This article was downloaded by: [RMIT University]

On: 25 March 2013, At: 16:19

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Psychology & Health

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/gpsh20

Perceived Emotional Intelligence, Stress

Reactivity, and Symptom Reports:
Further Explorations Using the Trait
Meta-Mood Scale
a b c
Peter Salovey , Laura R. Stroud , Alison Woolery & Elissa S.
d
Epel
a
Department of Psychology, Yale University, P.O. Box 208205, New
Haven, CT, 06520-8205, USA
b
Department of Psychiatry and Human Behavior, Brown University,
164 Summit Avenue Providence, RI, 02906, USA
c
Department of Psychology, University of California, P.O. Box
951563, Los Angeles, CA, 90095, USA
d
Department of Health Psychology, University of California, 3333
California Street, San Francisco, CA, 94143, USA
Version of record first published: 27 Oct 2010.

To cite this article: Peter Salovey , Laura R. Stroud , Alison Woolery & Elissa S. Epel (2002):
Perceived Emotional Intelligence, Stress Reactivity, and Symptom Reports: Further Explorations Using
the Trait Meta-Mood Scale, Psychology & Health, 17:5, 611-627

To link to this article: http://dx.doi.org/10.1080/08870440290025812

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
 demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
Downloaded by [RMIT University] at 16:19 25 March 2013
 Psychology and Health, 2002, Vol. 17, No. 5, pp. 611–627

PERCEIVED EMOTIONAL INTELLIGENCE,

STRESS REACTIVITY, AND SYMPTOM REPORTS:
FURTHER EXPLORATIONS USING THE TRAIT
META-MOOD SCALE*
Downloaded by [RMIT University] at 16:19 25 March 2013

PETER SALOVEYa,y, LAURA R. STROUDb, ALISON WOOLERYc

and ELISSA S. EPELd
a
Department of Psychology, Yale University, P.O. Box 208205, New Haven, CT 06520-8205, USA;
b
Department of Psychiatry and Human Behavior, Brown University, 164 Summit Avenue
Providence, RI 02906, USA; cDepartment of Psychology, University of California, Los Angeles,
P.O. Box 951563, Los Angeles, CA 90095, USA; dDepartment of Health Psychology, University of
California, San Francisco, 3333 California Street, San Francisco, CA 94143, USA

We examined the relationship between perceived emotional intelligence (PEI), measured by the Trait Meta-
Mood Scale (TMMS), and psychophysiological measures of adaptive coping. The TMMS assesses perceived
ability to (a) attend to moods (Attention), (b) discriminate clearly among moods (Clarity), and (c) regulate
moods (Repair). Study 1 showed significant positive associations between PEI and psychological and inter-
personal functioning. In Study 2, skill at mood Repair was associated with less passive coping and perceptions
of repeated laboratory stressors as less threatening; Clarity was related to greater increases in negative mood,
but lower cortisol release during repeated stress. In Study 3, Repair was associated with active coping and
lower levels of rumination; Attention was associated with lowered cortisol and blood pressure responses to
acute laboratory challenges. These findings suggest that psychophysiological responses to stress may be
one potential mechanism underlying the relationship between emotional functioning and health.

Keywords: Emotional intelligence; Health; Stress reactivity; Coping

Individuals differ in the skill with which they can identify their feelings and the feelings
of others, regulate these feelings, and use the information provided by their feelings to
motivate adaptive behavior. These competencies have been organized into a framework
called emotional intelligence (e.g., Salovey and Mayer, 1990; Mayer and Salovey, 1993,
1997; Salovey et al., 2000; Salovey et al., 2001). An important aspect of emotional
intelligence is the ability to reflect upon and manage one’s emotions. Mayer and
Gaschke (1988) demonstrated that individuals continually reflect upon their feelings
by monitoring, evaluating, and regulating them. They termed this process the

*Portions of this paper were presented at the Second International Conference on the (Non)Expression of
Emotions in Health and Disease, Department of Psychology, Tilburg University, The Netherlands, June 1999;
at the annual meeting of the American Psychological Association, Boston, MA, August 1999; and at the
annual meeting of the Eastern Psychological Association, Providence, RI, April 1999.
y
Corresponding author. E-mail: peter.salovey@yale.edu

ISSN 0887-0446 print: ISSN 1476-8321 online ß 2002 Taylor & Francis Ltd
DOI: 10.1080/08870440290025812
 612 P. SALOVEY et al.

meta-mood experience and developed what is now called the State Meta-Mood Scale to
measure individuals’ moment-by-moment changes in reflections about ongoing moods.
In order to measure more stable individual differences in the qualities of the reflective
mood experience, Salovey and colleagues (Salovey et al., 1995) developed the Trait
Meta-Mood Scale (TMMS). The TMMS is composed of three subscales: (a)
Attention – perceived ability to attend to moods and emotions, (b) Clarity – perceived
ability to discriminate clearly among feelings, and (c) Repair – perceived ability to reg-
ulate moods. Past research from our laboratory has shown links between perceived
emotional intelligence (PEI) as measured by the TMMS, and both psychological
responses to stress and physical health. In one study, individuals who were high in
Clarity showed greater rebound from induced negative mood and greater decline in
ruminative thought following an experimental stressor compared to individuals who
were low in Clarity (Salovey et al., 1995). In another, individuals who were higher in
Downloaded by [RMIT University] at 16:19 25 March 2013

