Content uploaded by Mary Burleson
Author content
All content in this area was uploaded by Mary Burleson on May 30, 2022
Content may be subject to copyright.
Content uploaded by Wenda Trevathan
Author content
All content in this area was uploaded by Wenda Trevathan on Mar 08, 2014
Content may be subject to copyright.
Arch Sex Behav (2007) 36:357–368
DOI 10.1007/s10508-006-9071-1
ORIGINAL PAPER
In the Mood for Love or Vice Versa? Exploring the Relations
Among Sexual Activity, Physical Affection, Affect, and Stress
in the Daily Lives of Mid-Aged Women
Mary H. Burleson ·Wenda R. Trevathan ·
Michael Todd
Received: 14 September 2005 / Revised: 20 February 2006 / Accepted: 1 May 2006 / Published online: 16 November 2006
C
Springer Science+Business Media, Inc. 2006
Abstract How do physical affection, sexual activity,mood,
and stress influence one another in the daily lives of mid-
aged women? Fifty-eight women (M age, 47.6 yrs) recorded
physical affection, several different sexual behaviors, stress-
ful events, and mood ratings every morning for 36 weeks.
Using multilevel modeling, we determined that physical af-
fection or sexual behavior with a partner on one day signifi-
cantly predicted lower negative mood and stress and higher
positive mood on the following day. The relation did not hold
for orgasm without a partner. Additionally, positive mood on
one day predicted more physical affection and sexual ac-
tivity with a partner, but fewer solo orgasms the following
day. Negative mood was mostly unrelated to next-day sexual
activity or physical affection. Sexual orientation, living with
a partner, and duration of relationship moderated some of
these effects. Results support a bidirectional causal model
in which dyadic sexual interaction and physical affection
improve mood and reduce stress, with improved mood and
reduced stress in turn increasing the likelihood of future sex
and physical affection.
M. H. Burleson ()
Department of Social and Behavioral Sciences,
Arizona State University, West Campus,
4701 W. Thunderbird Rd., Phoenix, Arizona 85069-7100
e-mail: mary.burleson@asu.edu
W. R. Treva tha n
Department of Sociology and Anthropology,
New Mexico State University,
Las Cruces, New Mexico
M. Todd
Prevention Research Center,
Berkeley, California
Keywords Sexual behavior .Mood .Mid-aged women .
Physical affection .Daily diary
Introduction
Although conventional wisdom holds that a satisfactory sex-
ual interaction will improve mood, little empirical evidence
exists to support this assumption. Surprisingly few studies
have investigated relations between sexual activity or physi-
cal intimacy and affect, and most address the issue of whether
mood (or mood disorder) affects sexual desire or arousal. To
our knowledge, none have asked whether sexual activity or
physical affection alters affective state or perceived stress on
a daily basis. The current study addresses these questions us-
ing daily diary methodology in a sample of mid-aged women.
The idea that sex and affect are related is not new. Almost
50 years ago, Wolpe (1958) postulated that sexual arousal
and anxiety were mutually inhibitory. Mood disorders, in-
cluding depression (Cyranowski et al., 2004; Kennedy,
Dickens, Eisfeld, & Bagby, 1999) and anxiety (Bodinger
et al., 2002), are linked to changes in sexual interest and
arousal. A rewarding sex life is considered an important con-
tributor to life satisfaction (Apt, Hurlbert, Pierce, & White,
1996), which in turn correlates with positive mood (Suh,
Diener, Oishi, & Triandis, 1998). Indeed, a recent panel
study of over 15,000 randomly selected adults showed that
amount of sexual activity was strongly positively related to
overall happiness (Blanchflower & Oswald, 2004). Many
models of sexual behavior, even in rodents, assume that
sex causes positive affect (Agmo, 1999). Moreover, there
are physiological reasons to expect sex or physical affec-
tion to influence emotions and stress. For example, orgasm
causes oxytocin release (Blaicher et al., 1999; Carmichael
et al., 1987), which is associated with stress relief and
positive social interactions in some contexts (Heinrichs,
Springer
358 Arch Sex Behav (2007) 36:357–368
Baumgartner, Kirschbaum, & Ehlert, 2003; Uvnas-Moberg,
1997).
Sexual function is multifaceted, and mood may influence
its various aspects (e.g., desire vs. arousal) in different ways.
Most non-clinical studies of these effects are of two main
types: laboratory studies with experimentally manipulated
mood and surveys or other studies where individuals describe
their usual sexual interest or response when they experience
particular moods or affective states.
Experimental studies suggest anger induction can reduce
sexual arousal in sexually functional men (Bozman & Beck,
1991); both anger and anxiety lead to lower reported sexual
desire in sexually functional men and women (Beck & Boz-
man, 1995; Bozman & Beck, 1991). On the other hand, in-
ducing anxiety before or during erotic stimulation enhances
sexual arousal in sexually functional men (Barlow, Sakheim,
& Beck, 1983; Wolchik et al., 1980) and women (Hoon,
Wincze, & Hoon, 1977; Palace & Gorzalka, 1990). Effects
of elated/happy and depressed/sad mood have also been in-
vestigated experimentally. Laan, Everaerd, van Berlo, and
Rijs (1995) and Meisler and Carey (1991) found induced
positive mood did not affect subsequent sexual arousal in
women, although it attenuated negative emotions elicited by
an erotic film. Similarly, Meisler and Carey (1991) found
no effect of elation on arousal or response in adult men;
depressed mood delayed subjective sexual arousal but did
not affect genital arousal. In contrast, Mitchell, DiBartolo,
Brown, and Barlow (1998) found induced elated mood facil-
itated both subjective and genital arousal, whereas induced
depressed mood reduced genital arousal.
As in experimental studies, the results from correlational
research suggest complex effects of affect on sexual func-
tioning. In a very large retrospective study of their previous
menstrual cycles, women’s sexual interest, feelings, and en-
joyment were strongly related to levels of positive mood
and energy (Warner & Bancroft, 1988). In other studies, gay
(Bancroft, Janssen, Strong, & Vukadinovic, 2003) and het-
erosexual (Bancroft et al., 2003) men were asked to describe
how they typically behaved sexually when depressed or anx-
ious. Most reported no change in sexual interest or erectile
response. For depressed mood, those who reported a change
in desire or response largely experienced reductions (only
9% of heterosexual and 16% of gay men reported an in-
crease). For anxiety, those who reported a change in desire
or response were more evenly distributed between increase
and decrease in sexual interest or performance. More than
half the participants in the heterosexual study reported occa-
sional masturbation to regulate mood, especially to reduce
anxiety and stress. Similarly, in a focus group study with
both lesbian and heterosexual participants, women most of-
ten reported that depressed mood and anger had a negative
effect on sexuality. Several stated that anxiety sometimes
added to arousal, and self-stimulation was occasionally used
to reduce anxiety or stress (Graham, Sanders, Milhausen, &
McBride, 2004).
One study of mood-sex relationships in adolescent
females used daily diary methodology. Participants prospec-
tively recorded mood and occurrence of sexual intercourse
(along with other variables) for an average of 194 days.
Multivariate results indicated that negative mood decreased
the likelihood of having intercourse on the same day
(Fortenberry et al., 2005); however, the potential effects of
prior-day intercourse on mood were not reported.
