AND CORTISOL AWAKENING

7
Hucklebridge, & Clow, 2001). The cortisol awakening response (CAR), the sharp increase in
cortisol in the thirty minutes following awakening, is frequently used as a measure of diurnal
HPA activity. Although a significant portion of variance in CAR is attributed to genetic factors
(Bartels, de Geus, Kirschbaum, Sluyter, & Boomsma, 2003; P. A. Gustafsson et al., 2011;
Kupper et al., 2005; Wust, Federenko, Hellhammer, & Kirschbaum, 2000), CAR is also highly
responsive to changes in the environment. In a meta-analysis of studies conducted primarily with
adults, greater experiences of life stress or work stress were associated with an elevated CAR
(Chida & Steptoe, 2009). For example, studies comparing workdays and weekends find that
CAR is greater on workdays, and even greater on weekdays when individuals anticipate more
work and have more demanding jobs (Devereux, Rydstedt, & Cropley, 2011). This may indicate
that one function of the CAR is to mobilize the body in preparation for a stressful day (Adam,
Hawkley, Kudielka, & Cacioppo, 2006; Fries, Dettenborn, & Kirschbaum, 2009; McEwen, 2000;
Miller et al., 2007; Tops, Riese, Oldehinkel, Rijsdijk, & Ormel, 2008).
Although undergoing daily stress is associated with increased CAR, a second pattern of
dysregulation, decreased CAR, has also been observed. A meta-analysis found reduced CAR in
association with post-traumatic stress disorder, fatigue, and burn-out (Chida & Steptoe, 2009).
CAR attenuation in adults has also been associated with early life trauma and early experiences
of loss (Meinlschmidt & Heim, 2005; Shea et al., 2007). Although some research suggests that
earlier-occurring or more severe stress is associated with attenuated diurnal HPA activity (Miller
et al., 2007), the opposite effect has also been reported (Engert, Efanov, Dedovic, Dagher, &
Pruessner, 2011). Therefore, although allostatic models posit that a history of elevated stress is
associated with decreased HPA activity and the concurrent experience of stress is associated with
ADOL STRESS AND CORTISOL AWAKENING

8
an increased HPA activity (Miller et al., 2007), whether distal and proximal stressors have
opposite effects on the CAR remains unclear.
More recently, research on the relationship between stress and CAR has been extended to
children and adolescents. Children who have experienced high levels of cumulative adversity
show reduced CAR in comparison to children who have experienced only moderate levels of
adversity (P. E. Gustafsson, Anckarsater, Lichtenstein, Nelson, & Gustafsson, 2010). Bevans,
Cerbone, and Overstreet (2008) found that children who had both a history of trauma and recent
exposure to traumatic events showed lower morning cortisol levels. These findings are consistent
with the hypothesis that severe stress attenuates CAR in children in the same manner as in adults.
Similarly, existing adolescent research points to a pattern of reduced CAR in association
with distal or severe stress and increased CAR in association with proximal or daily stress. For
example, young children who experienced early deprivation or neglect showed a blunted CAR
during early adolescence (Quevedo et al., 2012). In contrast, Latina/o adolescents who currently
perceived themselves as undergoing greater discrimination had an increased CAR (Zeiders,
Doane, & Roosa, 2012). However, the relatively limited number of studies examining CAR
during adolescence precludes a definitive understanding of how past and current stress may
affect HPA activity. Existing studies also often confound length of time since the stressor
occurred with severity of the stressor, as typically only severe past stressors are assessed. Only
one study has assessed how less severe forms of early stress are associated with adolescent CAR;
Roisman et al. (2009) found that greater maternal insensitivity and more time spent in daycare
were associated with reduced CAR during adolescence. No study has directly compared past and
current stress of the same type in an adolescent sample. However, without a direct comparison,
the effects of stress timing and stress type cannot be distinguished.
ADOL STRESS AND CORTISOL AWAKENING

9
Social Buffering of Stress
Social support has been identified as a consistent and robust buffer of HPA axis
functioning, across both human and animal studies (see Hostinar et al., 2014 for review).
However, the existing research has been conducted primarily with adults; less is known about
how relationships may protect adolescents from stress. Social relationships in adolescence are
not merely an upward extension of childhood or a downward extension of adulthood —
adolescence is a time of social transition and expansion. Two major sources of adolescent social
contact, parents and peers, therefore merit in-depth examination as potential moderators of stress.
Parents. Attachment theory proposes a mechanism by which individuals learn to socially
regulate stress (Bowlby, 1969). Infants first regulate their emotions through their interactions
with caregivers. Infants who have sensitive and responsive caregivers develop secure attachment,
which includes beliefs that others will provide comfort in times of distress and that unpleasant
emotions can be overcome (Ainsworth, 1978; Cassidy, 1994; Main, 1985). As children grow,
these se c ur e a tt a c hment be li e fs f a c il it a te the c hil d ’s a bil it y to i nde pe nd e ntl y r e g ulate both
physiology and emotion, even in the face of stressors (Cassidy, 1994; Schore, 2001).
The presence of caregivers was first shown to buffer the effects of stress on HPA activity
in animal models (e.g., Coe, Franklin, Smith, & Levine, 1982; Coe, Mendoza, Smotherman, &
Levine, 1978; Levine, 2001; Stanton & Levine, 1985). More recently, relationships with
caregivers have been shown to have a direct effect on HPA activity in humans. For example,
highly adverse and disrupted relationships with caregivers (e.g., Nicolson, 2004; van der Vegt,
van der Ende, Kirschbaum, Verhulst, & Tiemeier, 2009), low-quality (but not developmentally
atypical) caregiving (e.g., Ellenbogen & Hodgins, 2009; Engert et al., 2011; Essex et al., 2011;
Hackman et al., 2013), and attachment to caregivers (e.g., Oskis, Loveday, Hucklebridge, Thorn,
ADOL STRESS AND CORTISOL AWAKENING

10
& Clow, 2011), are all associated with HPA activity. Additionally, interventions designed to
increase parental sensitivity and responsiveness not only alter caregiving behaviors but also
produ c e c h a n g e in childr e n’s H P A f unc ti onin g (Cicchetti, Rogosch, Toth, & Sturge-Apple, 2011;
Dozier, Peloso, Lewis, Laurenceau, & Levine, 2008; Fisher, Stoolmiller, Gunnar, & Burraston,
2007). Taken together, these lines of research suggest that parent-child relationships directly
influence the development of HPA activity.
In contrast, relatively few studies have examined the moderating role of relationships
with parents on the association between stress and HPA activity. Although existing research
suggests that parents can be powerful moderators, most of this work has been conducted with
young children (e.g., Gunnar, Brodersen, Nachmias, Buss, & Rigatuso, 1996; Nachmias, Gunnar,
Mangelsdorf, Parritz, & Buss, 1996; Seltzer, Ziegler, & Pollak, 2010) or with adults (Hanson &
Chen, 2010) rather than with adolescents.
Peers. Adolescence is a period of decreasing reliance on caregivers and increasing time
spent with friends and romantic partners (Harris, 1995; Hartup, 1996; Hunter & Youniss, 1982).
Friends increasingly fill attachment roles, such as providing comfort in times of distress, which
were previously the sole domain of caregivers (Fraley & Davis, 1997); this change in the
hier a r c h y of a doles c e nts’ pre fe rr e d a tt a c h ment figures represents a shift of the locus of
attachment (Rosenthal & Kobak, 2010; Stein, Jacobs, Ferguson, Allen, & Fonagy, 1998). In light
of a dolesc e nts’ c ha n g in g soc ial wor lds, re lations hi ps wit h pe e rs ma y be c ome a n incr e a sin g l y important buffer of the effects of stress on HPA activity. However, this topic has gone largely
unexplored, leading to several reviews calling for the study of the protective role of adolescent
friendships on physiological stress responses (Hostinar & Gunnar, 2013; Hostinar et al., 2014;
Murray-Close, 2012). To our knowledge, only one study has examined the moderating role of
ADOL STRESS AND CORTISOL AWAKENING

11
friends on adolescent diurnal HPA activity: Doane and Zeiders (2014) found that relationship
quality with friends protect adolescents from the association between negative affect and
momentary changes in cortisol throughout the day.
Additionally, a small number of studies using samples of children have suggested that
friends serve as a buffer. For example, having at least one friend protects children from the effect
of bullying on cortisol production (Peters, Riksen-Walraven, Cillessen, & de Weerth, 2011).
Similarly, diary studies reveal that having a friend present at the time negative events occurred
reduced the cortisol response to negative events (Adams, Santo, & Bukowski, 2011). However,
research conducted with children has a limited ability to elucidate social processes in
adolescence, as the relationship between social relationships and physiology is likely to change
across development. For example, basal cortisol is associated with social relationship quality for
adolescents but not for children (Booth, 2008).
Although little research has examined the moderating effects of adolescent relationships,
several lines of research have examined their direct effects on HPA activity. However, even
these studies rarely examine the possible positive effects of peer friendships. In a recent review,
Murray-Close (2012) found that these studies instead focused on antisocial behaviors towards
peers, peer victimization, and status in adolescent social hierarchies as risk factors for adolescent
HPA dysregulation.
Given the paucity of the literature, it is difficult to predict whether peer relationships will
serve as a buffer or risk factor for adolescents who have experienced elevated stress. Although
the existing study addressing this topic suggests that peers are protective, a great deal of attention
ha s a lso bee n p a id t o the “ da r k si de ” of r e lations hips (Thoits, 2011), particularly during
adolescence. At-risk adolescents who spend more time with peers may engage in more risky or
ADOL STRESS AND CORTISOL AWAKENING

12
deviant behaviors as a means of coping with stress (Simons, Conger, & Whitbeck, 1988) and
risky behaviors in turn are associated with diurnal HPA dysregulation (e.g., Popma et al., 2007).
Therefore, the hypothesis that peer relationships might protect adolescents from stress must
remain a tentative one.
Emotion Regulation as a Buffer
Emotion regulation is composed of a range of adaptive emotional behaviors, including
maintaining of a homeostatic (as opposed to labile) emotional state, experiencing emotions that
are appropriate to the situation, recovering when negative emotions do occur, and having
awareness of on e ’s ow n e mot ional st a te (Shields & Cicchetti, 1997). Attachment theory states
that children learn emotion regulation in the context of relationships with caregivers; by
re spondi ng to i nf a nts’ e mot ional dis pla y s s e nsit iv ely, caregivers teach children to identify
emotions, provide them with strategies for self-soothing, and instill confidence that distressing
emotions can be overcome (Bowlby, 1969; Cassidy, 1994). Therefore, stressful events may be
easier to recover from for individuals with enhanced emotion regulation capacity.
Although positive and negative affective experiences have been related to changes in
diurnal cortisol (Adam, 2006), minimal research has examined the possible protective effect of
emotion regulation in the relationship between stress and HPA activity. One study of African-
American children did find that emotion regulation moderated the relationship between negative
life events and basal morning cortisol (Kliewer, Reid-Quinones, Shields, & Foutz, 2009).
Notably, one study has failed to find a direct relationship between emotion regulation and CAR
(Stadler, Evans, Hucklebridge, & Clow, 2011); however, emotion regulation may buffer the
effect of stress on CAR without affecting CAR directly.
HPA Activity, Social Relationships, and Gender
ADOL STRESS AND CORTISOL AWAKENING