Attention reported higher levels of physical symptoms, but individuals who perceived
themselves as skilled at mood repair reported fewer illnesses (Goldman et al., 1996).
In general, it appears that greater PEI, as assessed by the TMMS subscales, is associ-
ated with fewer physical symptoms and more adaptive reactions to stressors. It may be
that individuals who can clearly perceive their feelings and believe they can repair nega-
tive mood states turn their attentional resources toward coping and minimizing the
impact of stressful events. On the other hand, individuals low in Attention, Clarity,
and Repair may engage in extended rumination in order to understand how they
feel. Rumination and the absence of attempts to attend to, clarify, and repair mood
may then lead to prolonged physiological arousal and negative health outcomes
(Nolen-Hoeksema et al., 1994; Gross, 1998).
Although we have demonstrated relations between PEI and physical symptoms as
well as between PEI and psychological responses to stress, little is known about psycho-
physiological mechanisms underlying these associations. One construct that has been
linked to both emotion regulation and health, and thus represents a potential under-
lying mechanism, is physiological responses to stress. Physiological responses to
stress are believed to involve two systems: the hypothalamic pituitary adrenocortical
(HPA) axis, and the sympathetic axis. Activation of the HPA axis results in the secre-
tion of cortisol from the adrenal cortex into circulation. In the short term, cortisol both
increases arousal and defends against the body’s own stress reaction. However, chronic
over-exposure to the HPA stress response can itself damage many regulatory systems
(McEwen, 1998; Sapolsky et al., 2000). Activation of the sympathetic axis typically
leads to symptoms of arousal including cardiovascular changes (e.g., increased blood
pressure and heart rate). Although there are conflicting views of what constitutes adap-
tive HPA and sympathetic responses to stress, large increases in physiological responses
to acute stressors as well as the lack of ability to habituate to chronic stressors may over
time damage organs and lead to disease (Manuck and Krantz, 1986; Dientsbier, 1989;
Ratliff-Crain and Vingerhoets, 1996; McEwen, 1998). Given evidence for the relation-
ship between PEI and both health and psychological responses to stress (Salovey et al.,
1995; Goldman et al., 1996), and the importance of emotion regulation in determining
physiological stress reactivity (e.g., Nachmias et al., 1996), we predicted that greater
PEI will be related to adaptive physiological responses to stress.
Another construct related to both emotion regulation and health is coping. Coping
has been conceptualized as an individual’s pattern of response to external negative
events (Carver et al., 1989). Active coping is arguably more adaptive, and refers to
 TRAIT META-MOOD SCALE 613

active steps to change a stressful situation or to ameliorate its effects including both
emotion and problem-focused strategies. Passive coping, arguably less adaptive,
refers to giving up, avoiding, or inhibiting an active response. Given links between
coping and psychological and physical functioning (e.g., Weinberger, 1990), we also
examined the relation between PEI and state and trait coping. We believe that
Attention, Clarity, and Repair are essential ingredients for active coping, which
involves both engaging with emotions, then managing them in order to change the
situation. Although emotion-focused strategies have previously been linked with pas-
sive coping, PEI focuses on the tendency to engage with emotions, rather than avoid
them. Thus, we hypothesized that greater PEI will be associated with greater active
coping and less passive coping.
In the present series of studies, we re-examined relationships between PEI and
psychological and physical functioning, extending these findings to interpersonal
Downloaded by [RMIT University] at 16:19 25 March 2013

relationships (Study 1). We then investigated the relationships between PEI and psycho-
physiological measures of coping with stressors in an effort to elucidate potential
mechanisms underlying the relation between PEI and health (Studies 2 and 3). The
central thesis of this research is that attention to moods, clarity in perceiving mood,
and confidence in one’s ability to repair negative mood are critical for adaptive psycho-
physiological coping and subsequent well-being.

STUDY 1

In Study 1, we re-examined relationships between PEI and measures of psychological

and physical functioning. We expected that high Attention, Clarity, and Repair
would be positively associated with self-esteem and negatively associated with physical
symptom reporting, and depression. We also examined the relationship between
PEI and three measures of interpersonal functioning, empathy, social anxiety, and
general interpersonal satisfaction. We expected that greater PEI would be related
to greater empathy, lower social anxiety, and greater satisfaction in interpersonal
relationships (e.g., Mehrabian and Epstein, 1972; Gottman and Levenson, 1986;
Cooper et al., 1998).

Method
Participants
Participants were 104 undergraduates (29 males, 71 females, 4 unknown), ages 16–23,
who were recruited from an Introductory Psychology course, and received course
credit for their participation.

Procedure
Participants completed a packet of questionnaires including the TMMS as well as ques-
tionnaires measuring empathy, social anxiety, self-esteem, depression, physical symp-
toms, and interpersonal satisfaction.
 614 P. SALOVEY et al.

Measures
Trait Meta-Mood Scale The TMMS is a 48-item, self-report measure designed
to assess individuals’ beliefs about attending to moods (Attention), the clarity of
their own experiences of mood (Clarity), and their efforts to repair mood states
(Repair) (Salovey et al., 1995). The Attention scale includes items such as, ‘‘I pay a
lot of attention to how I feel,’’ and ‘‘I don’t think it’s worth paying attention to your
emotions or moods.’’ The Clarity scale includes items such as ‘‘I am usually very
clear about my feelings’’ and ‘‘I can’t make sense out of my feelings.’’ The third sub-
scale, Repair, reflects individuals’ efforts to repair negative mood by maintaining a
positive outlook. This subscale is characterized by items such as: ‘‘When I become
upset, I remind myself of all the pleasures in life’’ and ‘‘I try to have good thoughts
no matter how bad I feel.’’ The TMMS has been shown to have adequate internal con-
sistency and good convergent and discriminant validity (Salovey et al., 1995). Cronbach
Downloaded by [RMIT University] at 16:19 25 March 2013

alphas for this study are reported in Table I.

Empathy Empathy, or capacity for emotional sensitivity, vicarious arousal, and the
identification of affect in others, was measured using the 33-item Mehrabian and
Epstein (1972) measure of emotional empathy. Cronbach’s alpha in this study was 0.74.
Social Anxiety Social anxiety, or discomfort in the presence of others, was measured
with the Social Anxiety Subscale of the Self-Consciousness Scale (Fenigstein et al.,
1975). This 6-item scale included items such as ‘‘It takes me time to overcome my shy-
ness in new situations,’’ and ‘‘I feel anxious when I speak in front of a group’’ measured
along 5-point scales. Cronbach’s alpha was 0.72.
Self-Esteem Self-esteem, defined as global feelings of self-worth and self-acceptance,
was measured using the 10-item Rosenberg (1965) Self-Esteem Scale with standard
four-point scoring. Cronbach’s alpha was 0.87.
Depression Depression was measured with the Center for Epidemiologic Studies
Depression Scale (Radloff, 1977). This scale includes 20 items assessed along 4-point
scales measuring various components of depression including: depressed mood, feelings

TABLE I Means (SD’s), Cronbach’s alphas, and correlations among TMMS subscales
across Studies 1–3

Mean (SD) Alpha Attention Clarity Repair

Study 1 (n ¼ 104)
Attention 4.10 (0.52) 0.82 1.00
Clarity 3.27 (0.70) 0.88 0.02 1.00
Repair 3.59 (0.90) 0.85 0.13 0.52*** 1.00
Study 2 (n ¼ 60)
Attention 3.85 (0.39) 0.71 1.00
Clarity 3.66 (0.59) 0.86 0.01 1.00
Repair 3.74 (0.56) 0.64 0.24 0.30* 1.00
Study 3 (n ¼ 48)
Attention 3.88 (0.61) 0.88 1.00
Clarity 3.28 (0.55) 0.74 0.08 1.00
Repair 3.68 (0.94) 0.86 0.04 0.22 1.00
*p<0.05; ***p<0.001.
 TRAIT META-MOOD SCALE 615

of guilt and worthlessness, feelings of helplessness and hopelessness, psychomotor

retardation, loss of appetite, and sleep disturbance. Cronbach’s alpha was 0.86.
Physical Symptoms Physical symptoms were measured using the 23-item somatic sub-
scale of the Symptom Questionnaire (Kellner, 1987). Items from this questionnaire are
brief, and involve yes/no responses. Cronbach’s alpha was 0.83.
Interpersonal Satisfaction Scale Satisfaction with interpersonal functioning, was
measured with a 7-item scale designed for this study. Items included ‘‘I am happy
with my friendships’’ and ‘‘I have strong, secure family relationships.’’ Items were
assessed along 10-point scales. Cronbach’s alpha for the scale was 0.67.