Finally, several laboratory studies have examined affec-
tive responses after sexual stimulus presentation. One study
of adult men found high genital arousal during an erotic film
predicted higher elation and lower depression ratings after-
ward, independent of mood beforehand (Meisler & Carey,
1991). Heiman (1980) found women’s sexual arousal during
erotic audiotapes and films was positively correlated with
positive affective ratings of those stimuli. Three studies of
women (Hackbert & Heiman, 2002; Laan, Everaerd, van
Bellen, & Hanewald, 1994; Senn & Desmarais, 2004)re-
ported increased positive affect or decreased negative affect
after viewing erotic material. However, some of the affect
measures included self-ratings such as “sexy” and “easy to
arouse” along with nonsexual descriptors, making their in-
terpretation somewhat difficult.
Women’s sexuality changes throughout life, and there is
evidence to suggest that sexual interest and frequency often
decline with the natural menopause transition (Avis, Stel-
lato, Crawford, Johannes, & Longcope, 2000; Dennerstein,
Dudley, & Burger, 2001). To our knowledge, however, the
only study to longitudinally monitor both sexual functioning
and mood for a period of time during the menopause tran-
sition used annual interviews (e.g., Dennerstein, Randolph,
Taffe, Dudley, & Burger, 2002). Therefore, nothing is known
about the relations between daily sexual activity and mood
or stress in this age group.
Similarly, very little is known about how social relation-
ships and personal characteristics may influence the relations
between sex and mood. A previous study of adult women
found that sexual orientation and partner availability (i.e.,
living with one’s sexual partner) were related to amount of
sexual activity (Burleson, Trevathan, & Gregory, 2002), and
longer relationship duration is associated with lower sex-
ual frequency (Call, Sprecher, & Schwartz, 1995). Given
that these three characteristics have also been linked in some
studies with mood-related variables, such as global happiness
(Biernbaum & Ruscio, 2004), well being (Kurdek, 1991), or
depression (Thompson et al., 2005), we elected to explore
them as potential moderators of the relationships between
mood and sexual behavior.
In sum, most previous research suggests that negative
mood states, particularly depressed mood and hostility, are
likely to interfere with sexual functioning in some way, that
Springer
Arch Sex Behav (2007) 36:357–368 359
positive mood states may have the opposite effect, and that
sexual arousal may influence mood, at least in men. However,
all of the laboratory effects occurred within a very short time
frame, and the question remains as to whether these transi-
tory effects have discernible consequences in daily life. Only
one of the studies prospectively monitored sexual behavior
and mood on a daily basis, and this study was carried out
in adolescent girls. No study addressed directly the issue
of whether or how sexual interactions with others influence
moods. In the current study, we used daily diary methodol-
ogy to investigate temporal relations among sexual activity,
physical affection, mood, and stress in the daily lives of
mid-aged women, a group that has not been studied in this
context. In other words, we asked whether physical affection
or sexual activity predicted mood or stress on the following
day, and vice versa. We also asked whether sexual orien-
tation, partner availability, or relationship length moderated
these associations. Finally, we also tested whether physical
affection or sexual activity was related to mood or stress on
the same day.
Method
Participants
Women were recruited nationally through advertisements in
women’s publications to participate in a large longitudinal
study of the menopause transition “as experienced in the
context of daily life events.” To ensure a sufficient number
of lesbian participants, we also advertised in the gay and les-
bian press. Women were eligible if they were aged 42 to 52,
were not pregnant, lactating, or postmenopausal, had not had
reproductive surgery, and were not taking estrogen or pro-
gesterone or using an IUD. Of the 113 who wrote to request
study materials, 52 returned no data, 3 returned fewer than
4 weeks, and data from one woman were judged unreliable,
leaving 58 participants. Participants were not paid.
Mean age at entry was 47.6 years (SD =2.2) and mean
BMI at entry was 23.28 (SD =3.5). Forty-four (75.9%) par-
ticipants described themselves as exclusively heterosexual,
whereas 14 (24.1%) described themselves as exclusively or
primarily lesbian. Forty (69%) were married or cohabiting,
and 18 (31%) were not living with their sexual partners. For
those in a relationship (n=47), the mean duration of that
relationship was 15.6 years (SD =8.3, range 0.4–31 yrs).
Thirty-nine (67%) were premenopausal (i.e., showed little
or no variation in their menstrual cycles) and 19 (33%)
were perimenopausal (i.e., had irregular cycles) at study
entry. Six (10.3%) participants had less than a baccalau-
reate degree, six (10.3%) had earned a baccalaureate de-
gree, and 46 (79.3%) had post-graduate education. Fifty-two
(89.7%) were employed outside the home, and the mean
number of years employed was 21.1 (SD =6.3). Finally,
35 (60.3%) reported annual family incomes of more than
$50,000.
Procedure
Although the current study used only the first 36 weeks of
collected data (see below), women were recruited to partic-
ipate for five years or until one year without menstruation.
Each was mailed detailed instructions, an initial question-
naire, sufficient Daily Datasheets for 252 days (36 weeks)
of recording, a prepaid postcard for requesting a new packet
when she began recording the final 42 days, and a stamped
envelope for returning completed data sheets after 36 weeks.
Subsequent packets included only the Daily Datasheets for
252 days, the prepaid postcard, and the prepaid manila enve-
lope. At entry, participants recorded demographic and per-
sonal history data as described above. From then on, they
completed a datasheet upon arising each morning, covering
the previous 24-hour period. Morning data recording was
selected to maximize recollection of sexual activities.
Measures
To enhance accurate recording of physical and sexual in-
timacy variables, we included highly detailed instructions
regarding how these items should be interpreted. They were
reported as occurrence or non-occurrence since last data
recording and were defined as follows. Physical affection:
intimate physical contact not necessarily associated with sex;
breast stimulation: sexually-related breast stimulation by self
or other; genital stimulation: any form of genital stimulation
by another person, with or without intercourse; intercourse:
sexual intercourse with a man with intravaginal ejaculation
and no condom. We derived two additional variables: orgasm
with partner (orgasm in the context of genital stimulation or
intercourse, as defined above) and orgasm without partner
(orgasm without genital stimulation by another person or
intercourse).
The stress variable was occurrence of a stress event (de-
fined as any event that upset the participant), and thus incor-
porated perceived stress. Mood items included seven posi-
tive (enthusiastic, determined, alert, happy, cheerful, proud,
and contented) and eight negative (anxious, depressed, dis-
tressed, nervous, hostile, ashamed, guilty, and irritable) ad-
jectives, rated from 1 (experienced slightly or not at all)to
5(experienced extremely). Composite positive and negative
scores were derived for each day. In addition, to facilitate
comparisons with previous studies of mood effects on sex-
ual functioning, we created separate composite scores called
anxiety (anxious, distressed, nervous), and hostility (hos-
tile, irritable), and analyzed the depressed mood score as a
single-item indicator.