13
Although both males and females follow the same general pattern of diurnal cortisol
activity, the HPA axis is modulated by sex-hormones (Conrad et al., 2004), which begin to differ
dramatically between males and females during adolescence. However, gender differences in
HPA activity are only sometimes observed, for example, some studies indicate that males have a
a smaller awakening increase in cortisol, while other studies show no gender differences in the
diurnal cortisol pattern (see Bouma, Riese, Ormel, Verhulst, & Oldehinkel, 2009).
Additionally, there is some evidence that the effect of social relationships on HPA
activity varies somewhat between males and females. For example, a range of studies indicate
that wives are more physiologically reactive to negative marital interactions than are husbands
(e.g., Ewart, Taylor, Kraemer, & Agras, 1991; Gottman & Levenson, 1992; Kiecolt-Glaser et al.,
1996; Kiecolt-Glaser & Newton, 2001). In contrast, some research indicates that males mount a
greater stress response to social stress paradigms (Kudielka & Kirschbaum, 2005). Given the
inconsistent reporting of gender differences related to both HPA activity and social stress, the
present study includes gender as a possible moderator of the association between stress and HPA
activity.
The Present Study
Extensive research has demonstrated that stressful experiences are embedded in
physiology; diurnal functioning of the HPA axis, particularly the cortisol awakening response, is
one way in which these physiological changes can be observed. However, the existing literature
does not adequately address how past and current stress might differently impact the CAR,
particularly when past stress is not chronic, severe, or early in development. Additionally, little is
known about how relationships and emotion regulation might protect against the effects of stress.
Finally, both of the above questions are particularly under-studied among adolescents. The
ADOL STRESS AND CORTISOL AWAKENING

14
current study aims to fill these gaps by examining the relationship between past and current
stress and CAR among adolescents and then testing how each type of stress interacts with
possible protective factors, while also investigating the possible influences of gender.
First, we investigate how past and current stress are associated with CAR. Given the
literature suggesting that severe, early stress is associated with CAR attenuation and daily
stressors are associated with CAR exaggeration, we predict that past and current stress will have
opposite relationships with CAR, such that greater past stress is associated with a flatter
awakening increase while greater current stress is associated with a steeper awakening increase.
Second, we examine whether emotion regulation, attachment to parents, and attachment to peers
moderate the relationship between either past or current experiences of stress and CAR. We
predict that the three protective factors will act as buffers for the relationship between both types
of stress and current CAR. However, due to the shift in the locus of attachment experienced
during adolescence, we examine attachment to parents and peers separately. Finally, because
adolescence is a developmental period when females and males show differences in their social
functioning as well as in their HPA activity, we will explore whether any of the above
relationships vary by gender.
Method
Overview
The present study uses data from a longitudinal study that e x a mi ne s a doles c e nts’
exposure to violence as well as risk and resilience factors for adolescent adjustment. Participants
are a community sample of families who were recruited through flyers, newspaper
advertisements, and word-of-mouth. The larger study includes two cohorts of participants; the
first cohort includes families with a child age 9-10 at the start of data collection (n=119); the
ADOL STRESS AND CORTISOL AWAKENING

15
second cohort entered the study at the third wave of data collection, approximately 4 years later
(n=70). Participants in the second cohort had to have a child in middle school, to match the age
of the first cohort. Participation in both cohorts required that families lived together for at least 3
years before entering the study, that two parental figures resided in the home, and that all family
members could complete procedures in English (for further details, see Margolin, Vickerman,
Oliver, & Gordis, 2010). As part of a larger longitudinal study, participants visited the lab
approximately every two years. The present study utilized data from waves 3 and 5 of the
overall project. Wave 3 preceded wave 5 by approximately 5 years. Life stress data were
collected in waves 3 and 5; diurnal cortisol data and moderators were measured in wave 5. A
total of 131 adolescents participated in wave 5.
Participants
Of the 131 adolescents who participated in any procedures at wave 5, 99 youth (46
female) agreed to provide saliva samples as part of a home data collection procedure.
Participants were between 14 and 21 years old (M = 18.05, SD = 1.09). Participants identified as
37.4% Caucasian, 29.3% multi-racial, 20.2% Black/African-American, 5.1% Asian/Pacific
Islander, and 8.1% other/unknown; ethnically, 32.3% described themselves as Hispanic/Latino.
P a rtic ipants’ me a n house hold i nc ome w a s $91,828 (SD = $58,884); eight percent of participants
re porte d incom e be low t he pove rt y li ne . P a r e nts’ e duc a ti on va ri e d fr om 2 t o 20 years (M = 14.6,
SD = 2.8). In order to address the possibility of selective attrition, we conducted a number of
comparisons between the 99 participants who provided saliva samples and the other 32
adolescents who participated in wave 5 without providing saliva. In these comparisons, we found
no significant differences in ethnicity, race, a nnua l income , or p a r e nts’ leve l of e duc a ti on.
Procedures
ADOL STRESS AND CORTISOL AWAKENING

16
Participants completed both in-lab visits and home saliva collection. Home saliva
collection included providing five saliva samples per day and filling out questionnaires upon
provision of each saliva sample.
Home saliva collection. During their in-lab visits, participants were provided detailed
instructions for provision of saliva samples. Over three consecutive weekdays, participants were
asked to provide saliva samples five times per day: immediately upon awakening, twenty
minutes after awakening, forty minutes after awakening, at 4 P.M., and at 9 P.M. In order to
collect samples, experimenters provided partic ipa nts wi th a “ spit kit ,” whic h include d fif tee n
saliva collection swabs and storage tubes, fifteen questionnaires to complete upon provision of
each saliva sample, a thermal lunch box, an ice pack, and a watch with an alarm set to ring ten
minutes before each of the sampling times. The three morning sampling times were selected
ba se d on e a c h pa rtic ipant ’s t y pic a l we e kd a y ti me of a wa k e ning ;; t he ini ti a l a lar m wa s set no lat e r
than 8:50 A.M. The experimenter asked the participants about their normal time of awakening,
and set the watch alarms to allow for the specified times of saliva collection. The alarms for
samples one, four, and five were set 10 minutes before the scheduled saliva sampling time to
allow the participants to rinse their mouths with water to clear away debris and then to collect
saliva samples 10 minutes after rinsing.
After each saliva collection, participants were instructed to place the tubes in their home
freezers; if participants were away from home during one of the specified sampling times, they
were instructed to keep the sample in the lunchbox with an ice pack until they returned home.
Participants agreed not to consume caffeine or alcohol for 24 hours before beginning saliva
sampling. Additionally, they were instructed not to eat, drink, or brush their teeth before
ADOL STRESS AND CORTISOL AWAKENING

17
providing any of the three morning samples nor to eat, drink, or exercise for one hour before the
afternoon and evening samples.
Upon completion of saliva collection, participants either returned all materials to the lab,
or a researcher picked up materials from the participants at home. In the lab, saliva samples were
stored at -20 degrees Celsius before being shipped on dry ice for assay at Salimatrics, LLC (State
College, PA).
Measures
Diurnal cortisol. Saliva samples were assayed for cortisol in duplicate using a high-
sensitivity enzyme immunoassay, and the mean value of the two assays was used for analysis
(inter-assay correlation r(1,432) = .98, p < .001). For each time point, saliva samples that were
more than three standard deviations beyond the mean were considered outliers. A total of 15
samples from the first three time points (1.68% of all morning samples) were categorized as
outliers. After outliers were removed, we calculated the cortisol awakening response (CAR) by
subtracting the awakening cortisol value from whichever was higher of the second (awakening +
20 minutes) and third (awakening + 40 minutes) cortisol values of the day. CAR was
approximately normally distributed and did not require log transformation.
Covariates for cortisol analysis. In order to adjust for variables that may affect diurnal
cortisol, every day participants reported time of awakening and previous night hours of sleep.
These were entered in our models as continuous variables. Additionally, before each sample was
collected, participants reported whether they ate, drank, exercised, or had a mouth sore in the
prior hour; if they said yes to any of these activities at any data collection point, they were coded
as 1 for that day; if not, they were coded as 0. The large majority of participants (97.8%)
provided their awakening sample before 9 A.M. For 12% of samples, participants had consumed
ADOL STRESS AND CORTISOL AWAKENING

18
food or drink, exercised, or experienced a mouth sore in the hour before saliva provision. We
additionally included participant gender, age, medication use, and household income as
covariates. Thirteen percent of participants used medications, which included oral contraceptives
(6 participants), asthma medications (4 participants), thyroid medication (1 participant), and
medication to treat ADHD (2 participants).
I n o rde r to a djust fo r pa rt icipa nts’ sm oking , whic h c a n incr e a se c ortisol lev e ls (Granger et
al., 2007), the final saliva sample of each day was assayed for cotinine, a byproduct of nicotine
(Bramer & Kallungal, 2003). Eight participants had sufficiently elevated cotinine to suggest that
they had smoked nicotine (>10ng/ml; Caraballo, Giovino, & Pechacek, 2004). As over 85% of
adolescents had at least trace levels of cotinine in their saliva samples, cotinine level was
included as a covariate for all participants.
Life stress. During both wave 3 and wave 5, participants reported past year life stress
using the Life Events Checklist (LEC; modified from Johnson & McCutcheon, 1980). The LEC
presents participants with fifty-six items and asks them to rate whether each event occurred in
the past year, whether the event was good or bad, and to what extent the event affected the
participant (rated on a 1-4 Likert scale from No effect to Large effect). Items primarily focused
on stressors that children and adolescents are likely to experience in the course of typical
development (e.g., Changing to a new school, Trouble with friends), although the measure also
included a small number of major stressors (e.g., Death of family member or close family friend,
Parent going to jail). The measure was scored by counting only items that the participant
endorsed as having a negative effect; each negatively-rated item was multiplied by the extent to
which the participant reported that the event affected his/her life. Current stress is
operationalized as wave 5 LEC, while past stress is operationalized as wave 3 LEC.
ADOL STRESS AND CORTISOL AWAKENING