Results
Downloaded by [RMIT University] at 16:19 25 March 2013

Means and standard deviations for the TMMS subscales and intercorrelations among
them are shown in Table I. Internal consistency for all three subscales was strong, with
Cronbach’s alphas ranging from 0.82 to 0.85. Consistent with previous studies (Salovey
et al., 1995), no significant correlations emerged between Attention and Clarity or
Repair, but greater Clarity in distinguishing among moods was associated with greater
skill at mood Repair (r (104) ¼ 0.52, p<0.001).
Pearson correlations between the TMMS subscales and the criterion scales are shown
in Table II. Greater attention to mood was associated with greater empathy (r(104)
¼ 0.44, p<0.001), and perceptions of ability to distinguish among mood and skill at
mood repair were associated with lower levels of symptom reporting, social anxiety,
and depression (rs<0.30 , ps<0.01). Further, greater perceived skill at distinguishing
between emotions (Clarity) and repairing mood (Repair) were associated with greater
levels of satisfaction with interpersonal relationships (r(93) ¼ 0.39 for Clarity;
r(94) ¼ 0.31 for Repair, ps<0.01). All three TMMS subscales were positively associated
with self-esteem (rs > 0.22, ps<0.05).

Discussion
As in previous studies (e.g., Salovey et al., 1995), we found that clarity in distinguishing
between moods and skill at mood repair were associated with lower levels of depression.
Further, unlike Goldman et al. (1996), who found that under conditions of stress, per-
ceived ability to attend to moods was related to greater symptom reports, in Study 1,
skill at distinguishing between and repairing moods was linked with lower levels
of symptom reporting. However, findings from Study 1 are based on participants

TABLE II Correlations among the TMMS subscales and measures of

psychological, physical and interpersonal functioning, Study 1

Attention Clarity Repair

Empathy 0.44*** 0.04 0.11

Symptom reporting 0.04 0.30** 0.35**
Social anxiety 0.11 0.30** 0.37**
CES-depression 0.00 0.32** 0.47***
Rosenberg self-esteem scale 0.22* 0.34*** 0.65***
Interpersonal satisfaction scale 0.17 0.39*** 0.31**
*p<0.05; **p<0.01; ***p<0.001.
 616 P. SALOVEY et al.

reporting under non-stressed condition, whereas Goldman et al., found an interaction

of attention with distress in predicting greater symptom reports. Perhaps attending to
moods is associated with symptom reporting under conditions of stress, whereas
under basal conditions, skills at distinguishing between and repairing moods are
linked with less symptom reporting. Future research might examine associations
between the aspects of PEI and symptom reporting under both stressed and non-
stressed conditions. Finally, extending previous research with the TMMS, Clarity
and Repair were correlated with lower social anxiety, and greater empathy and
interpersonal satisfaction.

STUDY 2
Downloaded by [RMIT University] at 16:19 25 March 2013

Given these relations among PEI and measures of psychological, physical, and inter-
personal functioning, in Study 2, we examined potential mechanisms underlying
these associations. Specifically, we examined the relations between PEI, as measured
by the TMMS subscales, and psychological and physiological responses to repeated
laboratory stressors. Responses included salivary cortisol secretion, negative affect,
state and trait coping, and perceptions of threat from stressors. We expected that
greater PEI is associated with attenuated cortisol secretion in response to the repeated
stressors, and greater habituation to the stressors over time. Further, we expected that
greater PEI is related to adaptive coping and appraisal responses, including more active
coping, less passive coping, and perceptions of stressors as less threatening.

Method
Participants
Sixty women, ages 30–45, recruited from a Northeastern urban community, partici-
pated in a study of stress and cortisol reactivity among women with central and peri-
pheral body fat distribution (Epel et al., 2000). Participants were nonsmokers who
reported no history of psychological or physical illness, or use of prescription medica-
tions that might affect the HPA axis. Participants abstained from caffeine for 4 h and
from food and drink for 1 h before each laboratory session.

Procedure
At a baseline session, participants completed Trait Coping measures, followed by three
3-hour laboratory sessions on 3 consecutive days. Laboratory sessions were scheduled
between 4:00 and 5:30 p.m. to control for diurnal variation in cortisol levels. Stress
sessions consisted of a 30-minute baseline period during which subjects read neutral
magazines, 15 min of filling out neutral questionnaires (including the TMMS), then
45 min of stress. All stress periods involved an anticipation period and an identical
series of challenges, adapted from the Trier Social Stress Test (Kirschbaum et al.,
1993). Challenges included visiospatial puzzles, serial subtraction, and a videotaped
speech task. Tasks were made more stressful through the use of unrealistic time
constraints. Eight salivary cortisol samples were collected during each stress session,
including 2 baseline samples, 6 samples during stress, and 2 recovery samples 30 min
 TRAIT META-MOOD SCALE 617

and 1 h after the completion of the stress period. Saliva samples were collected at
matched time intervals throughout each session, at the following times: while resting
(15 and 30 min), before stress (45 min), during stress (60 and 70 min), at the cessation
of stress (90 min), and two recovery samples 30 and 60 min after stress (120 and
150 min).1 Self-reported mood was assessed before and after the stress periods on all
three sessions (see Mood Reactivity below). State coping and threat perceptions were
assessed following Session 1.

Measures

Trait Meta-Mood Scale (TMMS) Described in Study 1.

Downloaded by [RMIT University] at 16:19 25 March 2013

Negative Mood Reactivity Negative mood was assessed during baseline and immedi-
ately following the stress period for all 3 stress sessions using the Profile of Mood
States (POMS; McNair et al., 1981). Residual gains were calculated as a measure of
post-stress mood regressed on pre-stress mood. For a summary measure of increases
in negative mood after exposure to the stressors, we took the mean across all 3 sessions
of the negative mood residuals from the depression, anger, and anxiety subscales of
the POMS.