Springer
360 Arch Sex Behav (2007) 36:357–368
Data analysis
Missing data
For the entire study, durations of data recording ranged from
116 to over 2500 days. Nine participants did not fully com-
plete the first packet of questionnaires, and eight dropped
out after completing the first packet. Thus, to maximize the
number of participants included in the current analyses, we
chose to use the first 252 days (36 weeks) of data. Of the 58
participants included, 49 participated for at least 252 days,
and the remaining nine ranged from 116 to 250 days, for a
total of 14,218 possible person-days of data, and an aver-
age duration of 35 weeks.1For various reasons (e.g., illness,
travel, “forgot,” and unknown), participants failed to record
any data on 276 of these person-days, yielding a panelwise
missingness rate of 1.94%.
As noted above, nine participants contributed less than
252 days of data. They withdrew from the study for vari-
ous reasons, including their lives becoming too complicated
to continue participation (e.g., family illness; 2 participants)
and vacation periods interrupting their daily recordings so
extensively that they did not continue (2 participants). Five
participants stopped recording for no stated reason and re-
turned to us the records that they had completed.
Analysis plan
The data were intrinsically hierarchical with multiple daily
observations (Level 1) nested within participants (Level 2),
so we used multilevel modeling to take full advantage of the
data structure. We tested same-day relations using pooled
within-person correlations computed in Mplus 2.13 (Muthen
& Muthen, 1998) and prospective associations (predicting
outcomes from previous day’s predictor scores) via multi-
level regression models in SAS PROC MIXED under SAS
9.1.2The latter analyses allowed simultaneous estimation of
within-person relations between variables measured daily,
their average magnitude (and sign) across all participants,
and the effects of several potential moderators. To account
for cyclical dependencies and serial autocorrelation, we in-
cluded dummy predictor variables coding for month and
day of week and an AR(1) parameter in the lagged models
(West & Hepworth, 1991). Two sets of prospective mod-
els were specified, with sexual behavior or physical con-
tact predicting next-day mood or stress, and with mood or
1Complete information can be obtained from the corresponding author.
2When outcomes were dichotomous, we used the GLIMMIX macro
(under PROC MIXED) specifying binomial-distributed errors and a
logit link. Models were first estimated with random Level 1 slopes,
and if random components were non-significant or unestimable, re-
estimated with slopes fixed.
stress predicting next-day physical contact or sexual behavior
(see Appendix for example equations). Level 1 (daily-level)
predictor variables were person mean-centered and Level 2
(person-level) variables were grand mean-centered. In gen-
eral, centering helps to minimize multicollinearity among
interaction terms and their constituent “main effect” terms
(Aiken & West, 1991), and person-mean centering of daily
predictors facilitates interpretation of Level 1 coefficients as
purely within-person effects. However, one consequence of
person-mean centering of the Level 1 predictors is the re-
moval of person-level variation in the predictors, thus leav-
ing person-level covariation between daily-level predictors
and the outcomes unmodeled (Hofmann & Gavin, 1998).
Accordingly, we added grand-mean centered person-level
means on daily-level predictor variables as person-level pre-
dictors in our models. We also included sexual orientation,
proportion of study days cohabiting with sexual partner, and
duration of current sexual relationship, along with their in-
teractions with the primary variables, as predictors in the ini-
tial prospective models. In keeping with recommendations
for model building in the absence of strong theory-based
predictions, non-significant interaction terms were trimmed
from subsequent models (Aiken & West, 1991). Significant
interactions were probed by examining simple slopes at se-
lected values of the moderator variables (i.e., M±1SD for
relationship duration; 0 and 1 for cohabitation and sexual
orientation), using methods described in Bauer and Curran
(2005). An alpha level of .05 was used for all statistical tests.
Results
Preliminary analyses, descriptive statistics,
and Pearson correlations
Preliminary analyses were conducted controlling for month,
day of week, and serial autocorrelation. Positive mood was a
strong negative same-day predictor of overall negative mood,
hostility, depressed mood, and anxiety. Therefore, we in-
cluded positive mood as a predictor in all lagged models
with these four predictors. Older age predicted fewer stres-
sors and lower levels of all of the negative affective states,
so age was included in all subsequent models.
Mean daily frequencies of physical intimacy, sexual be-
havior, and stress variables, mean daily averages for the mood
variables, percentage of participants with the minimum pos-
sible values for these variables (e.g., zero frequency of in-
tercourse for the 36-wk data collection period), and Pearson
correlations across participants are shown in Table 1.Only
two significant correlations were found between physical
contact and mood variables: physical affection was posi-
tively related to positive mood, and breast stimulation was
negatively related to hostility.
Springer
Arch Sex Behav (2007) 36:357–368 361
Table 1 Means, standard deviations, percentage of sample with minimum value, and Pearson correlations for overall average daily frequencies
of physical affection, sexual activity, and stress, and overall daily means of mood
Vari able MSD%min1 234567891011
1. Physical affectiona.43 .36 6.9 —
2. Breast stimulationa.15 .18 17.2 .54∗∗∗ —
3. Genital stimulationa.12 .14 12.1 .57∗∗∗ .83∗∗∗ —
4. Intercoursea,b.11 .12 15.9 .47∗∗ .68∗∗∗ .84∗∗∗ —
5. Orgasm with partnera.10 .11 17.2 .58∗∗∗ .78∗∗∗ .95∗∗∗ .80∗∗∗ —
6. Orgasm no partnera.06 .09 25.9 −.20 −.06 −.17 −.22 −.16 —
7. Positive moodc2.84 .74 0.0 .29∗.24 .22 .03 .21 −.00 —
8. Negative moodc1.43 .33 0.0 −.07 −.20 −.16 −.15 −.17 .12 −.14 —
9. Hostilityc1.33 .31 0.0 −.19 −.26∗−.22 −.21 −.19 .17 −.19 .89∗∗∗ —
10. Depressed affectc1.46 .34 1.8 .02 −.26 −.17 −.17 −.16 .21 −.26 .81∗∗∗ .65∗∗∗ —
11. Anxietyc1.61 .43 0.0 −.05 −.16 −.11 −.06 −.12 .05 −.13 .96∗∗∗ .79∗∗∗ .80∗∗∗ —
12. Stress occurrencea.16 .14 3.4 −.08 .01 .08 .10 .04 .07 .01 .35∗∗ .23 .29∗.39∗∗
Note.N=58 for variables 1–6 and 12, N=55 for variables 7–11.
aValues could range from 0.00 (never occurred) to 1.00 (occurred at least one time every day of measurement period).
bStatistics involving intercourse were estimated excluding participants who self-identified as lesbian (N=44 for variables 1–6, and 12, N=41
for variables 7–11).
cValues could range from 1.00 to 5.00.
∗p<.05 ∗∗p<.01 ∗∗∗p<.001.
Same-day associations between daily mood or stress
and daily physical contact or sexual behavior
Pooled within-person correlations between physical contact
or sexual behavior and mood or stress reported concurrently
are shown in Table 2. Breast stimulation, genital stimulation,
and orgasm with partner were positively related to positive
mood and negatively related to all negative moods and stress.
Physical affection and intercourse showed very similar pat-
terns, except physical affection was unrelated to stress and
intercourse was unrelated to depressed mood. Orgasm with-
out partner was significantly related only to stress, and the
relation was negative.
Predicting next-day mood and stress from daily physical
contact and sexual behavior
Figure 1illustrates significant first order effects and their un-
standardized parameter estimates (γ) from the prospective
multilevel regression analyses.3In general, physical affec-
tion and sexual behavior were consistently associated with
reduced negative states, fewer stressors, and/or increased
positive mood the following day.