19
Emotion Regulation. In wave 5 of data collection, participants additionally reported
their emotion regulation using the Emotions Checklist (Shields & Cicchetti, 1997). Participants
we re a ske d to ra t e ten ite ms about t he ir ow n re g ul a tor y a bil it ies, suc h a s “ Overreacts to minor
fr ustra ti on” on a 4-point Likert scale ranging from Never true (0) to Almost always true (3).
Emot ion R e g ulation wa s sc ore d usi n g the me a n fr om all i tems (Cronba c h’s α = .692).
Attachment to parents and peers. During wave 5, participants reported attachment to
parents and peers using the Inventory of Parent and Peer Attachment (IPPA; Armsden &
Greenberg, 1987). The IPPA is a self- re port me a s ure of a dol e sc e nts’ p e rc e ived a tt a c hment
security in relationships with both caregivers and friends. Items are asked separately about
relationships with parents and relationships with peers, such that the IPPA yields independent
measures of attachment security for each group. Participants rate their agreement with each of 53
statements (28 items assessing perceived attachment to parents, 25 items assessing perceived
attachment to peers), such as “ I c a n c o unt on my parents when I need to get something off my
c he st,” on a 5-point Likert scale, ranging from Almost Never or Never (0) to Almost Always or
Always (4). Attachment to parents and attachment to peers was scored by using the means of
items assessing relationships with parents (Cron b a c h’s α = .900) and relationships with peers
(Cron ba c h’s α = .904), respectively.
Analytic Plan
We tested our hypotheses using a series of multi-level models. Analyses were conducted
using Mplus Version 6.12 (Muthén & Muthén, 2010). Although missingness of cortisol data was
low (3.6%), as was missingness for all other predictors and covariates of interest (4.4%), we
imputed missing values using FIML (Schafer & Graham, 2002). We conducted analyses using
multi-level models because we are interested in adjusting for covariates that vary either within
ADOL STRESS AND CORTISOL AWAKENING

20
individuals or between individuals. Some covariates of interest, such as eating, drinking, exercise,
cotinine concentration, hours of sleep, and time of awakening vary within individuals (level 1)
such that the same person could have a different value for each of these constructs depending on
the day. Other variables of interest, such as family income, age, medication status, gender, and
past-year stress, vary between individuals (level 2) such that the same person will have the same
value on all days of measurement, but different individuals will have different values. We
initially tested gender as a moderator of each hypothesized association (dummy coded such that
female = 1, male = 2). However, when gender was not a significant moderator, we re-ran the
analyses with gender as a covariate for purposes of parsimony and to increase our power. When
the analyses established that gender was a significant moderator, we re-ran the analyses
separately for males and females for better interpretation of the data.
For our first hypothesis, we tested the association between both past and current stress
using the following model:
Level 1: Y
ij
= β
0j
+ β
1j
(awakening time) + β
2j
(hours of sleep) + β
3j
(cotinine) +
β
4j
(whether ate/drank/exercised/had mouth sore) + e
ij

Level 2: β
0j
= γ
00
+ γ
01
(past life stress) + γ
02
(gender) + γ
03
(past stress x gender) +
γ
04
(current life stress) + γ
05
(age) + γ
06
(medication) + γ
07
(family income) + u
0j

For our second hypothesis, we tested whether the association between life stress and CAR
is moderated by emotion regulation, attachment to parents, or attachment to peers. Because we
were interested in first examining the possible moderating role of gender, we began by testing
our moderators in three-way interactions; if these interactions were not significant, we instead
tested moderation with gender treated as a covariate. For analyses that focused on the role of past
stress we adjusted for current stress; for analyses that focused on current life stress, we adjusted
ADOL STRESS AND CORTISOL AWAKENING

21
for past life stress. We examined how each of our moderators interacts with past stress and
current stress in separate equations. The following equations demonstrate our models using past
stress:
Level 1: Y
ij
= β
0j
+ β
1j
(awakening time) + β
2j
(hours of sleep) + β
3j
(cotinine) +
β
4j
(whether ate/drank/exercised/had mouth sore) + e
ij

Level 2: β
0j
= γ
00
+ γ
01
(past life stress) + γ
02
(moderator
1
) + γ
03
(gender) + γ
04
(past
life stress x moderator) + γ
05
(past stress x gender) + γ
06
(moderator x gender) + γ
07
(past
life stress x moderator x gender) + γ
08
(current life stress) + γ
09
(age) + γ
010
(medication) +
γ
011
(family income) + u
0j

Results
Inter-relationships among variables
Table 1 presents the means and standard deviations of all variables of interest. Across all
participants, mean levels of current stress were significantly higher than mean levels of past
stress, t(276) = -3.974, p < .001. Additionally, males reported significantly less past stress, t(246)
= 3.302, p = .001, and had a significantly flatter CAR, t(257) = 3.015, p = .003, than did females.
Finally, females were more likely to be on medication, t(267) = 2.268, p = .024, and less likely to
have high salivary concentrations of cotinine, t(170) = -2.753, p = .007.
Table 2 presents Pearson correlations among all variables used in this study for the entire
sample. For within-person variables, we used the mean values across the three days of data
collection. Current stress is positively associated with CAR, even before accounting for
covariates. Because gender was significantly associated with CAR and marginally associated
with past stress, we examined intercorrelations among variables for males and females separately

1
Attachment to parents, attachment to peers, or emotion regulation
ADOL STRESS AND CORTISOL AWAKENING

22
(see Table 3). For females, current stress is marginally positively correlated with CAR and past
stress is not significant. In contrast, for males, past stress is negatively correlated with CAR and
current stress is not significant. In addition, past stress is associated with current stress for
females but not males. Females also show significant correlations among the three hypothesized
buffer variables (emotion regulation, parent attachment and peer attachment) whereas males do
not.
Hypothesis 1: Past vs. current life stress
We examined the relationship between life stress and HPA activity using a multi-level
model, which included current and past life stress simultaneously, as well as all covariates that
may affect cortisol (age, medication, family income, awakening time, hours of sleep, cotinine
level, eating/drinking/exercising/mouth sores). All analyses first test for gender as a moderator;
when gender is not a significant moderator, it is instead entered as a covariate.
The interaction between current life stress and gender was not significant ( γ = -.181, SE
= .263, p = .491); this analysis yielded a marginally significant main effect of current stress on
CAR (γ = .451, SE=.232, p=.052). When the interaction term was removed and gender was
instead treated as a covariate, there was a main effect of current life stress on CAR ( γ = .306,
SE=.116, p=.008, see Table 4), such that adolescents who are experiencing more life stress have
a steeper morning cortisol increase. Figure 1 shows diurnal cortisol patterns for adolescents who
experience high vs. low levels of current stress.
Gender did significantly interact with past stress ( γ = -.674, SE = .081, p < .001, see Table
5), such that there was a significant main effect of past stress ( γ = .494, SE = .143, p = .001) but
no significant main effect of gender ( γ = .047, SE = .133, p = .724). We then analyzed the effects
of past stress for males and females separately. Past stress was significantly associated with CAR
ADOL STRESS AND CORTISOL AWAKENING

23
for males ( γ = -.727, SE = .192, p < .001, see Table 5), but not for females (γ = -.051, SE = .202,
p = .802), such that males who report more past stress exhibit a flatter CAR. Figure 2 shows the
diurnal cortisol patterns for males and females who report high or low past stress.
Hypothesis 2a: Moderating the Relationship Between Current Stress and CAR
No moderation was found for the relationship between current stress and CAR. Emotion
regulation did not interact with current stress ( γ = .289, SE = .271, p = .288), nor was there a
three-way interaction between emotion regulation, current stress, and gender ( γ = -.065, SE
= .335, p = .845). Similarly, for attachment to parents, there was no significant two-way
(attachment to parents x current stress, γ = -.293, SE = .212, p = .167) or three-way (attachment
to parents x current stress x gender, γ = -.131, SE = .178, p = .461) interaction. Finally, there
were no significant interactions between current stress and attachment to peers (attachment to
peers x current stress, γ = .221, SE = .293, p = .450; attachment to peers x current stress, gender,
γ = -.088, SE = .265, p = .740).
Hypothesis 2b: Moderating the Relationship Between Past Stress and CAR
Emotion Regulation. Next, we examined whether emotion regulation moderated the
associations between past stress and CAR. We began by running a three-way interaction between
past stress, emotion regulation, and gender, which was marginally significant ( γ = -.615, SE
= .328, p = .061, see Table 6). We therefore analyzed the interaction between past stress and
emotion regulation separately for males and females. For males, emotion regulation significantly
interacted with past stress ( γ = -.778, SE = .055, p < .001, see Table 7). Simple slope analyses
reveal that for males one standard deviation below the mean in emotion regulation, past stress is
associated with an increased CAR ( Β = 1.245, p < .001), but for males one standard deviation
above the mean, the association between past stress and CAR is not significant ( Β = -.311, p
ADOL STRESS AND CORTISOL AWAKENING

24
= .252, see Figure 3). For females, the interaction between past stress and emotion regulation
was not significant (γ = -.445, SE = .449, p = .322).
Attachment to Parents. We then examined whether attachment to parents interacted
with past stress in its association with diurnal HPA activity. The three-way interaction (stress x
attachment to parents x gender) was significant ( γ = -.606, SE = .219, p = .006, see Table 6).
Therefore, we examined the interaction between past stress and attachment to parents in males
and females separately. Results of these analyses were significant for males ( γ = -.804, SE = .048,
p < .001, see Table 8) but not for females (γ = .021, SE = 1.085, p = .985). Male adolescents who
are low in attachment to peers show greater CAR in association with greater past stress (Β =
1.290, p < .001); in contrast, males who are high in attachment to peers show some CAR
attenuation associated with increased past stress ( Β = -.318, p = .006, see Figure 4).
Attachment to Peers. Finally, we examined whether attachment to peers interacted with
past stress in its association CAR. The three-way interaction (past stress x attachment to peers x
gender) was not significant ( γ = -.051, SE = .274, p = .853, see Table 6). However, when gender
was treated as a covariate instead of a moderator, attachment to peers significantly interacted
with past stress (γ = .616, SE = .082, p < .001, see Table 9). Adolescents who are low in
attachment to peers show smaller CARs in association with greater past stress ( Β = -1.298, p
< .001), but the relationship between past stress and CAR is not significant for adolescents high
in attachment to peers ( Β = .325, p = .429, see Figure 5).
Discussion
The present study examined the relationship between past stress, current stress, and CAR
among adolescents. Across the sample, current stress was associated with an increased CAR. For
males only, past stress was associated with an attenuated CAR. When we assessed whether
ADOL STRESS AND CORTISOL AWAKENING