Trait Coping Trait coping was assessed through a combination of two instruments, the
COPE (Carver, et al., 1989), a multidimensional inventory designed to assess disposi-
tional strategies for coping with stress, and portions of Stanton’s coping questionnaire
(Stanton et al., 1994). The COPE consists of 60 items measured along 4-point scales
with items assessing different strategies for coping with stress (e.g., seeking of emotional
support, positive reinterpretation, acceptance, behavioral disengagement, denial). The
10 included items from Stanton’s measure assessed active coping through emotional
approach (processing and expressing one’s feelings). Following Tobin’s hierarchical
model of coping strategies (Tobin et al., 1989), we created two global constructs: (a)
approach or active coping (Cronbach’s alpha ¼ 0.93) and (b) avoidant or passive
coping (Cronbach’s alpha ¼ 0.80). Active coping included strategies such as concentrat-
ing, planning, seeking instrumental and emotional support, and processing and expres-
sing one’s emotions. Passive coping included disengagement strategies including
denial, giving up, and using drugs, alcohol, eating, sleep, or movies to feel better or
forget the problem.

State Coping State Coping was assessed immediately after participants completed the
challenges on the first session. An adapted version of the COPE assessing strategies for
coping with the specific laboratory challenges in this study was administered (Baggett
et al., 1996). Sample items included, ‘‘I saw the math as challenging,’’ and ‘‘I acted
as though you weren’t really serious about asking me to do the math.’’ Similar to

1
Peak cortisol response occurred at the cessation of stress (at 90 min), around 20 min after the speech
stressor began, which was likely the most stressful of the tasks.
 618 P. SALOVEY et al.

Trait Coping, State Coping was divided into two global constructs (a) active coping
(Cronbach’s alpha ¼ 0.71) and (b) passive coping (Cronbach’s alpha ¼ 0.84).

Cognitive Appraisal Cognitive Appraisal of the challenge was measured with items
assessing the extent to which participants perceived each task as challenging or threat-
ening. Items were rated along 4-point scales, and included two questions for each task
(e.g., ‘‘I saw the math task as challenging’’ or ‘‘I saw the math task as threatening’’).
Participants rated the questions from 1 (not at all) to 4 (a lot). As ratings across the
two items and across the math and puzzle tasks were highly intercorrelated, we created
a composite challenge/threat appraisal scale across both tasks (Cronbach’s
alpha ¼ 0.70).

Salivary Cortisol Saliva samples were collected using salivettes (Sarstedt, Rommelsdorf,
Downloaded by [RMIT University] at 16:19 25 March 2013

Germany) that were frozen until lab analysis. Samples were assayed in duplicate by
radioimmunoassay at the Yale Medical School General Clinical Research Center
(GCRC) using a commercial kit (Diagnostic Products Corporation, Los Angeles,
CA). Inter- and intra-assay coefficients of variation ranged from 4–5%. Cortisol area
under the curve (AUC), a summary measure of total cortisol secreted, was calculated
for each session. We also calculated mean AUC over the three stress sessions.
Resting baseline cortisol was calculated as the mean of the 2 baseline measures on
each stress day.

Results
Means, SDs, Cronbach’s alphas, and intercorrelations among the TMMS subscales are
shown in Table I. Similar to Study 1, we found satisfactory internal consistency for all
scales (Cronbach’s alphas ranged from 0.64 to 0.86). Again, Clarity in distinguishing
among feelings was significantly correlated with skill at mood repair (r(60) ¼ 0.30,
p<0.05), but no significant associations between Attention and either Clarity or
Repair emerged.
Next, Pearson correlations between the TMMS subscales and appraisal, mood reac-
tivity, and state and trait coping were calculated. Greater skill at mood Repair was
associated with perceptions of stressors as less threatening (r(60) ¼ 0.35, p<0.01)
and less trait and state passive coping (rs (60) ¼ 0.31 and 0.34, ps<0.05). Greater
Clarity in distinguishing among moods was associated with greater levels of negative
mood following the stressor, controlling for baseline negative mood (mood reactivity)
(r(60) ¼ 0.32, p<0.05). Relationships among the other TMMS subscales and appraisal,
coping, and mood reactivity were not significant.
Table III shows Pearson correlations among the TMMS subscales and baseline
and total cortisol levels (AUCs) over the 3 stress sessions. Greater clarity in
distinguishing between moods was associated with lower baseline cortisol levels on
days 1 and 2 (rs(60) ¼ 0.28 and 0.30, respectively, ps<0.05). Contrary to our predic-
tions, skill at mood Repair was associated with greater cortisol at baseline on day 3
(r(60) ¼ 0.30, p<0.05). Finally greater skill at distinguishing among moods (Clarity)
was associated with lower cortisol secretion (AUC) during stress across all days, and
particularly on day 2 (rs(60) ¼ 0.31 and 0.40, respectively, ps<0.05). Perceiving
one’s moods clearly, then, appears associated with lower adrenocortical activity
during a laboratory stress protocol.
 TRAIT META-MOOD SCALE 619

TABLE III Correlations among the TMMS subscales and cortisol secretion
during repeated stressors, Study 2

Attention Clarity Repair

Baseline Cortisol
Day 1 0.01 0.28* 0.18
Day 2 0.05 0.30* 0.13
Day 3 0.20 0.18 0.30*
Total Cortisol
Day 1 AUC 0.06 0.22 0.05
Day 2 AUC 0.15 0.40** 0.03
Day 3 AUC 0.15 0.19 0.11
Mean AUC 0.09 0.31* 0.06
*p<0.05; **p<0.01.
Baseline cortisol was calculated as the mean of 2 baseline measures on each stress day.
AUC ¼ area under the curve, a measure of total cortisol secreted during each session.
Downloaded by [RMIT University] at 16:19 25 March 2013

Mean AUC ¼ mean level of total cortisol secreted across the 3 stress sessions.

We also examined correlations between the TMMS subscales and degree of cortisol
habituation to the repeated stress.2 Habituation was defined as (Day 1 AUC)(Day 2
AUC)(Day 3 AUC). Although there were no significant correlations between degree
of habituation and Clarity and Repair (rs ¼ 0.19 and 0.13, respectively, ps ¼ ns), we
found a significant positive correlation between degree of habituation and Attention
(r(60) ¼ 0.27, p<0.05). Greater attention to moods was associated with greater habitua-
tion to repeated stressors.
Finally, we conducted a simultaneous multiple regression to examine the relation
among the 3 TMMS subscales and Mean Cortisol AUC over the 3 days, controlling
for the influence of baseline cortisol and overlap among the TMMS subscales. After
controlling for basal cortisol, Repair, and Attention, we found a significant relation-
ship between Clarity and average cortisol AUC ( ¼ 0.35, t ¼ 2.50, p<0.05). Again,
these findings suggest that even after having controlled for baseline cortisol levels,
Clarity in distinguishing between moods is related to attenuated cortisol secretion
across a laboratory stress protocol. However, as the regression model as a whole
accounted for a nonsignificant 12% of the variance in Mean Cortisol AUC (Model
R2 ¼ 0.12, F(4, 54) ¼ 1.80, p<0.15), these findings should be interpreted with caution.