Breast stimulation, genital stimulation, and orgasm with
a partner predicted higher next-day positive mood. However,
there were significant interactions with sexual orientation
for the effects of genital stimulation (γ=.08, p<.01) and
orgasm with a partner (γ=.09, p<.01) on positive mood.
3Complete results, including parameter estimates and standard errors
for all models, can be obtained from the corresponding author.
Probing of these interactions revealed that the relations be-
tween the sex variables and positive mood were significant
for the lesbian participants (ωs=.10, .10, both ps<.001,
where ωrepresents the parameter estimate for the simple
slope), but not for the heterosexual participants (ωs=.01,
.01, both ns).
Physical affection, breast stimulation, genital stimulation,
orgasm with a partner, and intercourse were all significant
predictors of lower overall negative mood and lower anxiety
on the following day. Daily physical affection, genital stimu-
lation, and partnered orgasm predicted lower next-day stress.
However, significant interactions with relationship duration
qualified the effects of genital stimulation (γ=.02, p<.01)
and orgasm with a partner (γ=.02, p<.01) on stress. In-
teraction probes revealed that both of these negative effects
were significant for women with relatively short-duration
relationships (respectively, ωsat2.5yrs=−.52, and −.46,
both ps<.01, ωsat12yrs=−.29, and −.23, both ps<.01),
but not for those with relatively long-duration relationships
(ωs at 21.5 yrs =−.06 and .01, both ns).
All of the significant associations involved behavior with
other individuals and improvements in mood except that
orgasm without partner predicted higher next-day negative
mood and anxiety. Next-day hostility and depressed mood
were associated with no predictors. Significant random ef-
fects noted in Fig. 1indicate instances where there was sig-
nificant variation between participants in the nature of the
relation between predictor and outcome variable. For ex-
ample, for genital stimulation predicting next-day negative
mood, the slopes ranged from −0.11 to 0.07.
Springer
362 Arch Sex Behav (2007) 36:357–368
Table 2 Pooled within-person correlations between physical affection or sexual behavior and same-day mood or stress
Positive mood Negative mood Hostility Depressed mood Anxiety Stress occurrence
Physical affection .09∗∗∗ −.04∗∗∗ −.04∗∗∗ −.03∗∗ −.05∗∗∗ −.01
Breast stimulation .08∗∗∗ −.04∗∗∗ −.03∗∗ −.02∗−.05∗∗∗ −.02∗
Genital stimulation .09∗∗∗ −.04∗∗∗ −.04∗∗∗ −.03∗∗ −.05∗∗∗ −.03∗∗∗
Intercoursea.07∗∗∗ −.03∗∗ −.04∗∗∗ −.01 −.03∗∗ −.02∗
Orgasm with partner .09∗∗∗ −.04∗∗∗ −.04∗∗∗ −.03∗∗ −.04∗∗∗ −.04∗∗∗
Orgasm no partner −.02 −.01 −.01 −.02 −.00 −.03∗∗∗
Note.Ns (number of person-days) for these analyses ranged from 13,011 to 13,919.
aCorrelations involving intercourse were estimated without participants who self-identified as lesbian.
∗p<.05; ∗∗p<.01; ∗∗∗p<.001.
Predicting next-day physical affection and sexual
behavior from daily mood and stress
As shown in Fig. 1, positive mood was the only significant
first order predictor of next-day physical affection or sexual
behavior. Regardless of which negative mood variable was
in the model, higher positive mood predicted more next-day
physical affection, breast stimulation, genital stimulation,
and orgasms with a partner and fewer orgasms without a
partner. However, for the physical affection outcome, there
were significant interactions between positive mood and sex-
ual orientation in all of these models (γs=.45 to .46, all
ps<.0001). Probing of these interactions revealed that all
of them were of the same form, such that the positive relation-
ship between positive mood and next-day physical affection
was significant for the lesbian participants (ωs=.46 to .50,
all ps<.0001), but not for the heterosexual participants
(ωs=.01 to 05, all ns). Interactions between positive mood
and sexual orientation did not approach significance for any
other sexual behavior outcome.
No negative mood variable or stress was significantly
related across all participants to any next-day physical
Fig. 1 Summary of significant conditional (main) effects and unstan-
dardized parameter estimates from prospective analyses. Coefficient
for positive mood predicting next-day affection or sex is average of co-
efficients for four separate models, each including positive mood along
with one of the negative mood indicators. Subscript r indicates random
effect. Subscript a indicates simple slope significant only for lesbian
participants; subscript b indicates effect stronger for shorter duration
relationships. See text for description. ∗p<.05, ∗∗p<.01, ∗∗∗p<.001
Springer
Arch Sex Behav (2007) 36:357–368 363
contact or sexual behavior variable. However, there were
three interactions involving negative mood variables. Sexual
orientation interacted significantly with anxiety (γ=−.29,
p<.05), such that higher anxiety predicted fewer orgasms
without a partner for heterosexual (ω=−.13, p<.05),
but not lesbian (ω=.15, ns), women. It also interacted
significantly with stressor occurrence (γ=−.53, p<.01)
such that higher stress predicted less breast stimulation in
lesbian (ω=−.50, p<.01), but not heterosexual (ω=.03,
ns), women. Finally, there was a significant interaction
between depressed mood and cohabitation (γ=.45,
p<.05), such that higher depressed mood predicted more
next-day intercourse for participants who lived with their
partners (ω=.13, p<.05), but not for those who did not
live with their partners (ω=−.33, ns).
Straightforward effect size measures paralleling those for
traditional “single-level” multiple regression models have
yet to be developed for multilevel regression approaches.
However, based on the rather small same-day within-person
correlations reported here, we can assume that the prospec-
tive effects were similarly small.
First order effects of moderators4
Models with mood or stress outcomes
In all models with affection or partnered sex as predictors,
lesbian participants reported less hostility than heterosexual
participants (γs=−.22 to −.25, all ps<.05). In addition,
they reported less overall negative mood in the breast stimu-
lation and genital stimulation models (γs=−.28 and −.24,
both ps<.05). Somewhat surprisingly, living with one’s
sexual partner predicted higher average levels of negative
affective states. In models with physical affection or genital
stimulation as predictors, cohabitation predicted higher
overall negative mood (γs=.43 and .30, both ps<.05);
and in models with breast stimulation, genital stimulation,
partnered orgasm, orgasm without a partner, or intercourse,
it predicted higher depressed mood (γs=.29 to .56, all ps<
.05). In general, however, as the duration of the relationship
increased, average levels of negative affect decreased. In
models with physical affection, breast stimulation, genital
4In addition to the effects that were the focus of the study, the lagged
models included grand mean-centered person means of the daily pre-
dictors. Inclusion of these variables was necessary in order to ensure
the correct apportionment of variance in outcomes and thus allow the
correct interpretation of parameter estimates associated with the vari-
ables of primary interest. Some of these between-subject variables were
statistically significant predictors. However, given that the current study
focuses on the relations between daily physical contact and daily mood
or stress, and that the sample size was 58 or fewer for these between-
subject effects, we elected not to present them in this report. Complete
information about these findings can be obtained from the authors.
stimulation, or intercourse as predictors, longer duration
of the relationship predicted lower overall negative mood
(γs=−.01 to −.02, all ps<.05) and in the model with
physical affection, it predicted lower hostility (γs=−.01,
p<.05). Older age predicted higher positive mood (γs=.26
to .37, all ps<.0001), lower anxiety (γs=−.05 to -.06,
all ps<.05), and fewer stressor occurrences in all models
(γs=−.25 to −.33, all ps<.001).