25
emotion regulation, attachment to parents, and attachment to peers moderated the association
between stress and CAR, no protective factors were found for current stress. For past stress,
emotion regulation and attachment to parents were protective for males only. Young men who
were low in emotion regulation or attachment to parents had a steeper CAR in association with
greater past stress, while males who were high in these constructs demonstrated either no
relationship or a small negative relationship between past stress and CAR. For both male and
female adolescents, high attachment to peers was protective for past stress; however, youths who
did not report close relationships with peers showed a flatter CAR with higher past stress.
Our study identified two distinct patterns of CAR associated with risk: exaggeration and
attenuation. Prior research has identified both very flat and very steep morning increases in CAR
as deleterious, and a meta-analysis suggests that different risk factors are associated with these
two types of CAR dysregulation (Chida & Steptoe, 2009). First, in our sample, current stress is
associated with increased CAR; a finding that is consistent with a range of studies suggesting
that anticipating a stressful day activates greater CAR increases (e.g., Adam et al., 2006). In
contrast, for male youth past stress is associated with reduced CAR. Although prior studies have
shown an association between extreme forms of early adversity and later CAR attenuation (e.g.,
Quevedo et al., 2012), the present study is among the first to demonstrate this relationship for
prior experiences of more developmentally-typical stressors. In our sample, having more
stressors such as performing poorly at school or not getting along with siblings only five years
prior to saliva collection is associated with decreased CAR for males. This study is notable both
for the relatively mild types of prior stressors associated with CAR attenuation and for the
relatively short period of time in which attenuation occurred. Whereas many studies have
examined the lasting effects of stressors experienced in early childhood, our past stress measure
ADOL STRESS AND CORTISOL AWAKENING

26
was administered in late childhood/early adolescence, at an average of age 12.69. Moreover, our
stud y c on c e ptualiz e d gre a ter stre ss a c c o rding to a dolesc e nts’ subj e c ti ve re port of how muc h a variety of difficulties affected them rather than merely counting the number of difficulties the
adolescent had experienced or focusing on only one type of stressor.
When we added protective factors to our models, the pattern of CAR dysregulation
associated with past stress changed. For male adolescents who are low in emotion regulation or
who report low-quality relationships with parents, greater experiences of past stress are
associated with a larger CAR. This pattern runs opposite to the association between greater past
stress and reduced CAR in our larger sample of males. For the subset of adolescents who are low
in these protective factors, historical stress has the same positive association with CAR as does
current stress; without the benefit of good regulatory abilities and supportive relationships with
parents, stressful events that occurred years in the past continue to affect adolescents just like
stressors that are presently ongoing. HPA attenuation in response to a history of high stress may
be adaptive (Saxbe, Margolin, Spies Shapiro, & Baucom, 2012); indeed, even male adolescents
who reported high attachment to parents showed at least some attenuation associated with greater
past stress. However, without good regulatory ability and close relationships with parents, this
adaptive physiological process may not occur.
Our study is unique for its examination of relationships with friends as a protective factor
for adolescent HPA functioning. As a protective factor, relationships with peers differ from
relationships with parents and from emotion regulation in several important respects. First,
a tt a c hment to par e nts an d e mot ion re g ul a ti on e mer g e e a rlie r in a n a dolesc e nts’ li fe a nd show
greater stability over time (Waters, Weinfield, & Hamilton, 2000); in contrast, relationships with
peers (including both how much closeness the adolescent feels with peers and with which peers
ADOL STRESS AND CORTISOL AWAKENING

27
the adolescents feel close) are likely to significantly shift from early to late adolescence. Second,
adolescents may play a more active role in creating their relationships with peers. Whereas
nearly every adolescent has caregivers with whom to form an attachment bond, creating peer
relationships requires effortful engagement with the social world. Therefore, we predicted that
attachment to peers might have a different association with CAR than does attachment to parents.
The present study is the first to examine the moderating role of adolescent friendship on CAR;
therefore there was little precedent in the literature to guide our hypotheses.
I nd e e d, w e foun d that a d olesc e nts’ a tt a c hment to pee rs pla y e d a diff e re nt r ole tha n did
attachment to parents or emotion regulation. Although adolescents high in peer attachment
showed a similar buffering effect, such there was no relationship between past stress and CAR
for youth high in peer attachment, adolescents low in peer attachment appeared to be a distinct
risk group. Whereas adolescents low in emotion regulation or low in attachment to parents
showed increasing CAR associated with increasing past stress, adolescents low in attachment to
peers showed the opposite pattern: more past stress was associated with a reduced CAR. This
may be because adolescents who do not have close friendships are not as involved with the world
around them, and steeper CARs serve to prepare the body to engage the day (McEwen, 2000).
Being both high in past stress and without close peer relationships may represent a socially
disengaged, withdrawn pattern, which is associated with attenuated HPA activity (Ben-Dat
Fischer et al., 2007). Because HPA activity and attachment to friends was measured
simultaneously, this study cannot speculate as to whether a lack of close friend relationships
contributed to attenuation or whether a flat CAR made adolescents less likely to engage socially.
Additionally, peer relationships emerge later and show less stability than do relationships
with ca re g ive rs or e mot i on re g ulation. Adolesc e n ts’ pa tt e rn of a tt a c hment to pee rs ma y ha v e
ADOL STRESS AND CORTISOL AWAKENING

28
changed considerably in the five years between the measurement of past stress and the
measurement of salivary cortisol. In contrast, those adolescents who were high in attachment to
parents and emotion regulation were also likely to have been high in these constructs at the time
the past stress occurred. Because our study only measured protective factors concurrently with
CAR, we cannot examine whether adolescents who had protective factors at the time the stress
occurred show different HPA activity than do adolescents who only develop protective factors
after the stress passed. Future studies should examine the different roles of social support at the
time the stressors occur vs. social support at the time HPA functioning is measured. Such studies
could uncover whether there are critical periods for the protective influence of relationships or
whe ther late r de v e lopi ng supportive r e lations hips ca n “ undo ” the e f fe c ts of e a rl y stre ss.
Notably, we were unable to identify any protective factors for the association between
experiencing current stress and having an elevated CAR. Adolescents who have supportive
relationships and good regulatory abilities are buffered from the effects of stress in the long-term,
but in our sample, even the adolescents who had high levels of protective factors showed an
increased CAR in the face of current stress. These findings highlight the adaptive role an
increase in CAR plays in facing short-term stress, as elevated morning cortisol may help an
adolescent prepare to face a challenging day. Indeed, having an HPA axis that flexibly adapts to
changing stressors indicates good, rather than poor health (Mikolajczak et al., 2010). However,
in our sa mpl e a dolesc e nts who a re hi g h in prote c ti ve fa c to rs a r e e ve ntu a ll y a ble to “ re c ove r” from the effects of stress, whereas adolescents who are low in protective factors appear to be
affected by past stressors as though they were still occurring in the present.
Across our analyses, we observed several gender differences. First, on average, males in
our sample reported less past stress and had a flatter CAR. Additionally, not only did males alone
ADOL STRESS AND CORTISOL AWAKENING

29
show a relationship between past stress and CAR, but it was only for males that the protective
role of attachment to parents and emotion regulation held. These gender differences were
unexpected because the literature does not consistently indicate that males are more susceptible
to the effects of stress (e.g., Stroud, Salovey, & Epel, 2002). Similarly, there is mixed evidence
as to whether social support buffers are more or less effective for males (Shumaker & Hill, 1991).
Doom et al. (2013) recently found an interaction between childhood maltreatment history and
gender in its association with diurnal HPA activity; maltreated boys exhibited elevated diurnal
cortisol patterns as compared to maltreated girls, a finding which ran counter to their hypotheses.
Our finding that past stress appears to have a greater legacy for boys than for girls may
reflect the time frame during which our past stress data was collected. During wave 3, our
participants had a mean age of 12.69 years. On average, girls begin puberty at an earlier age than
do boys; thus it is likely that a greater proportion of girls than boys in our sample had begun
puberty at the time our past stress measure was collected. Puberty is a time of HPA axis
reorganization and consolidation (Gunnar & Quevedo, 2007). According to allostatic load theory,
stressors that occur earlier in development should have a greater impact on physiology (Repetti
et al., 2011). Although there were not significant differences between the average ages of males
and females at the time the past stress measure was collected, boys may have been
physiologically less mature at the time they experienced the stressors; therefore, these stressors
would have exerted a larger influence on later HPA functioning. Future studies should examine
whether experiencing stressors shortly before or after the onset of puberty is associated with
different patterns of HPA activity in late adolescence and early adulthood. Additionally, this
study underscores the importance of examining gender not just as a covariate but as a potential
ADOL STRESS AND CORTISOL AWAKENING

30
moderator; future studies of adolescent HPA activity should examine whether findings differ
between boys and girls.
Several limitations of this study should be noted. First, our measures of emotion
regulation and attachment relied on self-report. Future studies should measure these constructs
more comprehensively, including physiological measures of emotion regulation (e.g., vagal tone,
Porges, Doussard-Roosevelt, & Maiti, 1994) and interview measures of attachment (e.g., Adult
Attachment Interview, George, Kaplan, & Main, 1984). Additionally, all adolescents were
sa mpl e d on we e kd a y s. A lt houg h thi s de sig n max im iz e s the c onsi stenc y of a dolesc e nts’
awakening time, it prevents us from examining how CAR differs between weekdays and
weekends. Parent and peer attachment may play different roles on days when adolescents are
anticipating spending more time with peers (e.g., at school on weekdays) versus with parents
(e.g., at home on weekends). Furthermore, our study measured current stress as an overall report
of the past year; we did not measure stress or anticipation of stressful events on the day level.
However, the CAR is sensitive to changes in perceived stress from one day to the next; our study
was unable to adjust for these day-today variations. Finally, we were unable to adjust for
pubertal status in our sample. Although our youth were sufficiently old (mean age of 18) that
they all likely completed puberty before HPA measures were collected, we do not have
information regarding pubertal status at the time past stress was measured. Future longitudinal
studi e s shoul d mea sure a dolesc e nts’ pube rta l deve lopm e nt at a ll ti me points of da ta c oll e c ti on.
Despite these limitations, the present study has several unique strengths. First, in contrast
to prior studies that have focused on either only one type of stressor (e.g., bullying, reviewed in
Murray-Close, 2012), only severe stress (e.g., deprivation and neglect, Quevedo et al., 2012), or
only stress that occurred early in development (e.g., time spent in daycare at age three, Roisman
ADOL STRESS AND CORTISOL AWAKENING

31
et al., 2009), our investigation measures a range of stressors which occurred only five years prior,
most of which are a typical part of adolescent development (e.g., conflict with friends, doing
poorl y in a c lass) . W e c a lcula te total st re ss base d o n a dolesc e nts’ subj e c ti ve re ports of how muc h
each stressor affected them. Even though past stress was measured relatively late in development
(early adolescence), a greater history of stress was nevertheless associated with attenuated CAR
for males, the same pattern that is seen for early/severe stress. Our study indicates that even
typical stressors have long-term implications for physiological regulation. Second, we assessed
the same stresses at two points in time, which has not been reported in prior studies. Based on
the low association between past and present stress, we know that we are not just measuring the
continuation of the same phenomena over time. Third, our study is the first to examine how
re lations hips wi th fr iends buff e r the e ff e c t of str e s s on a dolesc e nts’ C AR. Adole sc e nc e is a ti me
of reorganization of the social world, such that relationships with peers take on greater
importance and the influence of parents diminishes (Harris, 1995; Hartup, 1996; Hunter &
Youniss, 1982). Therefore, examining the moderating role of peer relationships is an important
first step towards developing a comprehensive understanding of how friends influence HPA
activity in adolescence. Finally, our study identified three factors that moderate the association
between past stress and CAR dysregulation. As stressful experiences are a common part of
adolescent development, identifying protective factors is a first step towards developing
interventions to reduce the impact of stress on HPA functioning. In the long term, HPA
dysregulation is associated with a range of disorders, including heart disease and depression
(Adam et al., 2010; Kumari et al., 2011). Clinicians who are aware of the protective effects of
emotion regulation, attachment to parents, and attachment to peers can intervene to strengthen
these factors for adolescents who have histories of stress; intervention administered during
ADOL STRESS AND CORTISOL AWAKENING

32
adolescence may prevent the development of adverse physical and mental health sequelae of
stress.