Discussion
As predicted, we found that greater PEI was associated with adaptive psychological
coping, attenuated cortisol release following repeated stressors, and greater habitua-
tion to repeated stressors. In particular, skill at mood regulation (Repair) was related
to perceptions of stressors as less threatening, and less use of state and trait passive
coping strategies. Interestingly, Repair was also linked with greater baseline
cortisol levels on the third day of repeated stress suggesting possible associations
with repression.
Clarity, although associated with increased negative mood following stress, was also
linked with lowered cortisol release at baseline, and throughout repeated stress sessions.

2
Overall, there were significant differences between Day 1 AUC and Day 2 AUC (t ¼ 3.23, p<0.01), but no
significant differences between Day 1 and Day 3, and Day 2 and Day 3 AUCs.
 620 P. SALOVEY et al.

Thus, although individuals who are able to distinguish their moods clearly may experi-
ence more negative mood following repeated stress, the increased negative mood does
not translate into physiological arousal, which may reduce damage to physical systems
and poor health outcomes (e.g., McEwen, 1998).
Finally, greater attention to moods was correlated with greater adrenocortical habi-
tuation to repeated stressors. Thus, unlike Attention’s association with an arguably
maladaptive outcome (greater symptom perception) in Goldman et al. (1996), in
Study 2, attention to mood was related to an adaptive physiological outcome (greater
habituation to repeated stressors).

STUDY 3

In Study 3, we further examined links between PEI and psychophysiological measures

Downloaded by [RMIT University] at 16:19 25 March 2013

of adaptive coping with stress. This study extended findings from Study 2 to cardio-
vascular (blood pressure) as well as HPA responses to stress, and to both men and
women. As in Studies 1 and 2, we expected that greater PEI would correlate with
more adaptive coping (more active strategies, less rumination), and, as in Study 2,
with attenuated cortisol and blood pressure changes following laboratory stress.

Method
Participants
Forty-eight (21 male and 27 female) undergraduates (ages 17–23) recruited from
Introductory Psychology classes and from signs around campus participated in a
larger study of adrenocortical and blood pressure responses to interpersonal and achieve-
ment stress (Stroud et al., 1999). All participants reported that they were in good physical
and mental health and did not use prescription medications. Participants abstained from
eating or drinking (except water) for 2 h before the stress session, from exercise or alcohol
for 24 h, and from caffeine for 12 h before the stress session.

Procedure
To allow physiological habituation to the laboratory, participants initially completed a
1.5-h rest session involving filling out a large packet of questionnaires. On a later day,
participants were scheduled between 3:00 and 5:30 PM to complete a 2.5-h stress session.
The stress session included baseline, stress, and recovery periods. During the baseline
period, participants read neutral magazines for 10 min, then filled out questionnaires
for 10 min. For the stress period, participants were randomly assigned to an achievement
or interpersonal condition. The achievement condition lasted 45 min and involved chal-
lenging arithmetic problems using a hypothetical new numbering system (30 min), and
memorization and recitation of a difficult passage (15 min). Time pressure was applied
by the experimenter to increase participants’ distress. The interpersonal condition
lasted 30 min and involved a rejection paradigm in which participants were asked to
engage in two 15-min conversation segments (fun activities on weekends, friendships
at Yale) while being excluded by 2 confederates (Stroud et al., 2000). Following the stres-
sors, participants completed more questionnaires, including the TMMS, while
listening to soft, classical music. Salivary cortisol and blood pressure measures were
 TRAIT META-MOOD SCALE 621

collected at 6 time points throughout each stress session, including 2 baseline samples
(one following the reading of magazines, and one following the 10-min questionnaire
period), 2 samples during stress (one after each stressor), and 2 post-stress samples 15
and 30 min after termination of the stressors.

Measures
Trait Meta-Mood Scale (TMMS) Described in Study 1.

Salivary Cortisol As in Study 2, saliva samples were collected from each participant
using the Salivette sampling device (Sarstedt, Rommelsdorf, Germany). Cortisol
samples were again assayed by the General Clinical Research Center (GCRC) at
Yale with inter- and intra-assay coefficients of variation ranging from 6 to 16%.
Downloaded by [RMIT University] at 16:19 25 March 2013

Cortisol reactivity was calculated by subtracting peak cortisol levels from the mean
of the 2 baseline measures. As cortisol values were positively skewed, analyses were
based on logarithmic transformations of all cortisol values.

Systolic and Diastolic Blood Pressure (SBP and DBP) were measured with an Omron
Auto-Inflation digital sphygmomanometer (Model HEM-706) at 6 time points during
the stress sessions. The HEM-706 has passed accuracy standards of the Association
of Medical Instrumentation, with mean differences between HEM-706 and stethoscopic
readings ranging from 1.2 to 4.4 mm Hg (Foster et al., 1994). SBP and DBP reactivity
scores were calculated by subtracting peak SBP or DBP from the mean of the 2 baseline
SBP or DBP measures.

Trait Coping Trait coping was assessed with the COPE (Carver et al., 1989), described
in Study 2. Active and passive coping scales were formed. Internal consistency for the
active scale was good (Cronbach’s alpha ¼ 0.84); for the passive scale, internal consis-
tency was low (Cronbach’s alpha ¼ 0.41).

Rumination Tendency to ruminate and engage in distraction in response to negative

mood was measured using the Response Styles Questionnaire (Nolen-Hoeksema and
Morrow, 1991). Participants were asked to indicate how they typically respond to
depressed mood. The rumination scale included items such as ‘‘Think ‘Why do I always
react this way’,’’ and ‘‘Think, ‘I need to understand these feelings’.’’ Distraction was
measured with items such as ‘‘Think, ‘I’ve got to get up and do something to make
myself feel better’,’’ and ‘‘Think, ‘These feelings won’t last’.’’ Internal consistency for
both scales was good (Cronbach’s alpha were 0.90 and 0.89, respectively).