Models with affection or sexual behavior outcomes
In all models with positive mood and one of the negative
affect variables as predictors, living with one’s sexual
partner predicted more physical affection (γs=3.02 to 3.12,
all ps<.001) and fewer unpartnered orgasms (γs=−1.96
to −2.21, all ps<.01). In models including positive mood
with hostility or anxiety, or stressor occurrence, it predicted
more partnered orgasms (γs=1.43 to 1.72, all ps<.05).
Older age was associated with lower average frequency of
breast stimulation (γs=−.23 to −.35, all ps<.05) and
genital stimulation (γs=−.19 to −.34, all ps<.05) in all
models. Neither sexual orientation nor relationship duration
was associated with average frequency of physical affection
or any of the sexual behaviors.
Discussion
In the current study, mid-aged women provided detailed daily
records of sexual behavior, physical affection, mood, and
stress for an average duration of 35 weeks. To our knowl-
edge, this study is the first to evaluate relations between
physical contact or sexual behavior and mood or stress in
adults on a prospective daily basis. Two advantages of our
study over most previous work are enhanced ecological va-
lidity compared to laboratory studies and collection of daily
rather than long-term retrospective reports. This design al-
lowed us to examine the extent to which occasions of relative
increases or decreases in sexual behavior or physical affec-
tion were temporally associated with occasions of relatively
more intense mood or stress within individual participants.
We also evaluated potential causal effects by testing lagged
relations between variables. The results were consistent with
bidirectional causal relations in which dyadic sexual inter-
action and physical affection improved mood and reduced
stress, which in turn increased the likelihood of future sex
and physical intimacy. Sexual orientation, living with one’s
partner, and relationship duration moderated some of these
lagged daily effects.
Same-day relationships
With few exceptions, physical affection or sexual activ-
ity variables were positively related to same-day positive
Springer
364 Arch Sex Behav (2007) 36:357–368
mood and negatively related to same-day negative moods
and stress. These results are similar to the findings of Forten-
berry and colleagues (Fortenberry et al., 2005), in which
univariate analyses showed negative and positive mood were
negatively and positively related, respectively, to same-day
intercourse in adolescent women. Given that sex is an ef-
fective reinforcer in many species and can elicit positive
affective states in male and female rodents that persist af-
ter sexual behavior ends (Agmo & Berenfeld, 1990; Paredes
& Vazquez, 1999), perhaps these results are not surprising.
However, it is important to note that, whereas orgasm with
a partner was significantly related to all mood and stress
variables in the current study, orgasm without a partner was
related only to stress. For our participants, all orgasms were
not equal, at least in terms of their associations with mood
states.
Because the variables were assessed concurrently, their
temporal relations to one another could not be established
and no causal pathway could be eliminated by these analyses.
We note, however, that surveys suggest most sexual interac-
tion between individuals occurs after retiring to bed and be-
fore arising (Dutton, 2003), whereas moods are experienced
throughout waking hours. With this in mind, it is arguably
more likely that these within-subject correlations reflect a
mood effect on same-day sexual behavior than vice-versa. If
true, this suggests positive moods may have promoted sexual
expression in our participants, whereas negative moods may
have discouraged it.
Does sex or physical affection influence mood
or experience of stress?
Affectionate physical contact or any of several forms of sex-
ual activity with another person (e.g., genital stimulation)
on a given day predicted reports of lower levels of negative
affective states and fewer stress reports the following day,
particularly for participants whose relationships were rela-
tively newer. Further, genital stimulation or orgasm with a
partner on a given day predicted more positive mood the
following day for lesbian participants. These prospective as-
sociations are consistent with the idea that physical affection
and partnered sexual activity may have improved partici-
pants’ mood and reduced perceived stress. Previous studies
have reported that anxiety or stress might serve to motivate
sexual interactions as a form of mood repair for some individ-
uals (Bancroft et al., 2003; Graham, Sanders, Milhausen, &
McBride, 2004). Although we have little evidence to support
this idea in our sample (the only predictive effects of stress
or anxiety on sexual behavior were negative and differed
according to sexual orientation), such interactions might ef-
fectively improve mood on a continuing basis, especially in
the earlier stages of a relationship.
In contrast, we found that orgasm without a partner was
actually associated with higher anxiety and negative mood
the following day. Thus, for our mid-aged female partici-
pants, social aspects of a dyadic relationship were important
for any mood-elevating effects of sexual activity; sensory as-
pects alone (or physiological changes accompanying them)
were insufficient. We speculate that unpartnered sexual re-
lations may have served to highlight what was missing in
our participants’ social lives, which in turn could have led
to lower mood on the following day. Alternatively, because
self-stimulation was viewed as socially unacceptable during
their formative years, some of our participants may have had
negative feelings associated with it.
Beneficial effects of social connectedness on stress
management and health are noted in many studies (House,
Landis, & Umberson, 1988), but the pathways through
which they occur are not fully explained (House, 2001).
Our results suggest they are mediated, at least partially,
through physical affection or sexual interactions (Carter,
1998; Uvnas-Moberg, 1997,1998). Those with effective
social support networks may experience more pleasing
physical contact in daily life, or a loving social relation-
ship may bring about more, or at least more satisfying,
sexual activity. Both could enhance mood and lower
stress.
If affection and sex do enhance mood and reduce stress,
what physiological mechanisms might mediate these effects?
Giving or receiving positive physical contact (e.g., touch,
massage) can influence cardiovascular and neuroendocrine
functioning to counteract some effects of stress (Moyer,
Rounds, & Hannum, 2004; Uvnas-Moberg, 1997). For ex-
ample, massage-like touch causes release of oxytocin in rats
(Lund et al., 2005) and sexual arousal and orgasm cause
enhanced release of oxytocin in human males and females
(Carmichael et al., 1987). Oxytocin is a neuropeptide that
promotes attachment, prosocial, and reproductive behav-
iors in many mammalian species, possibly including hu-
mans (Carter, 1998). Intranasally administered exogenous
oxytocin can reduce men’s anxiety and cortisol responses
to stress (Heinrichs et al., 2003), likely through its regula-
tory effects on the hypothalamic-pituitary-adrenal axis (Neu-
mann, Wigger, Torner, Holsboer, & Landgraf, 2000). In ad-
dition, greater oxytocin responses to laboratory stress were
associated with lower blood pressure (BP) before and after
the stressor and faster diastolic BP recovery (Light et al.,
2000). In rats, repeated oxytocin administration causes long-
term decreases in BP, suggesting potential oxytocin-related
stress-reducing effects of satisfying sexual activity could be
relatively long-lasting (Uvnas-Moberg, 1997).
Both lagged and concurrent relations reported here were
consistent with previous research suggesting that sexual
arousal leads to higher positive or lower negative mood
Springer
Arch Sex Behav (2007) 36:357–368 365
(Meisler & Carey, 1991) on a short-term basis. Indeed, our
results go further and indicate that mood-enhancing effects
of sexual behavior and physical affection may remain active
for at least an entire day and night.