ADOL STRESS AND CORTISOL AWAKENING

33
References
Adam, E. K. (2006). Transactions among adolescent trait and state emotion and diurnal and
momentary cortisol activity in naturalistic settings. Psychoneuroendocrinology, 31(5),
664-679. doi: 10.1016/j.psyneuen.2006.01.010
Adam, E. K., Doane, L. D., Zinbarg, R. E., Mineka, S., Craske, M. G., & Griffith, J. W. (2010).
Prospective prediction of major depressive disorder from cortisol awakening responses in
adolescence. Psychoneuroendocrinology, 35(6), 921-931. doi:
10.1016/j.psyneuen.2009.12.007
Adam, E. K., Hawkley, L. C., Kudielka, B. M., & Cacioppo, J. T. (2006). Day-to-day dynamics
of experience--cortisol associations in a population-based sample of older adults. Proc
Natl Acad Sci U S A, 103(45), 17058-17063. doi: 10.1073/pnas.0605053103
Adams, R. E., Santo, J. B., & Bukowski, W. M. (2011). The presence of a best friend buffers the
effects of negative experiences. Dev Psychol, 47(6), 1786-1791. doi: 10.1037/a0025401
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of Attachment: A
Psychological Study of the Strange Situation. Hillsdale, NJ: Erlbaum.
Armsden, G. C., & Greenberg, M. T. (1987). The inventory of parent and peer attachment:
Individual differences and their relationship to psychological well-being in adolescence. J
Youth Adolesc, 16(5), 427-454. doi: 10.1007/BF02202939
Bartels, M., de Geus, E. J., Kirschbaum, C., Sluyter, F., & Boomsma, D. I. (2003). Heritability of
daytime cortisol levels in children. Behav Genet, 33(4), 421-433.
Ben-Dat Fischer, D., Serbin, L. A., Stack, D. M., Rutte, P. L., Ledingham, J. E., & Schwartzman,
A. E. (2007). Intergenerational predictors of diurnal cortisol secretion in early childhood.
Infant and Child Development, 16, 151-170.
ADOL STRESS AND CORTISOL AWAKENING

34
Bevans, K., Cerbone, A., & Overstreet, S. (2008). Relations between recurrent trauma exposure
and recent life stress and salivary cortisol among children. Dev Psychopathol, 20(1), 257-
272. doi: 10.1017/S0954579408000126
Booth, A., Granger, D. A., Shirtcliff, E. A. (2008). Gender- and age-related differences in the
association between social relationship quality and trait-levels of salivary cortisol.
Journal of Research on Adolescence, 18, 239-260.
Bouma, E. M., Riese, H., Ormel, J., Verhulst, F. C., & Oldehinkel, A. J. (2009). Adolescents'
cortisol responses to awakening and social stress; effects of gender, menstrual phase and
oral contraceptives. The TRAILS study. Psychoneuroendocrinology, 34(6), 884-893. doi:
10.1016/j.psyneuen.2009.01.003
Bowlby, John. (1969). Attachment and loss. New York,: Basic Books.
Bramer, S. L., & Kallungal, B. A. (2003). Clinical considerations in study designs that use
cotinine as a biomarker. Biomarkers, 8(3-4), 187-203. doi: 10.1080/13547500310012545
Caraballo, R. S., Giovino, G. A., & Pechacek, T. F. (2004). Self-reported cigarette smoking vs.
serum cotinine among U.S. adolescents. Nicotine Tob Res, 6(1), 19-25. doi:
10.1080/14622200310001656821
Cassidy, J. (1994). Emotion regulation: Infulences of attachment relationships. In N. Fox (Ed.),
The development of emotion regulation: Biological and behavioral considerations.
Monographs of the Society for Research in Child Development (pp. 228-283).
Chida, Y., & Steptoe, A. (2009). Cortisol awakening response and psychosocial factors: a
systematic review and meta-analysis. Biol Psychol, 80(3), 265-278. doi:
10.1016/j.biopsycho.2008.10.004
ADOL STRESS AND CORTISOL AWAKENING

35
Cicchetti, D., Rogosch, F. A., Toth, S. L., & Sturge-Apple, M. L. (2011). Normalizing the
development of cortisol regulation in maltreated infants through preventive interventions.
Dev Psychopathol, 23(3), 789-800. doi: 10.1017/S0954579411000307
Coe, C. L., Franklin, D., Smith, E. R., & Levine, S. (1982). Hormonal responses accompanying
fear and agitation in the squirrel monkey. Physiol Behav, 29(6), 1051-1057.
Coe, C. L., Mendoza, S. P., Smotherman, W. P., & Levine, S. (1978). Mother-infant attachment
in the squirrel monkey: adrenal response to separation. Behav Biol, 22(2), 256-263.
Conrad, C. D., Jackson, J. L., Wieczorek, L., Baran, S. E., Harman, J. S., Wright, R. L., & Korol,
D. L. (2004). Acute stress impairs spatial memory in male but not female rats: influence
of estrous cycle. Pharmacol Biochem Behav, 78(3), 569-579. doi:
10.1016/j.pbb.2004.04.025
Devereux, J., Rydstedt, L. W., & Cropley, M. (2011). An exploratory study to assess the impact
of work demands and the anticipation of work on awakening saliva cortisol. Psychol Rep,
108(1), 274-280.
Doane, L. D., & Zeiders, K. H. (2014). Contextual moderators of momentary cortisol and
negative affect in adolescents' daily lives. J Adolesc Health, 54(5), 536-542. doi:
10.1016/j.jadohealth.2013.10.007
Dozier, M., Peloso, E., Lewis, E., Laurenceau, J. P., & Levine, S. (2008). Effects of an
attachment-based intervention on the cortisol production of infants and toddlers in foster
care. Dev Psychopathol, 20(3), 845-859. doi: 10.1017/S0954579408000400
Edwards, S., Evans, P., Hucklebridge, F., & Clow, A. (2001). Association between time of
awakening and diurnal cortisol secretory activity. Psychoneuroendocrinology, 26(6),
613-622.
ADOL STRESS AND CORTISOL AWAKENING

36
Ellenbogen, M. A., & Hodgins, S. (2009). Structure provided by parents in middle childhood
predicts cortisol reactivity in adolescence among the offspring of parents with bipolar
disorder and controls. Psychoneuroendocrinology, 34(5), 773-785. doi:
10.1016/j.psyneuen.2008.12.011
Engert, V., Efanov, S. I., Dedovic, K., Dagher, A., & Pruessner, J. C. (2011). Increased cortisol
awakening response and afternoon/evening cortisol output in healthy young adults with
low early life parental care. Psychopharmacology (Berl), 214(1), 261-268. doi:
10.1007/s00213-010-1918-4
Essex, M. J., Shirtcliff, E. A., Burk, L. R., Ruttle, P. L., Klein, M. H., Slattery, M. J., . . .
Armstrong, J. M. (2011). Influence of early life stress on later hypothalamic-pituitary-
adrenal axis functioning and its covariation with mental health symptoms: a study of the
allostatic process from childhood into adolescence. Dev Psychopathol, 23(4), 1039-1058.
doi: 10.1017/S0954579411000484
Ewart, C. K., Taylor, C. B., Kraemer, H. C., & Agras, W. S. (1991). High blood pressure and
marital discord: not being nasty matters more than being nice. Health Psychol, 10(3),
155-163.
Fisher, P. A., Stoolmiller, M., Gunnar, M. R., & Burraston, B. O. (2007). Effects of a therapeutic
intervention for foster preschoolers on diurnal cortisol activity.
Psychoneuroendocrinology, 32(8-10), 892-905. doi: 10.1016/j.psyneuen.2007.06.008
Fraley, C. R. , & Davis, K. E. (1997). Attachment formation and transfer in young adults' close
friendships and romantic relationships. Personal Relationships, 4, 131-144.
ADOL STRESS AND CORTISOL AWAKENING

37
Fries, E., Dettenborn, L., & Kirschbaum, C. (2009). The cortisol awakening response (CAR):
facts and future directions. Int J Psychophysiol, 72(1), 67-73. doi:
10.1016/j.ijpsycho.2008.03.014
George, C., Kaplan, M., & Main, M. (1984). Adult Attachment Interview. University of
California, Berkeley.
Gottman, J. M., & Levenson, R. W. (1992). Marital processes predictive of later dissolution:
behavior, physiology, and health. J Pers Soc Psychol, 63(2), 221-233.
Granger, D. A., Blair, C., Willoughby, M., Kivlighan, K. T., Hibel, L. C., Fortunato, C. K., . . .
Family Life Project, Investigators. (2007). Individual differences in salivary cortisol and
alpha-amylase in mothers and their infants: relation to tobacco smoke exposure. Dev
Psychobiol, 49(7), 692-701. doi: 10.1002/dev.20247
Gunnar, M. R., Brodersen, L., Nachmias, M., Buss, K., & Rigatuso, J. (1996). Stress reactivity
and attachment security. Dev Psychobiol, 29(3), 191-204. doi: 10.1002/(SICI)1098-
2302(199604)29:3<191::AID-DEV1>3.0.CO;2-M
Gunnar, M. R., & Quevedo, K. (2007). The neurobiology of stress and development. Annu Rev
Psychol, 58, 145-173. doi: 10.1146/annurev.psych.58.110405.085605
Gustafsson, P. A., Gustafsson, P. E., Anckarsater, H., Lichtenstein, P., Ljung, T., Nelson, N., &
Larsson, H. (2011). Heritability of cortisol regulation in children. Twin Res Hum Genet,
14(6), 553-561.
Gustafsson, P. E., Anckarsater, H., Lichtenstein, P., Nelson, N., & Gustafsson, P. A. (2010).
Does quantity have a quality all its own? Cumulative adversity and up- and down-
regulation of circadian salivary cortisol levels in healthy children.
Psychoneuroendocrinology, 35(9), 1410-1415. doi: 10.1016/j.psyneuen.2010.04.004
ADOL STRESS AND CORTISOL AWAKENING