Results
Means, SDs, Cronbach’s alphas, and intercorrelations among the TMMS subscales are
shown in Table I. Means and SD’s were similar to those shown in Studies 1 and 2.
Internal consistency was good for all scales (Cronbach’s alphas > 0.74) and similar
to Studies 1 and 2. Intercorrelations followed the same pattern as those from Studies
1 and 2, with little relation between Attention and either Clarity or Repair, but a
larger correlation between Clarity and Repair, which however was not significant in
this study (r(48) ¼ 0.22).
 622 P. SALOVEY et al.

TABLE IV Correlations among TMMS subscales and state and trait coping,
cortisol, systolic blood pressure (SBP), and diastolic blood pressure (DBP)
reactivity to acute stress, Study 3

Attention Clarity Repair

Trait coping active 0.02 0.23 0.44**

Passive 0.08 0.03 0.10
Response styles rumination 0.26 0.03 0.56***
Distraction 0.08 0.17 0.55***
Cortisol reactivity 0.30* 0.01 0.18
SBP reactivity 0.36** 0.07 0.18
DBP reactivity 0.22 0.10 0.05
*p<0.05; **p<0.01; ***p<0.001.

Pearson correlations between TMMS subscales and trait coping, rumination, and dis-
Downloaded by [RMIT University] at 16:19 25 March 2013

traction, and salivary cortisol, SBP, and DBP reactivity are shown in Table IV. Skill at
mood repair was associated with higher levels of active trait coping, higher levels of dis-
traction following negative mood (rs (48) > 0.44, ps<0.05), and lower levels of rumina-
tion following negative mood (r(48) ¼ 0.56, p<0.001). No associations between the
TMMS subscales and trait passive coping were found; however internal consistency
for this scale was low. Finally, greater attention to moods was associated with attenuated
cortisol and SBP reactivity to stress (rs(48)<0.30, ps<0.05).
We then conducted a series of simultaneous multiple regression analyses to examine
the relations among the 3 TMMS subscales and cortisol, and SBP reactivity, controlling
for baseline levels and overlap between the TMMS subscales. Controlling for baseline
levels as well as other TMMS subscales, greater attention to mood was associated with
lower peak levels of cortisol ( ¼ 0.24, t ¼ 2.10, p<0.05) and SBP ( ¼ 0.23, t ¼ 2.55,
p<0.05). Regression analyses accounted for 47% and 67% of the variance in peak phy-
siologic responses, respectively. (For cortisol, Model R2 ¼ 0.47, F(4, 43) ¼ 9.53,
p<0.0001, and for SBP, R2 ¼ 0.67, F(4, 43) ¼ 21.93, p<0.0001).

Discussion
As in Study 2, we found that skill at mood repair was associated with psychological
changes reflecting adaptive coping with stress. Complementing the findings of Study 2,
in which skill at mood Repair was related to lower levels of passive coping, in
Study 3, Repair was correlated with greater levels of active coping. Skill at mood
Repair was also associated with lower levels of rumination, a construct that has been
linked to physiological reactivity and poor health outcomes (Pennebaker, 1995), and
higher levels of distraction. With respect to physiological stress reactivity, Study 3
showed a relationship between PEI and attenuated cortisol as well as cardiovascular
(SBP) responses to stress. In particular, increased attention to mood was related to
more adaptive (lowered) physiological responses to acute stress.

GENERAL DISCUSSION

In three studies, we examined the relations between aspects of PEI, defined as the ability
to attend to, distinguish among, and regulate moods, and a number of indices of adap-
tive coping. We extended previous research in which PEI was associated with both
 TRAIT META-MOOD SCALE 623

health outcomes and psychological responses to laboratory stressors (Salovey et al.,

1995; Goldman et al., 1996), to interpersonal functioning, coping styles, and physiolo-
gical responses to stressors. Given links between both emotional functioning and phy-
siological stress reactivity and health outcomes (McEwen, 1998), these studies lend
preliminary support for psychophysiological stress reactivity as a potential mechanism
linking emotional functioning with health.
Similar to previous studies, in Study 1, we found that aspects of PEI (specifically, the
ability to distinguish between and repair moods) were related to lower levels of depres-
sion. Perceived ability to distinguish between and repair moods was also found to be
related to lower levels of symptom reporting, a finding different from our previous
work (Goldman et al., 1996), in which we found that perceived attention to mood
was associated with greater levels of symptom reporting. However, Goldman et al.’s
study described associations under conditions of stress, whereas the present study repre-
Downloaded by [RMIT University] at 16:19 25 March 2013

sented associations under non-stressed conditions. Extending previous studies, we also

found that aspects of perceived emotional intelligence were also related to interpersonal
functioning. Specifically, consistent with literature showing associations between
emotional competence and interpersonal functioning (Cooper et al., 1998; Saarni,
1999), clarity in distinguishing among moods and skill at mood repair were associated
with lower levels of social anxiety, and greater interpersonal satisfaction.
Across Studies 2 and 3, we found relationships between aspects of PEI and psycho-
logical and physiological measures of coping with acute and repeated laboratory
stressors. Perceptions of skill at mood repair were associated with less state and trait
passive coping (giving up, avoiding, or inhibiting active responses to stressful situa-
tions), greater trait active coping (active steps to change a stressful situation or amelio-
rate its effects), less use of rumination in response to negative mood, and perceptions of
stressors as less threatening. However, although skill at mood repair was associated
with arguably more adaptive coping strategies, it was also associated with greater use
of distraction (participating in other activities to take one’s mind off one’s mood),
which might be considered an avoidant coping strategy, and greater baseline cortisol
levels on the third day of repeated stress. Heightened physiological responses, combined
with avoidant coping might suggest an association between Repair and repression that
might be further examined in future studies.
Clarity, or perceptions of ability to distinguish among moods, was associated with
greater increases in negative mood following repeated stressors, but lowered cortisol
release during baseline and repeated stress. Thus, although individuals who are able
to distinguish their moods clearly may experience more negative mood following
repeated stress, the increased negative mood does not translate into physiological arou-
sal. Although preliminary, these findings suggest that willingness to distinguish negative
feelings when stressed may play a role in reducing the damaging physiological conse-
quences from stress and potentially reducing negative health outcomes. Results are con-
sistent with studies by Pennebaker and colleagues in which individuals who disclosed
emotions and details relating to a traumatic event evidenced greater negative mood
immediately following the disclosure, but lower skin conductance levels, and better
health outcomes in the long run (Pennebaker and Beall, 1986; Pennebaker, 1995).
Results also extend those of Salovey et al. (1995), in which Clarity was associated with
a greater decline in ruminative thought following an experimental stressor.
Attention to moods was associated with greater cortisol habituation to repeated
stressors, and lowered cortisol and cardiovascular (systolic blood pressure) responses
 624 P. SALOVEY et al.