Does mood or experience of stress influence occurrence
of sex or physical affection?
Increased positive mood predicted more next-day physical
affection for the lesbian participants. It also predicted higher
levels across participants of all partnered sexual behaviors
except for intercourse, but predicted fewer unpartnered or-
gasms. These results lend causal weight to prior reports
that women are more likely to express themselves sexu-
ally when experiencing a positive affective state (Sanders,
Warner, Backstrom, & Bancroft, 1983; Warner & Bancroft,
1988). It is important to note they do not mean that happier
individuals had more affection or sex. The results indicated
instead that when positive mood was higher than usual, par-
ticipants were more likely to be affectionate or engage in
partnered sexual activity (except intercourse) the next day.
Because of the strong interaction between sexual orientation
and positive mood in predicting next-day affection, along
with the absence of a significant relationship between posi-
tive mood and intercourse (which was estimated excluding
lesbian participants), we were concerned that all of the re-
lationships between positive mood and next-day sexual be-
havior might be due primarily to effects for the lesbian but
not the heterosexual participants. However, none of the other
orientation by mood interactions was significant. Thus, the
strength of the relations between positive mood and next-day
sexual behaviors did not differ by sexual orientation.
As noted above, intercourse was the only sexual activity
that was unrelated to women’s prior-day positive mood in the
current study. By definition, sexual intercourse occurs only
when a male partner maintains an erection long enough for
vaginal penetration, whereas the other partnered sexual activ-
ities we monitored do not have this requirement. Because of
this difference, we speculate that occurrence of intercourse
was more heavily influenced by men’s mood state, rather
than women’s.
In the current study, daily negative affect and stress gen-
erally were not related to affection or sexual behavior on the
following day. Thus, if negative moods or stress influenced
the occurrence of affectionate or sexual behaviors, most of
the effects did not carry over to the next day. One exception
to this involved the only significant lagged daily relationship
found for depressed mood as either predictor or outcome, and
the only case in which women’s mood predicted next-day in-
tercourse. Higher depressed mood predicted more next-day
intercourse for women who lived with their partners. Given
that induced depressed mood can reduce both subjective and
genital arousal (Meisler & Carey, 1991; Mitchell et al., 1998),
this result was somewhat unexpected, but may represent a
previously unreported form of mood-repair.
In previous research, the best-established causal effect of
mood on sexual arousal is the ability of anxiety to enhance
it. Anxiety activates the sympathetic nervous system, and
experiments show that non-sex-related anxiety (Hoon
et al., 1977; Palace & Gorzalka, 1990), sex-related anxiety
(Barlow et al., 1983; Bozman & Beck, 1991), and even
sympathetic activation from exercise (Meston & Gorzalka,
1996) can all enhance sexual arousal. However, all of
these effects occurred within a very short time frame—any
influence of mood induction or sympathetic activation was
detected within the laboratory session. In the current study,
we measured mood and sexual behavior only once daily,
and did not assess arousal per se; thus, it is difficult to make
direct comparisons with our findings. However, our design
could potentially detect whether short-term mood effects
accumulated over time to influence actual occurrence of
sexual activity. For example, if anxiety intensified arousal
within each sexual episode, it might become associated with
enhanced pleasure and hence increase the likelihood of sex
occurring. Our results did not support that possibility. The
only lagged effect of higher anxiety was a negative effect
on orgasm without a partner, and the effect was limited to
heterosexual women. In our same-day analyses, we found all
negative mood indicators and stress, including anxiety, were
negatively related to affection and partnered sex variables.
Perhaps anxiety interfered with sexual desire before the
arousal stage, as reported by Bozman and Beck (1991).
Experimental studies of adults also suggest that anger and
depressed mood are likely to interfere with sexual arousal
(Bozman & Beck, 1991; Mitchell et al., 1998), but positive
mood is more likely to enhance it (Mitchell et al., 1998).
Our same-day results were consistent with these findings,
as were most of the lagged analyses showing positive mood
predicted more sexual behavior on the following day. That
is, if positive mood enhanced arousal in our participants, it
may have also enhanced sexual motivation, leading to more
sexual interactions.
As found in previous research (Burleson et al., 2002), part-
ner availability was positively related to frequency of part-
nered sexual behavior and negatively related to frequency
of unpartnered sexual behavior. However, sexual orientation
was unrelated to frequency of either partnered or unpart-
nered sexual behavior. Average age in the current study was
14 years older than in the previous study, which may help to
account for this difference in findings. In the current study,
older age was associated with less frequent breast and genital
stimulation, higher positive mood and lower anxiety ratings,
and fewer stressor occurrences. These findings fit well with
previous cross-sectional results showing less frequent sexual
activity (Leigh, Temple, & Trocki, 1993) and higher levels
of positive affect correlated with increasing age (Carstensen,
Springer
366 Arch Sex Behav (2007) 36:357–368
Pasupathi, Mayr, & Nesselroade, 2000; Mroczek & Kolarz,
1998). Interestingly, living with one’s sexual partner pre-
dicted higher overall levels of negative affective states in the
current study. However, as the duration of the relationship
increased (controlling for age), average levels of negative
affect declined. Finally, the lesbian participants reported less
hostility and depression than the heterosexual participants.
Limitations and implications for future research
The current study has several limitations. Perhaps the most
consequential is the nature of our sample, which comprised
a relatively small group of mid-aged women with a limited
age range. As in many longitudinal daily studies, our par-
ticipants were very highly motivated. They were also more
educated and wealthier than average, and comprised a higher
than average proportion of lesbian women. Therefore, fur-
ther study is needed to determine whether the results we
found are borne out in the general population. Although we
report differences between lesbian and heterosexual women,
there were relatively few lesbian participants, so these find-
ings should be interpreted cautiously. In addition, although
many of the effects were highly significant, they were small.
The small effect sizes are not surprising, given that the data
were collected in a natural setting, within the context of mul-
tiple social and environmental influences, rather than in a
laboratory.
Measurement issues also deserve mention. Because the
study continued for several years, we reduced measurement
burden by using a short daily instrument with dichotomous
response formats where possible. Therefore, our sexual ac-
tivity and affection measures included no ratings of desire,
arousal, satisfaction, or other qualitative aspects. Nor did
we ask who initiated the encounter or gather data about any
interpersonal activities not listed on the daily datasheet. Sim-
ilarly, our stress measure asked only if a stressor occurred.
We did not obtain intensity ratings or inquire about content
or type of stressor. Approximately one third of the sample
had entered the menopause transition. Because we did not
measure hormones, any potential effects of changes in hor-
monal status were not detected. All of this information would
be useful to better understand relations among the variables
we measured.
To facilitate regular data collection, our participants
completed daily recordings immediately upon arising,
before the day’s activities had been initiated and distractions
could occur. Therefore, a sleep period occurred between
data recording and most moods, stressors, or incidences of
sexual activity or affection that were reported. Processes
that occur during sleep may influence memory (Benington
& Frank, 2003). In fact, Fahrenberg and colleagues (Fahren-
berg, Bruegner, Foerster, & Kaeppler, 1999) reported a bias
towards reporting more negative affect when moods through-
out a day were summarized in the evening or the following
morning, but the summary followed the same pattern of rela-
tive mood levels as the target day’s multiple ratings. Thus, as
relative ranks are more important than absolute levels in our
analyses, this potential bias may be relatively unproblematic.