38
Hackman, D. A., Betancourt, L. M., Brodsky, N. L., Kobrin, L., Hurt, H., & Farah, M. J. (2013).
Selective impact of early parental responsivity on adolescent stress reactivity. PLoS One,
8(3), e58250. doi: 10.1371/journal.pone.0058250
Hanson, M. D., & Chen, E. (2010). Daily stress, cortisol, and sleep: the moderating role of
childhood psychosocial environments. Health Psychol, 29(4), 394-402. doi:
10.1037/a0019879
Harris, J. (1995). Where is the child's environment? A group socialization theory of development.
Psychological Review, 2, 458-489.
Hartup, W. W. (1996). The company they keep: friendships and their developmental significance.
Child Dev, 67(1), 1-13.
Hostinar, C. E., & Gunnar, M. R. (2013). Future directions in the study of social relationships as
regulators of the HPA axis across development. J Clin Child Adolesc Psychol, 42(4),
564-575. doi: 10.1080/15374416.2013.804387
Hostinar, C. E., Sullivan, R. M., & Gunnar, M. R. (2014). Psychobiological mechanisms
underlying the social buffering of the hypothalamic-pituitary-adrenocortical axis: a
review of animal models and human studies across development. Psychol Bull, 140(1),
256-282. doi: 10.1037/a0032671
Hunter, F. T., & Youniss, J. (1982). Changes in functions of three relations during adolescence.
Developmental Psychology, 18, 806-811.
Johnson, J. H., & McCutcheon, S. M. (1980). Assessing life stress in older children and
adolescents: preliminary findings with the Life Events Checklist. In I. G. Sarason & C. D.
Spielberger (Eds.), Stress and Anxiety. Washington, D.C.: Hemisphere.
ADOL STRESS AND CORTISOL AWAKENING

39
Kiecolt-Glaser, J. K., Newton, T., Cacioppo, J. T., MacCallum, R. C., Glaser, R., & Malarkey, W.
B. (1996). Marital conflict and endocrine function: are men really more physiologically
affected than women? J Consult Clin Psychol, 64(2), 324-332.
Kiecolt-Glaser, J. K., & Newton, T. L. (2001). Marriage and health: his and hers. Psychol Bull,
127(4), 472-503.
Kliewer, W., Reid-Quinones, K., Shields, B. J., & Foutz, L. (2009). Multiple risks, emotion
regulation skill, and cortisol in low-income African American youth: A prospective study.
Journal of Black Psychology, 35, 24-43.
Kudielka, B. M., & Kirschbaum, C. (2005). Sex differences in HPA axis responses to stress: a
review. Biol Psychol, 69(1), 113-132. doi: 10.1016/j.biopsycho.2004.11.009
Kumari, M., Shipley, M., Stafford, M., & Kivimaki, M. (2011). Association of diurnal patterns in
salivary cortisol with all-cause and cardiovascular mortality: findings from the Whitehall
II study. J Clin Endocrinol Metab, 96(5), 1478-1485. doi: 10.1210/jc.2010-2137
Kupper, N., de Geus, E. J., van den Berg, M., Kirschbaum, C., Boomsma, D. I., & Willemsen, G.
(2005). Familial influences on basal salivary cortisol in an adult population.
Psychoneuroendocrinology, 30(9), 857-868. doi: 10.1016/j.psyneuen.2005.04.003
Levine, S. (2001). Primary social relationships influence the development of the hypothalamic--
pituitary--adrenal axis in the rat. Physiol Behav, 73(3), 255-260.
Lovallo, W. R., Farag, N. H., Sorocco, K. H., Cohoon, A. J., & Vincent, A. S. (2012). Lifetime
adversity leads to blunted stress axis reactivity: studies from the Oklahoma Family Health
Patterns Project. Biol Psychiatry, 71(4), 344-349. doi: 10.1016/j.biopsych.2011.10.018
ADOL STRESS AND CORTISOL AWAKENING

40
Lyons, D. M., Parker, K. J., Katz, M., & Schatzberg, A. F. (2009). Developmental cascades
linking stress inoculation, arousal regulation, and resilience. Front Behav Neurosci, 3, 32.
doi: 10.3389/neuro.08.032.2009
Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A
move to the level of representation. Monographs of the Society for Research in Child
Development, 50, 66-106.
Margolin, G., Vickerman, K. A., Oliver, P. H., & Gordis, E. B. (2010). Violence exposure in
multiple interpersonal domains: cumulative and differential effects. J Adolesc Health,
47(2), 198-205. doi: 10.1016/j.jadohealth.2010.01.020
McEwen, B. S. (1998). Stress, adaptation, and disease. Allostasis and allostatic load. Ann N Y
Acad Sci, 840, 33-44.
McEwen, B. S. (2000). The neurobiology of stress: from serendipity to clinical relevance. Brain
Res, 886(1-2), 172-189.
McEwen, B. S., & Wingfield, J. C. (2003). The concept of allostasis in biology and biomedicine.
Horm Behav, 43(1), 2-15.
Meinlschmidt, G., & Heim, C. (2005). Decreased cortisol awakening response after early loss
experience. Psychoneuroendocrinology, 30(6), 568-576. doi:
10.1016/j.psyneuen.2005.01.006
Mikolajczak, M., Quoidbach, J., Vanootighem, V., Lambert, F., Lahaye, M., Fillee, C., & de
Timary, P. (2010). Cortisol awakening response (CAR)'s flexibility leads to larger and
more consistent associations with psychological factors than CAR magnitude.
Psychoneuroendocrinology, 35(5), 752-757. doi: 10.1016/j.psyneuen.2009.11.003
ADOL STRESS AND CORTISOL AWAKENING

41
Miller, G. E., Chen, E., & Zhou, E. S. (2007). If it goes up, must it come down? Chronic stress
and the hypothalamic-pituitary-adrenocortical axis in humans. Psychol Bull, 133(1), 25-
45. doi: 10.1037/0033-2909.133.1.25
Murray-Close, D. (2012). Psychophysiology of Adolescent Peer Relations II: Recent Advances
and Future Directions. Journal of Research on Adolescence, 23(2), 260-273.
Muthén, L. K., & Muthén, B. O. (2010). Mplus User's Guide (Sixth Edition ed.). Los Angeles,
CA: Muthén & Muthén.
Nachmias, M., Gunnar, M. R., Mangelsdorf, S., Parritz, R. H., & Buss, K. (1996). Behavioral
inhibition and stress reactivity: the moderating role of attachment security. Child Dev,
67(2), 508-522.
Nicolson, N. A. (2004). Childhood parental loss and cortisol levels in adult men.
Psychoneuroendocrinology, 29(8), 1012-1018. doi: 10.1016/j.psyneuen.2003.09.005
Oskis, A., Loveday, C., Hucklebridge, F., Thorn, L., & Clow, A. (2011). Anxious attachment
style and salivary cortisol dysregulation in healthy female children and adolescents. J
Child Psychol Psychiatry, 52(2), 111-118. doi: 10.1111/j.1469-7610.2010.02296.x
Peters, E., Riksen-Walraven, J. M., Cillessen, A. H., & de Weerth, C. (2011). Peer rejection and
HPA activity in middle childhood: friendship makes a difference. Child Dev, 82(6),
1906-1920. doi: 10.1111/j.1467-8624.2011.01647.x
Popma, A., Doreleijers, T. A., Jansen, L. M., Van Goozen, S. H., Van Engeland, H., &
Vermeiren, R. (2007). The diurnal cortisol cycle in delinquent male adolescents and
normal controls. Neuropsychopharmacology, 32(7), 1622-1628. doi:
10.1038/sj.npp.1301289
ADOL STRESS AND CORTISOL AWAKENING

42
Porges, S. W., Doussard-Roosevelt, J. A., & Maiti, A. K. (1994). Vagal tone and the
physiological regulation of emotion. Monogr Soc Res Child Dev, 59(2-3), 167-186.
Quevedo, K., Johnson, A., Loman, M., Lafavor, T., & Gunnar, M. R. (2012). The Confluence of
Adverse Early Experience and Puberty on the Cortisol Awakening Response. Int J Behav
Dev, 36(1), 19-28. doi: 10.1177/0165025411406860
Repetti, R. L., Robles, T. F., & Reynolds, B. (2011). Allostatic processes in the family. Dev
Psychopathol, 23(3), 921-938. doi: 10.1017/S095457941100040X
Roisman, G. I., Susman, E., Barnett-Walker, K., Booth-LaForce, C., Owen, M. T., Belsky, J., . . .
Network, Nichd Early Child Care Research. (2009). Early family and child-care
antecedents of awakening cortisol levels in adolescence. Child Dev, 80(3), 907-920. doi:
10.1111/j.1467-8624.2009.01305.x
Rosenthal, N. L., & Kobak, R. (2010). Assessing Adolescents' Attachment Hierarchies:
Differences Across Developmental Periods and Associations With Individual Adaptation.
J Res Adolesc, 20(3), 678-706. doi: 10.1111/j.1532-7795.2010.00655.x
Sapolsky, R. M. (1994). The physiological relevance of glucocorticoid endangerment of the
hippocampus. Ann N Y Acad Sci, 746, 294-304; discussion 304-297.
Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorticoids influence
stress responses? Integrating permissive, suppressive, stimulatory, and preparative
actions. Endocr Rev, 21(1), 55-89. doi: 10.1210/edrv.21.1.0389
Saxbe, D. E., Margolin, G., Spies Shapiro, L. A., & Baucom, B. R. (2012). Does dampened
physiological reactivity protect youth in aggressive family environments? Child Dev,
83(3), 821-830. doi: 10.1111/j.1467-8624.2012.01752.x
ADOL STRESS AND CORTISOL AWAKENING

43
Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art.
Psychological Methods, 7, 147-177.
Schore, A. N. (2001). Effects of a secure attachment relationship on right brain development,
affect regulation, and infant mental health. Infant Mental Health Journal, 22, 7-66.
Seeman, T. E., Singer, B. H., Rowe, J. W., Horwitz, R. I., & McEwen, B. S. (1997). Price of
adaptation--allostatic load and its health consequences. MacArthur studies of successful
aging. Arch Intern Med, 157(19), 2259-2268.
Seltzer, L. J., Ziegler, T. E., & Pollak, S. D. (2010). Social vocalizations can release oxytocin in
humans. Proc Biol Sci, 277(1694), 2661-2666. doi: 10.1098/rspb.2010.0567
Shea, A. K., Streiner, D. L., Fleming, A., Kamath, M. V., Broad, K., & Steiner, M. (2007). The
effect of depression, anxiety and early life trauma on the cortisol awakening response
during pregnancy: preliminary results. Psychoneuroendocrinology, 32(8-10), 1013-1020.
doi: 10.1016/j.psyneuen.2007.07.006
Shields, A., & Cicchetti, D. (1997). Emotion regulation among school-age children: the
development and validation of a new criterion Q-sort scale. Dev Psychol, 33(6), 906-916.
Shumaker, S. A., & Hill, D. R. (1991). Gender differences in social support and physical health.
Health Psychol, 10(2), 102-111.
Simons, R. L., Conger, R. D., & Whitbeck, L. B. (1988). A multistage social learning model of
the influences of family and peers upon adolescent substance use. Journal of Drug Issues,
18, 293-316.
Stadler, T., Evans, P., Hucklebridge, F., & Clow, A. (2011). Associations between the cortisol
awakening response and heart rate variability. Psychoneuroendocrinology, 36, 454-462.
ADOL STRESS AND CORTISOL AWAKENING