to acute laboratory challenges. Thus, under differing circumstances, both attention to

mood, and the ability to distinguish among moods may lead to lowered physiological
responses to stress, which may eventuate in less damage to physiological systems and
better health outcomes (e.g., McEwen, 1998). Interestingly, in a previous study, greater
attention to mood was associated with greater symptom perception, but not reports of
illness under increasing distress (Goldman et al., 1996), while attention was related to
arguably positive consequences in the present study. Although future research is
needed to look at more complex relationships between attention to moods, symptom
perception, physiological responses to stress, and health, it is possible that attending
more closely to moods may be linked with greater attention to physical symptoms;
however, this does not appear to correlate with either health outcomes or physiological
responses to stress. Last, we note that only skills at identifying and paying attention to
moods were associated with physiological stress reactivity; perceived skill at mood
Downloaded by [RMIT University] at 16:19 25 March 2013

repair was not. Perhaps if moods are not identified and attended to, physiological arou-
sal results too quickly for repair to have an impact.

Limitations
Although suggestive, results from these studies should be interpreted cautiously. First,
with respect to the validity of the emotional intelligence construct, the obvious limita-
tion of these studies is their reliance on self-report measures. Emotional intelligence – as
a set of competencies concerning the processing of emotion-relevant information – is
likely to be measured with greatest validity when it is assessed as a set of competencies
or skills (Mayer et al., 1999). Self-reported assessments in this domain may not be espe-
cially accurate or even available to conscious introspection. It is unlikely that test items
such as ‘‘I think I’m a pretty smart person’’ would make for a valid measure of IQ; the
usefulness of analogous questions about one’s emotional intelligence is also doubtful.
What we have measured here should not really be considered emotional intelligence
per se, but rather beliefs about emotional intelligence, a kind of emotional intelligence
self-efficacy. Nonetheless, such self-reported appraisals have been important in other
domains, and we expect that studies in which both task-based and self-report measures
are included may reveal that both account for significant variance in important physio-
logical, cognitive, and social outcomes.
Second, it is difficult to determine the direction of causality in these studies. For
example, the relation between PEI and stress reactivity brings up the classic
emotion–physiology debate. On one hand, as we have described thus far, Clarity in dis-
tinguishing among feelings and Attention to mood may predict lower physiological
arousal in response to stress. On the other hand, it is possible that individuals with
lower physiological arousal in response to stress are better able to attend to and identify
their feelings, because physiological arousal is not impeding these processes. Likewise,
individuals high in arousal may be unable to distinguish between and attend to mood as
they are overwhelmed by the physiological arousal they experience. Future research to
tease apart directional effects between PEI, stress reactivity, and health is warranted.
Third, in a small subset of participants in Study 3, we found significant correlations
between the Repair subscale and a measure of social desirability. Although such asso-
ciations were not examined for all participants in all studies, it appears that desire to be
perceived favorably may influence responding to the Repair subscale of the TMMS.
Future studies might further examine the influence of social desirability on participants’
 TRAIT META-MOOD SCALE 625

perceptions of their ability to repair mood, and control for such effects in examining the
influence of Repair on other outcomes.
Finally, although physiological habituation to repeated stressors has consistently
been considered an adaptive response (and lack of habituation a maladaptive response),
what constitutes an adaptive response to acute stress remains a question (Dientsbier,
1989; Ratliff-Crain and Vingerhoets, 1996; McEwen, 1998). In Study 3, we have
described attenuated physiological responses to acute stress as adaptive; however, it
is also possible that greater physiological responses to acute stress may be more adap-
tive, depending on the individual and the specific situation. Thus, it is possible that
attention to moods is correlated with a less adaptive physiological response to stress.
Future research to determine what constitutes adaptive and maladaptive acute stress
responses may help to elucidate this issue.
Downloaded by [RMIT University] at 16:19 25 March 2013

CONCLUSIONS

Although there are some limitations to the studies presented here, PEI, in general, was
related to a wide array of indices of adaptive psychophysiological responses across a
variety of methodologies. These results suggest some relationship between perceived
emotional competence and health that is deserving of further study. The importance
of this relationship is further highlighted by the failure of many trait measures to
predict physiological (particularly cortisol) reactivity to laboratory stressors in the
past research (Kirschbaum et al., 1992; Van Eck et al., 1996). We advocate further
examination of psychophysiological stress reactivity as a potential mechanism linking
emotional intelligence with health, and believe such research will further the fields of
emotions and health, as well as general psychobiology.

Acknowledgements
Preparation of this manuscript was facilitated by the following grants: American
Cancer Society (RPG-93-028-05-PBP), National Cancer Institute (R01-CA68427),
and National Institute of Mental Health (P01-MH/DA56826) as well as funding
from the Ethel F. Donaghue Women’s Health Investigator Program at Yale
University. We also thank the General Clinical Research Center at the Yale School
of Medicine for carrying out the salivary cortisol analyses for Studies 2 and 3.

References
Baggett, H.L., Saab, P.G. and Carver, C.S. (1996). Appraisal, coping, task performance, and cardiovascular
responses during the evaluated speaking task. Personality and Social Psychology Bulletin, 22, 483–494.
Carver, C., Scheier, M. and Weintraub, J. (1989). Assessing coping strategies: a theoretically based approach.
Journal of Personality and Social Psychology, 56, 267–283.
Cooper, M.L., Shaver, P.L. and Collins, N.L. (1998). Attachment styles, emotional regulation, and adjust-
ment in adolescence. Journal of Personality and Social Psychology, 74, 1380–1397.
Dienstbier, R.A. (1989). Arousal and physiological toughness: implications for mental and physical health.
Psychology Review, 96, 84–100.
Epel, E., McEwen, B., Seeman, T., Matthews, K., Castellazzo, G., Brownell, K., Bell, J. and Ickovics, J.
(2000). Stress and body shape: consistently greater stress-induced cortisol reactivity among women
with abdominal fat. Psychosomatic Medicine, 62, 623–632.
 626 P. SALOVEY et al.