With notable exceptions (Exton et al., 2001), much of
what is known about neuroendocrine changes during and
after sexual activity has been learned studying arousal and
orgasm in the laboratory, with no partner present. In all our
analyses, we found differences between effects or correlates
of partnered versus non-partnered sexual activities. If these
effects are mediated, even partially, by biochemical changes,
it follows that self-stimulated arousal and orgasms may be
physiologically different from those occurring with a partner.
This emphasizes the importance of studying sexuality in
a more realistic social context whenever possible (Senn &
Desmarais, 2004).
Future longitudinal field studies should make more fre-
quent assessments. This would improve the accuracy of the
data, but more importantly, allow stronger causal tests. For
example, Bancroft et al. (2003) reported that many men use
sex (either alone or with a partner) to improve their mood
or reduce stress. If true, this suggests that for some indi-
viduals, bad mood or stress causes sexual activity to occur,
which in turn causes improved mood or reduced stress. To
test this pathway, assessments of mood both before and after
sexual activity are needed, which requires, at minimum, two
measurements per day.
Despite its limitations, our study is the first to report
prospective daily associations between sexual behavior or
physical affection and affective state or stress in an adult
sample. Our results suggest the conventional wisdom may
be true. For our sample of mid-aged women, physical affec-
tion or sexual activity with another person were related to
improved mood, for at least one day. Unpartnered sexual ac-
tivity, however, did not have the same effect, highlighting the
importance of social relationships in the association between
sex and mood. In addition, positive mood was related to in-
creased likelihood of next-day sex or affection. These results
suggest that the associations between physical contact and
mood are driven by complex and intertwined processes that
unfold over time.
Acknowledgments This project was supported by an internal grant
from the College of Arts and Sciences at New Mexico State University.
Data analysis and manuscript preparation were additionally supported
by the National Science Foundation (BCS 0129922) and the New Col-
lege of Interdisciplinary Arts and Sciences at Arizona State University.
References
Agmo, A. (1999). Sexual motivation–an inquiry into events determining
the occurrence of sexual behavior. Behavioural Brain Research,
105, 129–150.
Springer
Arch Sex Behav (2007) 36:357–368 367
Agmo, A., & Berenfeld, R. (1990). Reinforcing properties of ejacula-
tion in the male rat: Role of opioids and dopamine. Behavioral
Neuroscience,104, 177–182.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
interpreting interactions. Newbury Park, CA: Sage Publications.
Apt, C., Hurlbert, D. F., Pierce, A. P., & White, L. C. (1996). Rela-
tionship satisfaction, sexual characteristics and the psychosocial
well-being of women. Canadian Journal of Human Sexuality,5,
195–210.
Avis, N. E., Stellato, R., Crawford, S., Johannes, C., & Longcope,
C. (2000). Is there an association between menopause status and
sexual functioning? Menopause,7, 297–309.
Bancroft, J., Janssen, E., Strong, D., Carnes, L., Vukadinovic, Z., &
Long, J. S. (2003). The relation between mood and sexuality in
heterosexual men. Archives of Sexual Behavior,32, 217–230.
Bancroft, J., Janssen, E., Strong, D., & Vukadinovic, Z. (2003). The
relation between mood and sexuality in gay men. Archives of
Sexual Behavior,32, 231–242.
Barlow, D. H., Sakheim, D. K., & Beck, J. G. (1983). Anxiety increases
sexual arousal. Journal of Abnormal Psychology,92, 49–54.
Bauer, D. J., & Curran, P. J. (2005). Probing interactions in fixed and
multilevel regression: Inferential and graphical techniques. Multi-
variate Behavioral Research,40, 373–400.
Beck, J. G., & Bozman, A. W. (1995). Gender differences in sexual de-
sire: The effects of anger and anxiety. Archives of Sexual Behavior,
24, 595–612.
Benington, J. H., & Frank, M. G. (2003). Cellular and molecular con-
nections between sleep and synaptic plasticity. Progress in Neu-
robiology,69, 71–101.
Biernbaum, M. A., & Ruscio, M. (2004). Differences between matched
heterosexual and non-heterosexual college students on defense
mechanisms and psychopathological symptoms. Journal of Ho-
mosexuality,48, 125–141.
Blaicher, W., Gruber, D., Bieglmayer, C., Blaicher, A. M., Knogler,
W., & Huber, J. C. (1999). The role of oxytocin in relation to
female sexual arousal. Gynecologic and Obstetric Investigation,
47, 125–126.
Blanchflower, D. G., & Oswald, A. J. (2004). Money, sex and happi-
ness: An empirical study. National Bureau of Economic Research
Working Paper Series (Working paper no. 10499).
Bodinger, L., Hermesh, H., Aizenberg, D., Valevski, A., Marom, S.,
Shiloh, R., et al. (2002). Sexual function and behavior in social
phobia. Journal of Clinical Psychiatry,63, 874–879.
Bozman, A. W., & Beck, J. G. (1991). Covariation of sexual desire
and sexual arousal: The effects of anger and anxiety. Archives of
Sexual Behavior,20, 47–60.
Burleson, M. H., Trevathan, W. R., & Gregory, W. L. (2002). Sexual
behavior in lesbian and heterosexual women: Effects of menstrual
cycle phase and partner availability. Psychoneuroendocrinology,
27, 489–503.
Call, V., Sprecher, S., & Schwartz, P. (1995). The incidence and fre-
quency of marital sex in a national sample. Journal of Marriage
and the Family,57, 639–652.
Carmichael, M. S., Humbert, R., Dixen, J., Palmisano, G., Greenleaf,
W., & Davidson, J. M. (1987). Plasma oxytocin increases in the
human sexual response. Journal of Clinical Endocrinology and
Metabolism,64, 27–31.
Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R. (2000).
Emotional experience in everyday life across the adult life span.
Journal of Personality and Social Psychology,79, 644–655.
Carter, C. S. (1998). Neuroendocrine perspectives on social attachment
and love. Psychoneuroendocrinology,23, 779–818.
Cyranowski, J. M., Bromberger, J., Youk, A., Matthews, K., Kravitz,
H. M., & Powell, L. H. (2004). Lifetime depression history and
sexual function in women at midlife. Archives of Sexual Behavior,
33, 539–548.
Dennerstein, L., Dudley, E., & Burger, H. (2001). Are changes in sexual
functioning during midlife due to aging or menopause? Fertility
and Sterility,76, 456–460.
Dennerstein, L., Randolph, J., Taffe, J., Dudley, E.,& B urger, H. (2002).
Hormones, mood, sexuality, and the menopausal transition. Fe r-
tility and Sterility,77, S42–S50.
Dutton, J. (2003). Redbook’s steamiest sex survey ever! Redbook,200,
192–193, 209–210.
Exton, M. S., Krueger, T. H. C., Koch, M., Paulson, E., Knapp, W.,
Hartmann, U., et al. (2001). Coitus-induced orgasm stimulates
prolactin secretion in healthy subjects. Psychoneuroendocrinol-
ogy,26, 287–294.