44
Stanton, M. E., & Levine, S. (1985). Brief separation elevates cortisol in mother and infant
squirrel monkeys. Physiol Behav, 34(6), 1007-1008.
Stein, H., Jacobs, N. J., Ferguson, K. S., Allen, J. G., & Fonagy, P. (1998). What do adult
attachment scales measure? Bull Menninger Clin, 62(1), 33-82.
Stroud, L. R., Salovey, P., & Epel, E. S. (2002). Sex differences in stress responses: social
rejection versus achievement stress. Biol Psychiatry, 52(4), 318-327.
Thoits, P. A. (2011). Mechanisms linking social ties and support to physical and mental health. J
Health Soc Behav, 52(2), 145-161. doi: 10.1177/0022146510395592
Tops, M., Riese, H., Oldehinkel, A. J., Rijsdijk, F. V., & Ormel, J. (2008). Rejection sensitivity
relates to hypocortisolism and depressed mood state in young women.
Psychoneuroendocrinology, 33(5), 551-559. doi: 10.1016/j.psyneuen.2008.01.011
van der Vegt, E. J., van der Ende, J., Kirschbaum, C., Verhulst, F. C., & Tiemeier, H. (2009).
Early neglect and abuse predict diurnal cortisol patterns in adults A study of international
adoptees. Psychoneuroendocrinology, 34(5), 660-669. doi:
10.1016/j.psyneuen.2008.11.004
Waters, E., Weinfield, N. S., & Hamilton, C. E. (2000). The stability of attachment security from
infancy to adolescence and early adulthood: general discussion. Child Dev, 71(3), 703-
706.
Wust, S., Federenko, I., Hellhammer, D. H., & Kirschbaum, C. (2000). Genetic factors,
perceived chronic stress, and the free cortisol response to awakening.
Psychoneuroendocrinology, 25(7), 707-720.
ADOL STRESS AND CORTISOL AWAKENING

45
Zeiders, K. H., Doane, L. D., & Roosa, M. W. (2012). Perceived discrimination and diurnal
cortisol: examining relations among Mexican American adolescents. Horm Behav, 61(4),
541-548. doi: 10.1016/j.yhbeh.2012.01.018

ADOL STRESS AND CORTISOL AWAKENING

46
Table 1

Descriptive Statistics for Cortisol Measures, Stress and Moderator Variables, and Covariates
a
These measures are means across three days.

Total Sample
Mean (SD)
Males
Mean (SD)
Females
Mean (SD)
Cortisol Awakening Response .16 (.27) .12 (.25) .21 (.29)
Cortisol 1: awakening
a
.37 (.23) .38 (.24) .36 (.21)
Cortisol 2: awake+20min
a
.50 (.30) .49 (.31) .51 (.28)
Cortisol 3: awake+40min
a
.46 (.29) .42 (.27) .50 (.31)
Past Stress 10.00 (8.14) 8.47 (6.67) 11.63 (9.25)
Current Stress 12.53 (9.76) 11.98 (8.81) 13.13 (10.79)
Emotion Regulation 2.27 (.36) 2.31 (.38) 2.23 (.34)
Attachment to Parents 2.50 (.57) 2.51 (.55) 2.49 (.61)
Attachment to Peers 2.96 (.54) 2.91 (.52) 3.02 (.57)
Age 18.05 (1.09) 18.13 (1.12) 17.96 (1.05)
Family Income 91,828 (58,884) 92,343 (56,839) 91,225 (61,898)
Medication .22 (.42) .17 (.38) .28 (.46)
Time of Awakening
a

7.39 (.97)
7.42 (1.05) 7.35 (.86)
Hours of Sleep
a

7.25 (1.59)
7.30 (1.62) 7.20 (1.57)
Cotinine
a

4.76 (21.61)
7.81 (28.62) 1.25 (6.50)
Ate/Drank/Exercised/Sores
a

.29 (.46)
.34 (.47) .24 (43)
ADOL STRESS AND CORTISOL AWAKENING

47

Table 2

Intercorrelations Among All Model Variables

Note. CAR = cortisol awakening response, Cort = cortisol sample, P Stress = past stress, C Stress = current stress, EmotReg = emotion regulation, Att Par =
attachment to parents, Att Peer = attachment to peers, Meds = medications, Wake = time of awakening, Hrs Slp = hours of sleep, Cotin = cotinine, and Ate =
ate/drank/exercised/sores.
†p < .10, *p < .05, ** p < .01, *** p < .001
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1. CAR -
2. Cort 1 -.23* -
3. Cort 2 .39** .74*** -
4. Cort 3 .61** .53*** .80*** -
5. P Stress -.01 .10 .14 .03 -
6. C Stress .21* .05 .22* .16 .29** -
7. EmotReg .12 -.02 .02 .00 -.08 -.19 † -
8. Att Par -.07 .03 -.02 -.01 -.10
-
.35***
.35*** -
9. Att Peer .11 .00 .13 .04 .19 † .15 .29** .27** -
10. Age -.20* -.08 -.19 † -.22* -.05 -.02 .08 .23* .02 -
11. Gender -.25* .09 -.06 -.17 † -.20 † -.06 .11 .02 -.10 .08 -
12. Income .08 .10 .15 .13 .08 -.17 .20 † .07 .18 † -.14 .01 -
13. Meds .13 .00 .13 .12 .21* .14 -.01 -.06 .13 -.12 -.13 -.09 -
14. Wake -.19 † .01 -.03 -.18 † .10 .12 .09 -.07 .07 .30** .05 -.06 .14 -
15. Hrs Slp .01 .00 .05 .01 .09 -.02 .10 .06 -.04 .06 .01 .15 .16 .25* -
16. Cotin .00 -.03 -.06 -.03 -.06 -.07 .07 -.09 .01 -.03 .15 .10 -.12 .15 -.35** -
17. Ate .04 .07 .14 .03 -.02 .19 † -.09 -.03 .10 .08 .13 .11 .04 .13 .01 -.08 -
ADOL STRESS AND CORTISOL AWAKENING

48
Table 3.

Intercorrelations Among All Model Variables, Presented for Males and Females Separately

Note: Intercorrelations for males presented on top half. CAR = cortisol awakening response, Cort = cortisol sample, P Stress = past stress, C Stress = current
stress, EmotReg = emotion regulation, Att Par = attachment to parents, Att Peer = attachment to peers, Meds = medications, Wake = time of awakening, Hrs Slp
= hours of sleep, Cotin = cotinine, and Ate = ate/drank/exercised/sores.
†p < .10, *p < .05, ** p < .01, *** p < .001
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1. CAR - -.35* .08 .30* -.38** .10 .32* .06 -.01 -.02 -.12 .06 -.19 .10 .07 -.03
2. Cort 1 -.04 - .85** .67** .25 † .07 -.03 .10 .15 -.12 .20 .03 .06 -.04 -.05 .06
3. Cort 2 .72** .57** - .80** .07 .13 .04 .12 .14 -.10 .06 .17 .01 .01 -.06 .07
4. Cort 3 .84** .44** .81** - -.11 .07 .05 .14 .04 -.14 .04 .06 -.13 .06 .03 -.02
5. P Stress .12 -.01 .18 .06 - .19 .00 .10 .16 .11 .09 .07 .14 -.03 -.08 -.01
6. C Stress .27 † .04 .311* .23 .33* - -.20 -.18 .16 .07 -.12 .12 .06 -.05 -.11 .19
7. EmotReg -.01 -.02 .03 -.01 -.11 -.18 - .22 .18 -.01 .14 .05 .10 .00 .10 -.15
8. Att Par -.17 -.08 -.17 -.15 -.24 -.49** .50** - .22 .29* .04 -.15 -.10 .06 -.16 .16
9. Att Peer .16 -.22 .11 .00 .17 .14 .44** .33* - .17 .15 .08 .16 -.14 .04 .25 †
1. Age -.36* -.04 -.32* -.29* -.16 -.10 .20 .17 -.11 - -.10 -.19 .22 .02 -.05 .13
11. Income .24 -.02 .24 .22 .09 -.21 .26 .11 .22 -.19 - -.24 .08 .22 .20 .20
12. Meds .13 -.01 .09 .13 .26 † .14 -.04 .02 .15 -.04 .05 - .28* .35* -.12 -.06
13. Wake -.19 -.13 -.09 -.24 .09 .20 .06 -.04 -.06 .42** -.23 .00 - .18 .17 .17
14. Hrs Slp -.06 .08 .10 -.05 .21 .01 .25 .06 .10 .11 .08 -.03 .39* - -.46** -.06
15. Cotin -.05 -.12 -.06 -.14 .07 .00 -.09 .13 .00 -.03 -.23 -.13 .10 -.01 - .17
16. Ate .18 .05 .24 .13 .02 .21 -.06 -.24 -.04 -.02 .00 .17 .07 .11 .13 -
ADOL STRESS AND CORTISOL AWAKENING

49
Table 4.

Standardized Multi-level Model of Current Stress Regressed on CAR

Coefficient Standard
Error
Current Stress .31** .12
Level 2 Covariates
Past Stress -.228 .17
Age -.25† .13
Gender -.35** .12
Family Income .16 .13
Medications .12 .16
Level 1 Covariates
Time of Awakening -.03 .03
Hours of Sleep -.13† .07
Cotinine .07† .04
Ate/Drank/Exercised/Sores -.01 .84
†p < .10, *p < .05, ** p < .01, *** p < .001

ADOL STRESS AND CORTISOL AWAKENING

50
Table 5.

Standardized Multi-level model of Interaction Between Past Stress and Gender regressed on CAR and Past Stress Regressed on CAR
for Males and Females Separately