Fenigstein, A., Scheier, M.F. and Buss, A.H. (1975). Public and private self-consciousness: assessment and
theory. Journal of Consulting and Clinical Psychology, 43, 522–527.
Foster, C., McKinlay, S., Cruickshank, J.M. and Coats, A.J.S. (1994). Accuracy of the Omron HEM 706 por-
table monitor for home measurement of blood pressure. Journal of Human Hypertension, 8, 661–664.
Goldman, S.L., Kraemer, D.T. and Salovey, P. (1996). Beliefs about mood moderate the relationship of stress
to illness and symptom reporting. Journal of Psychosomatic Research, 41, 115–128.
Gottman, J.M. and Levenson, R.W. (1986). Assessing the role of emotion in marriage. Behavioral Assessment,
8, 31–48.
Gross, J.J. (1998). The emerging field of emotion regulation: an integrative review. Review of General
Psychology, 3, 271–299.
Kellner, R. (1987). A symptom questionnaire. The Journal of Clinical Psychiatry, 48, 268–274.
Kirschbaum, C., Bartussek, D. and Strasburger, C.J. (1992). Cortisol responses to psychological stress and
correlations with personality traits. Personality and Individual Differences, 13, 1353–1357.
Kirschbaum, C., Pirke, K. and Hellhammer, D. (1993). The ‘‘Trier Social Stress Test’’ – a tool for investigat-
ing psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28, 76–81.
Manuck, S.B. and Krantz, D.S. (1986). Psychophysiologic reactivity in coronary heart disease and essential
hypertension. In: Matthews, K.A., Weiss, S.B., Deter, T., Dembroski, T., Falkner, B., Manuck, S.B.
and Williams, R.B. Jr. (Eds.), Handbook of Stress, Reactivity and Cardiovascular Disease, pp. 11–47.
Downloaded by [RMIT University] at 16:19 25 March 2013

Wiley-Interscience, New York.

Mayer, J.D., Caruso, D.R. and Salovey, P. (1999). Emotional intelligence meets traditional standards for an
intelligence. Intelligence, 27, 267–298.
Mayer, J.D. and Gaschke, Y.N. (1988). The experience and meta-experience of mood. Journal of Personality
and Social Psychology, 55, 102–111.
Mayer, J.D. and Salovey, P. (1993). The intelligence of emotional intelligence. Intelligence, 22, 89–113.
Mayer, J.D. and Salovey, P. (1997). What is emotional intelligence? In: Salovey, P. and Sluyter, D. (Eds.),
Emotional Development and Emotional Intelligence:Implications for Educators, pp. 3–31. Basic Books,
New York.
McEwen, B. (1998). Protective and damaging effects of stress mediators. New England Journal of Medicine,
338, 171–179.
McNair, D.M., Lorr, M.M. and Droppleman, L.F. (1981). Profile of Mood States Manual. Education and
Industrial Testing Service, San Diego.
Mehrabian, A. and Epstein, N. (1972). A measure of emotional empathy. Journal of Personality, 40, 525–543.
Nachmias, M., Gunnar, M., Mangelsdorf, S., Paritz, R.H. and Buss, K. (1996). Behavioral inhibition and
stress reactivity: the moderating role of attachment security. Child Development, 67, 508–522.
Nolen-Hoeksema, S. and Morrow, J. (1991). A prospective study of depression and posttraumatic stress
symptoms after a natural disaster: the 1989 Loma Prieta earthquake. Journal of Personality and Social
Psychology, 61, 115–121.
Nolen-Hoeksema, S., Parker, L.E. and Larson, J. (1994). Ruminative coping with depressed mood following
loss. Journal of Personality and Social Psychology, 67, 92–104.
Pennebaker, J.W. (1995). Emotion, Disclosure, and Health. American Psychological Association, Washington,
DC.
Pennebaker, J.W. and Beall, S.K. (1986). Confronting a traumatic event: toward an understanding of inhibi-
tion and disease. Journal of Abnormal Psychology, 95, 274–281.
Radloff, L.S. (1977). The CES-D scale: a self-report depression scale for research in the general population.
Applied Psychological Measurement, 1, 385–401.
Ratliff-Crain, J. and Vingerhoets, A. (1996). Letter to the editor. Psychosomatic Medicine, 58, 392–393.
Rosenberg, M. (1965). Society and the Adolescent Self-Image. Princeton University Press, Princeton, NJ.
Saarni, C. (1999). The Development of Emotional Competence. Guilford Press, New York.
Salovey, P., Bedell, B.T., Detweiler, J.B. and Mayer, J.D. (2000). Current directions in emotional intelligence
research. In: Lewis, M. and Haviland-Jones, J.M. (Eds.), Handbook of Emotions, 2nd Edition, pp. 504–
520. Guilford Press, New York.
Salovey, P. and Mayer, J.D. (1990). Emotional intelligence. Imagination, Cognition, and Personality, 9,
185–211.
Salovey, P., Mayer, J.D., Goldman, S., Turvey, C. and Palfai, T. (1995). Emotional attention, clarity and
repair: exploring emotional intelligence using the trait meta-mood scale. In: Pennebaker, J.D. (Ed.),
Emotion, Disclosure, and Health, pp. 125–154. American Psychological Association, Washington, DC.
Salovey, P., Woolery, A. and Mayer, J.D. (2001). Emotional intelligence: conceptualization and measurement.
In: Fletcher, G. and Clark, M. (Eds.), The Blackwell Handbook of Social Psychology, pp. 279–307.
Blackwell, London.
Sapolsky, R.M., Romero, L.M. and Munck, A.U. (2000). How do glucocorticoids influence stress responses?
Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews, 21, 55–89.
Stanton, A., Danoff-Burg, S., Cameron, C. and Ellis, A. (1994). Coping through emotional approach:
problems of conceptualization and confounding. Journal of Personality and Social Psychology, 66,
350–362.
 TRAIT META-MOOD SCALE 627

Stroud, L.R., Salovey, P. and Epel, E.S. (June, 1999). Sex Differences in Adrenocortical and Blood Pressure
Responses to Achievement and Interpersonal Stress. Paper presented at the annual meeting of the
American Psychological Society, Denver, CO.
Stroud, L.R., Tanofsky-Kraff, M., Wilfley, D.E. and Salovey, P. (2000). The Yale Interpersonal Stressor
(YIPS): affective, physiological, and behavioral responses to a novel interpersonal rejection paradigm.
Annals of Behavioral Medicine, 22, 204–213.
Tobin, D., Holroyd, K., Reynold, R. and Wigal, J. (1989). The hierarchical factor structure of the Coping
Strategies Inventory. Cognitive Therapy Research, 13, 343–361.
Van Eck, M., Berkhof, H., Nicolson, N. and Sulon, J. (1996). The effects of perceived stress, traits, mood
states and stressful daily events on salivary cortisol. Psychosomatic Medicine, 58, 447–458.
Weinberger, D.A. (1990). The construct validity of repressive coping. In: Singer, J.L. (Ed.), Repression and
Dissociation: Implications for Personality Theory, Psychopathology and Health, pp. 337–386. University
of Chicago Press, Chicago.
Downloaded by [RMIT University] at 16:19 25 March 2013