Fahrenberg, J., Bruegner,G., Foerster, F., & Kaeppler, C. (1999). Ambu-
latory assessment of diurnal changes with a hand-held computer:
Mood, attention and morningness-eveningness. Personality and
Individual Differences,26, 641–656.
Fortenberry, J. D., Temkit, M. H., Tu, W., Graham, C. A., Katz, B. P.,
& Orr, D. P. (2005). Daily mood, partner support, sexual interest,
and sexual activity among adolescent women. Health Psychology,
24, 252–257.
Graham, C. A., Sanders, S. A., Milhausen, R. R., & McBride, K. R.
(2004). Turning on and turning off: A focus group study of the
factors that affect women’s sexual arousal. Archives of Sexual
Behavior,33, 527–538.
Hackbert, L., & Heiman, J. R. (2002). Acute dehydroepiandrosterone
(dhea) effects on sexual arousal in postmenopausal women. Jour-
nal of Women’s Health and Gender Based Medicine,11, 155–
162.
Heiman, J. R. (1980). Female sexual response patterns: Interactions of
physiological, affective, and contextual cues. Archives of General
Psychiatry,37, 1311–1316.
Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003).
Social support and oxytocin interact to suppress cortisol and sub-
jective responses to psychosocial stress. Biological Psychiatry,54,
1389–1398.
Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in hier-
archical linear models: Implications for research in organizations.
Journal of Management,24, 623–641.
Hoon, P., Wincze, J. P., & Hoon, E. (1977). A test of reciprocal in-
hibition: Are anxiety and sexual arousal in women mutually in-
hibitory? Journal of Abnormal Psychology,86, 65–74.
House, J. S. (2001). Social isolation kills, but how and why? Psycho-
somatic Medicine,63, 273–274.
House, J. S., Landis, K. R., & Umberson, D. (1988). Social relationships
and health. Science,241, 540–545.
Kennedy, S. H., Dickens, S. E., Eisfeld, B. S., & Bagby,
R. M. (1999). Sexual dysfunction before antidepressant therapy
in major depression. Journal of Affective Disorders,56, 201–
208.
Kurdek, L. A. (1991). The relations between reported well-being and
divorce history, availability of a proximate adult, and gender. Jour-
nal of Marriage and the Family,53, 71–78.
Laan, E., Everaerd, W., van Bellen, G., & Hanewald, G. (1994).
Women’s sexual and emotional responses to male- and female-
produced erotica. Archives of Sexual Behavior,23, 153–169.
Laan, E., Everaerd, W., van Berlo, R., & Rijs, L. (1995). Mood and
sexual arousal in women. Behaviour Research and Therapy,33,
441–443.
Leigh, B. C., Temple, M. T., & Trocki, K. F. (1993). The sexual behavior
of U.S. adults: Results from a national survey. American Journal
of Public Health,83, 1400–1408.
Light, K. C., Smith, T. E., Johns, J. M., Brownley, K. A., Hofheimer,
J. A., & Amico, J. A. (2000). Oxytocin responsivity in mothers of
infants: A preliminary study of relationships with blood pressure
during laboratory stress and normal ambulatory activity. Health
Psychology,19, 560–567.
Springer
368 Arch Sex Behav (2007) 36:357–368
Lund, I., Ge, Y., Yu, L. C., Uvnas-Moberg, K., Wang, J., Yu, C., et al.
(2005). Repeated massage-like stimulation induces long-term ef-
fects on nociception: Contribution of oxytocinergic mechanisms.
European Journal of Neuroscience,22, 1553–1554.
Meisler, A. W., & Carey, M. P. (1991). Depressed affect and male sexual
arousal. Archives of Sexual Behavior,20, 541–554.
Meston, C. M., & Gorzalka, B. B. (1996). Differential effects of sym-
pathetic activation on sexual arousal in sexually dysfunctional and
functional women. Journal of Abnormal Psychology,105, 582–
591.
Mitchell, W. B., DiBartolo, P. M., Brown, T. A., & Barlow, D. H.
(1998). Effects of positive and negative mood on sexual arousal in
sexually functional males. Archives of Sexual Behavior,27, 197–
207.
Moyer, C. A., Rounds, J., & Hannum, J. W. (2004). A meta-analysis of
massage therapy research. Psychological Bulletin,130, 3–18.
Mroczek, D. K., & Kolarz, C. M. (1998). The effect of age on posi-
tive and negative affect: A developmental perspective on happi-
ness. Journal of Personality and Social Psychology,75, 1333–
1349.
Muthen, L. K., & Muthen, B. O. (1998). Mplus user’s guide.Los
Angeles, CA: Muthen & Muthen.
Neumann, I., Wigger, A., Torner, L., Holsboer, F., & Landgraf, R.
(2000). Brain oxytocin inhibits basal and stress-induced activity
of the hypothalamo-pituitary-adrenal axis in male and female rats:
Partial action within the paraventricular nucleus. Journal of Neu-
roendocrinology,12, 235–243.
Palace, E. M., & Gorzalka, B. B. (1990). The enhancing effects of anx-
iety on arousal in sexually dysfunctional and functional women.
Journal of Abnormal Psychology,99, 403–411.
Paredes, R. G., & Vazquez, B. (1999). What do female rats like
about sex? Paced mating. Behavioural Brain Research,105, 117–
127.
Sanders, D., Warner, P., Backstrom, T., & Bancroft, J. (1983). Mood,
sexuality, hormones and the menstrual cycle: I. Changes in mood
and physical state: Description of subjects and method. Psychoso-
matic Medicine,45, 487–501.
Senn, C. Y., & Desmarais, S. (2004). Impact of interaction with a part-
ner or friend on the exposure effects of pornography and erotica.
Violence and Victims,19, 645–658.
Suh, E., Diener, E., Oishi, S., & Triandis, H. C. (1998). The shift-
ing basis of life satisfaction judgments across cultures: Emotions
versus norms. Journal of Personality and Social Psychology,74,
484–493.
Thompson, L. Y., Snyder, C. R., Hoffman, L., Michael, S. T.,
Rasmussen, H. N., Billings, L. S., et al. (2005). Dispositional
forgiveness of self, others, and situations. Journal of Personality,
73, 313–359.
Uvnas-Moberg, K. (1997). Oxytocin linked antistress effects–the relax-
ation and growth response. Acta Physiologica Scandinavica 161
(Suppl. 640), 38–42.
Uvnas-Moberg, K. (1998). Oxytocin may mediate the benefits of pos-
itive social interaction and emotions. Psychoneuroendocrinology,
23, 819–835.
Warner, P., & Bancroft, J. (1988). Mood, sexuality, oral contraceptives
and the menstrual cycle. Journal of Psychosomatic Research, 32,
417–427.
West, S. G., & Hepworth, J. T. (1991). Statistical issues in the study
of temporal data: Daily experiences. Journal of Personality, 59,
609–662.
Wolchik, S. A., Beggs, V. E., Wincze, J. P., Sakheim, D. K., Barlow,
D. H., & Mavissakalian, M. (1980). The effect of emotional arousal
on subsequent sexual arousal in men. Journal of Abnormal Psy-
chology, 89, 595–598.
Wolpe, J. (1958). Psychotherapy by reciprocal inhibition. Stanford,
CA: Stanford University Press.
Springer