Entire Sample Interaction Males Only Females Only
Coefficient Standard
Error
Coefficient Standard
Error
Coefficient Standard
Error
Past Stress x Gender -.67*** .08 - - - -
Past Stress .49** .14 -.73*** .19 -.05 .20
Gender .05 .13 - - - -
Level 2 Covariates
Current Stress .19† .11 .27† .16 .32† .17
Age
-.14
.09 -.04 .16 -.41* .18
Family Income .11 .10 -.14 .19 .34* .16
Medications .10 .10 .18 .29 .07 .19
Level 1 Covariates
Time of Awakening -.03 .03 -.08* .03 .07 .05
Hours of Sleep -.12† .07 .03 .06 -.28** .10
Cotinine .08* .04 .09 .06 .05** .02
Ate/Drank/Exercised/Sores -.01 .05 .03 .08 -.03 .06
†p < .10, *p < .05, ** p < .01, *** p < .001

ADOL STRESS AND CORTISOL AWAKENING

51
Table 6.

Standardized Multi-level Model of Three-way Interactions Between Past Stress, Gender, and each moderator regressed CAR

Emotion Regulation Attachment to Parents Attachment to Peers
Coefficient Standard
Error
Coefficient Standard
Error
Coefficient Standard
Error
Past Stress x Gender x
Moderator
-.615† .33 -.61** .22 -.05 .28
Past Stress x Gender .23 .26 .31 .20 -.19 .38
Past Stress x Moderator .32 .36 .31 .25 .70*** .11
Gender x Moderator .40 .33 .42* .18 -.06 .28
Past Stress -.08 .45 -.15 .30 -.55** .21
Gender -.38 .29 -.36* .15 .00 .26
Moderator .00 .30 -.13 .20 -.04 .15
Level 2 Covariates
Current Stress .16 .11 .11 .07 .16† .09
Age
-.13
.12 -.10 .08 -.11 .08
Family Income .06 .07 .06 .06 .06 .07
Medications .07 .08 .07 .06 .09 .08
Level 1 Covariates
Time of Awakening -.03 .03 -.03 .03 -.03 .03
Hours of Sleep -.12† .07 -.13 † .07 -.10 .06
Cotinine .07* .04 .08* .04 .07* .04
Ate/Drank/Exercised/Sores -.01 .05 -.01 .05 -.01 .05
†p < .10, *p < .05, ** p < .01, *** p < .001

ADOL STRESS AND CORTISOL AWAKENING

52
Table 7.

Standardized Multi-level Model of Interaction Between Past Stress and Emotion Regulation Regressed on CAR, Presented Separately
for Total Sample, Males, and Females

Total Sample Males Only Females Only
Coefficient Standard
Error
Coefficient Standard
Error
Coefficient Standard
Error
Past Stress x Emot Reg -.61*** .13 -.78*** .06 -.45 .45
Past Stress .47** .17 .47** .14 .40 .48
Emotion Regulation .25** .10 .34** .13 .18 .15
Level 2 Covariates
Current Stress .21* .11 .14† .08 .26 .24
Age - .18† .10 0.01 .08 -.36* .17
Gender -.22† .11 - - - -
Family Income .08 .09 -.08 .07 .25 .21
Medications .09 .09 .12 .10 .04 .15
Level 1 Covariates
Time of Awakening -.03 .03 -.08* .04 .04 .05
Hours of Sleep -.13† .07 .04 .05 -.29** .11
Cotinine .07† .04 .09 .06 .05** .02
Ate/Drank/Exercised/Sores -.01 .05 .02 .09 -.03 .06
†p < .10, *p < .05, ** p < .01, *** p < .001

ADOL STRESS AND CORTISOL AWAKENING

53
Table 8.

Standardized Multi-Level Model of Interaction Between Past stress and Attachment to Parents Regressed on CAR, Presented
Separately for Total Sample, Males, and Females

Total Sample Males Only Females Only
Coefficient Standard
Error
Coefficient Standard
Error
Coefficient Standard
Error
Past Stress x Att Parents -.65*** .08 -.80*** .05 .02 1.09
Past Stress .51*** .12 .49*** .10 -.07 1.17
Attachment to Parents .24** .08 .26** .08 .05 .45
Level 2 Covariates
Current Stress .15 .10 .11 .07 .35† .20
Age -.16* .08 -.03 .07 -.41* .21
Gender -.16† .10 - - - -
Family Income .08 .09 -.05 .06 .33* .16
Medications .09 .08 .12 .09 .07 .19
Level 1 Covariates
Time of Awakening -.03 .03 -.07† .04 .07 .05
Hours of Sleep -.14* .07 -.01 .05 -.28** .11
Cotinine .07* .04 .08 .06 .04** .02
Ate/Drank/Exercised/Sores -.01 .05 .03 .08 -.03 .06
†p < .10, *p < .05, ** p < .01, *** p < .001

ADOL STRESS AND CORTISOL AWAKENING

54
Table 9.

Standardized Multi-level Model of Interaction Between Past Stress and Attachment to Peers
Regressed on CAR

Coefficient Standard
Error
Past Stress x Att Peers .62*** .08
Past Stress -.68*** .05
Attachment to Peers -.07 .06
Level 2 Covariates
Current Stress .14* .07
Age -.11 .07
Gender -.13 .08
Family Income .05 .06
Medications .09 .07
Level 1 Covariates
Time of Awakening -.03 .03
Hours of Sleep -.11† .07
Cotinine .07† .04
Ate/Drank/Exercised/Sores -.01 .05
†p < .10, *p < .05, ** p < .01, *** p < .001

ADOL STRESS AND CORTISOL AWAKENING

55

Figure 1. Diurnal cortisol patterns of adolescents high (above the median) or low (below the
median) in current stress.
0
0.1
0.2
0.3
0.4
0.5
0.6
C ort iso l
C onc e ntr a tion
( μg/d L )
High Current Stress
Low Current Stress
ADOL STRESS AND CORTISOL AWAKENING

56

Figure 2. Diurnal cortisol patterns of adolescents high (above the median) or low (below the median) in past stress, presented for
males and females separately.
0
0.1
0.2
0.3
0.4
0.5
0.6
Females Males
C o r tis o l Co n ce n tr atio n ( μ g /d L )
High Past Stress
Low Past Stress
ADOL STRESS AND CORTISOL AWAKENING

57

Figure 3. Interaction between past stress and emotion regulation for male adolescents. High and
low values are one standard deviation above or below the mean.

-1.5
-1
-0.5
0
0.5
1
1.5
2
Low Past Stress High Past Stress
Corisol Awakening Response
Low Emotion
Regulation
High Emotion
Regulation
ADOL STRESS AND CORTISOL AWAKENING

58

Figure 4. Interaction between past stress and attachment to parents for male adolescents. High
and low values are one standard deviation above or below the mean.

-1.5
-1
-0.5
0
0.5
1
1.5
2
Low Past Stress High Past Stress
Cortisol Awakening Response
Low Attachment to
Parents
High Attachment
to Parents
ADOL STRESS AND CORTISOL AWAKENING

59

Figure 5. Interaction between past stress and attachment to peers for entire sample. High and low
values are one standard deviation above or below the mean.

-1.5
-1
-0.5
0
0.5
1
1.5
2
Low Past Stress High Past Stress
Cortisol Awakening Response
Low Attachment
to Peers
High Attachment
to Peers

Abstract (if available)

Abstract The cortisol awakening response (CAR) is an index of hypothalamic‐pituitary‐adrenal (HPA) activity that is sensitive to the effects of stress among adults (Chida & Steptoe, 2009). However, less is known about the relationship between stress and CAR in adolescence. The present study examines whether adolescents’ past and current stress have different associations with CAR and tests whether emotion regulation, attachment to parents, or attachment to peers moderates these relationships. Ninety‐nine adolescents were recruited from a longitudinal study to provide home saliva samples for cortisol analyses. In the laboratory, adolescents additionally reported on stressful life experiences, emotion regulation, and attachment to parents and peers. Multi‐level models reveal that current stress is associated with increased CAR. In contrast, past stress is associated with decreased CAR for males only. No moderation was found for the association between current stress and CAR. In contrast, several moderators were identified for the relationship between past stress and CAR: attachment to peers was protective for all adolescents, while emotion regulation and attachment to parents were protective for males only. Additionally, adolescents who are low in attachment to peers appear to represent a distinct risk group from adolescents who are low in attachment to parents or low in emotion regulation. Results highlight the distinct roles peers play in adolescents’ changing social worlds and underscore the gender differences in bio‐psycho‐social development across middle childhood and adolescence.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Effects of childhood adversity on physiology and health in emerging adulthood

PDF

Couples’ neuroendocrine activity in response to family conflict discussions: the role of self-reported anger and previous marital aggression

PDF

Using observed peer discussions to understand adolescent depressive symptoms and interpersonal interactions

PDF

A multiple systems approach to examining physiological stress and its association with internalizing disorders in adolescence

PDF

The acute relationship between affective states and physiological stress response, and the moderating role of moderate-to-vigorous physical activity

PDF

The old ball and linkage: couples’ prenatal conflict behavior, cortisol linkage, and postpartum depression risk

PDF

Couple conflict during pregnancy: Do early family adversity and oxytocin play a role?

PDF

Biological and behavioral correlates of emotional flexibility and associations with exposure to family aggression

PDF

The role of emotions: gender differences in different displayed emotions and witness credibility

PDF

Family aggression, prosocial friends, and the risk of dating and friend victimization in late adolescence

PDF

Dynamics of victimization, aggression, and popularity in adolescence

PDF

Community violence exposure and children's subsequent rejection within the school peer group: the mediating roles of emotion dysregulation and depressive symptoms

PDF

Family aggression exposure and community violence exposure associated with brain volume in late adolescence: a comparison of automated versus manual segmentation

PDF

You always say that: physical aggression perpetration, linguistic behavior, and conflict intensity in young adult dating couples

PDF

Friendship as a protective factor in adolescence

PDF

Mediators and moderators in the link between maternal psychological control and peer victimization for Hong Kong Chinese boys

PDF

The physiology of compassion in couples’ discussions about loss

PDF

Emotion regulation as a mechanism linking parents’ marital aggression to adolescent behavioral problems: a longitudinal analysis

PDF

Family conflict, negative mood, and adolescents' daily problems in school

PDF

Predictors and outcomes across the transition to fatherhood

Asset Metadata

Creator Miller, Kelly Frances (author)

Core Title Adolescent life stress and the cortisol awakening response: the moderating roles of emotion regulation, attachment, and gender

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 07/01/2014

Defense Date 05/28/2014

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag adolescence,cortisol awakening response,HPA axis,OAI-PMH Harvest,Stress

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Margolin, Gayla (committee chair), Farver, Jo Ann M. (committee member), Saxbe, Darby E. (committee member)

Creator Email kellyfmiller@gmail.com,kfmiller@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-428324

Unique identifier UC11286721

Identifier etd-MillerKell-2601.pdf (filename),usctheses-c3-428324 (legacy record id)

Legacy Identifier etd-MillerKell-2601.pdf

Dmrecord 428324

Document Type Thesis

Format application/pdf (imt)

Rights Miller, Kelly Frances

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA