Emotion
2011, Vol. 11, No. 6, 1368 –1378

© 2011 American Psychological Association
1528-3542/11/$12.00 DOI: 10.1037/a0024278

Feeling Good:
Autonomic Nervous System Responding in Five Positive Emotions
Michelle N. Shiota, Samantha L. Neufeld, Wan H. Yeung, Stephanie E. Moser, and Elaine F. Perea
Arizona State University
Although dozens of studies have examined the autonomic nervous system (ANS) aspects of negative
emotions, less is known about ANS responding in positive emotion. An evolutionary framework was
used to define five positive emotions in terms of fitness-enhancing function, and to guide hypotheses
regarding autonomic responding. In a repeated measures design, participants viewed sets of visual images
eliciting these positive emotions (anticipatory enthusiasm, attachment love, nurturant love, amusement,
and awe) plus an emotionally neutral state. Peripheral measures of sympathetic and vagal parasympathetic activation were assessed. Results indicated that the emotion conditions were characterized by
qualitatively distinct profiles of autonomic activation, suggesting the existence of multiple, physiologically distinct positive emotions.
Keywords: positive emotion, evolutionary psychology, autonomic nervous system, psychophysiology

emotions, and the findings of individual studies have been mixed,
major reviews of this literature typically point to some degree of
specificity (e.g., Cacioppo et al., 2000; Friedman, 2010; Kreibig,
2010), especially when subtypes of major emotion categories are
considered (Kreibig, 2010).
Much less is known about autonomic responding in the positive
emotions. This lack of attention is consistent with the historical
underrepresentation of positive emotions in psychological research, and with the still-common perception among theorists that
positive emotions have fewer implications for evolutionary fitness,
are less differentiated, and have less distinct impact on motivation
and behavior than is true of the negative emotions (noted by
Fredrickson, 1998, 2001). The present study aims to help correct
this imbalance by examining and comparing the autonomic nervous system aspects of five positive emotions.

From William James (1884, p. 190), who proposed that “bodily
changes follow directly the perception of the exciting fact, and our
feeling of the same changes as they occur IS the emotion,” to
Antonio Damasio (1999, p. 51), who wrote that “emotions use the
body as their theater,” many theorists have commented on the role
of physiological changes controlled by the autonomic nervous
system (ANS) in emotional experience. The autonomic aspects of
emotion are thought to serve important adaptive functions, preparing the body for an appropriate behavioral response to the eliciting
situation (Cannon, 1915; Frijda, 1986; Lang, 1985; Tooby &
Cosmides, 2008). Many negative emotions involve increased activation of the sympathetic branch of the ANS (see Cacioppo,
Berntson, Larsen, Poehlmann, & Ito, 2000 for a review), which
generally facilitates physical exertion needed for “fight or flight.”
To the extent that different emotion-eliciting situations call for
different behavioral responses, however, qualitatively different
profiles of autonomic responding are expected to occur. In particular, those who posit the existence of multiple, functionally distinct “discrete” emotions predict some degree of emotion-specific
autonomic patterning (e.g., Ekman, 1992; Tooby & Cosmides,
2008). For example, overall vascular resistance and blood flow to
the hands are greater during anger than during fear (Cacioppo et
al., 2000; Ekman, Levenson, & Friesen, 1983; Levenson, 1992),
and disgust appears to involve an increase in parasympathetic
activation as well as some degree of sympathetic activation (Rozin,
Haidt, & McCauley, 1999). Although controversy still exists regarding the extent of autonomic differentiation among negative

Positive Emotion Psychophysiology: Conflicting
Messages
Some studies have examined the autonomic aspects of positive
emotion, but these send conflicting messages. On one hand, many
studies suggest that positive emotion involves little or no change in
ANS activation. Early studies comparing “happiness” with several
negative emotions found that happiness led to minimal cardiovascular and electrodermal change, relative to anger, fear, and sadness
(e.g., Ekman et al., 1983; Levenson, Ekman, Heider, & Friesen,
1992; Levenson, 1992). Based upon a meta-analysis of existing
research, Cacioppo and colleagues (2000) concluded that positive
emotions were characterized by less autonomic reactivity than the
negative emotions, likely reflecting a lesser degree of motivational
output.
Other studies suggest that positive emotion is associated with
increased physiological arousal. In particular, studies using comedy film clips to elicit amusement have documented changes in
skin conductance, heart rate, and/or other cardiovascular measures
consistent with increased sympathetic nervous system activation

Michelle N. Shiota, Samantha L. Neufeld, Wan H. Yeung, Stephanie E.
Moser, and Elaine F. Perea, Department of Psychology, Arizona State
University.
Stephanie Moser is now at Albion College. Elaine Perea is now at
Daytona State College.
Correspondence concerning this article should be addressed to Michelle
N. Shiota, Department of Psychology, Arizona State University, P.O. Box
871104, Tempe, AZ 85287-1104. E-mail: Lani.Shiota@asu.edu
1368

 ANS RESPONDING IN FIVE POSITIVE EMOTIONS

(e.g., Christie & Friedman, 2003; Demaree, Schmeichel, Robinson, & Everhart, 2004; Giuliani, McRae, & Gross, 2008; Mauss,
Levenson, McCarter, Wilhelm, & Gross, 2005). Other studies have
observed signs of sympathetic activation associated with “happiness” or “joy” (e.g., Neumann & Waldstein, 2001; Tsai,
Chentsova-Dutton, Freire-Bebeau, & Przymus, 2002). Recently,
researchers have emphasized that sympathetic activation in positive emotions may reflect increased motivational engagement, and
have documented signs of sympathetic activation in participants
receiving positive feedback on a challenging task (e.g., Kreibig,
Gendolla, & Scherer, 2010).
Still other research suggests that positive emotions have the
opposite effect, reducing cardiovascular arousal. Several studies
have documented an “undoing effect” of experimentally elicited
amusement and contentment, finding that these emotions speed
recovery from arousal associated with negative emotion (e.g.,
Fredrickson & Levenson, 1998; Fredrickson, Mancuso, Branigan,
& Tugade, 2000). This might be due to withdrawal of sympathetic
activation, but increased parasympathetic activation could also
explain the effect. It has been proposed that positive emotions
associated with social bonding, in particular, are characterized by
enhanced activation of the vaso-vagal branch of the parasympathetic nervous system (Porges, 1997). Oveis and colleagues (2009)
found that high tonic/resting respiratory sinus arrhythmia was
associated with higher levels of dispositional positive affect, and a
few studies have observed increases in vagal parasympathetic
activation during exposure to pleasant stimuli (e.g., Matsunaga et
al., 2009; McCraty, Atkinson, Tiller, Rein, & Watkins, 1995). Still,
the extent to which specific positive emotion states are characterized by increased parasympathetic activation is quite uncertain.

1369

by ␣-adrenergic, ␤-adrenergic, and/or cholinergic innervation),
and distinguish these from the effects of parasympathetic responding.
Third, few studies have examined more than one positive emotion. Many studies have examined only “happiness” or “joy” as
elicited by relived emotional experience, directed facial action, or
imagining an eliciting situation (e.g., Collet, Vernet-Maury, Delhomme, & Dittmar, 1997; Ekman et al., 1983; Levenson et al.,
1992; Sinha, Lovallo, & Parsons, 1992; Stemmler, 1989). Other
studies have only examined amusement in response to a comedy
film (e.g., Demaree et al., 2004; Guiliani et al., 2008; Mauss et al.,
2005). Recently, a greater number of studies have included two
positive emotions, often with contentment as a presumed lowarousal positive emotion (e.g., Stephens, Christie, & Friedman,
2010), but these also have important limitations. In some studies,
researchers allowed participants to define the emotion constructs
in a relived experience task, so that the content of the eliciting
stimulus was not consistent across participants (e.g., Neumann &
Waldstein, 2001; Tsai et al., 2002). In others, researchers examined physiological differences between participants who reported
different positive emotions in the context of the same experimental
task (e.g., Kreibig et al., 2010). Reports of yet other studies
provide little detail about the content of experimental stimuli (e.g.,
Christie & Friedman, 2003; Stephens et al., 2010; Vrana, 1993).
Although all of these studies offer valuable insight into the autonomic aspects of positive emotion, there is still a need for studies
comparing autonomic responses to controlled stimuli targeting
several specific positive emotions.

The Present Study
Limitations of Prior Research
Prior studies of autonomic responding in positive emotion have
a number of limitations. This is partly because many of these
studies were not designed to study the autonomic aspects of
positive emotion per se, but instead targeted some other psychological process (e.g., the effects of suppressing behavioral expression; emotion response system coherence). First, many studies do
not include a neutral control condition against which the positive
emotion condition(s) can be compared (e.g., Ekman et al., 1983;
Levenson et al., 1992; Neumann & Waldstein, 2001; Tsai et al.,
2002; Vrana, 1993). Emotion induction methods may themselves
have effects on autonomic responding, and any effects of the actual
emotion will be layered on top of these (Christie & Friedman,
2003). In particular, studies using stimuli such as photographs,
film clips, or audio recordings need to account for the “orienting
effect” when assessing the effects of emotion (Bradley, 2009;
Sokolov, 1990; Stekelenburg & van Boxtel, 2002).
Second, many prior studies assessed variables that limit detection of complex ANS responses by confounding the effects of
multiple systems (Stemmler, Grossman, Schmid, & Foerster,
1991). For example, increased heart rate could be caused by an
increase in ␤-adrenergic sympathetic influence, by withdrawal of
vagal parasympathetic influence, or both. Because the physiological aspects of discrete emotions might involve profiles across
multiple systems rather than change in a single system, the ideal
study would include measures that help tease apart different aspects of sympathetic responding (e.g., effects mediated primarily

The present study examined the autonomic aspects of five
positive emotions, as elicited experimentally using slides showing
emotional images in a repeated-measures design.1 Six peripheral
measures of autonomic responding were assessed, allowing for
some discrimination among different sympathetic and vagal parasympathetic mechanisms of physiological reactivity. The five positive emotion constructs were defined by identifying particular
adaptive problems, the cues of a prototypical opportunity to solve
each problem, and the functional behavioral response to each
opportunity. These theoretical definitions formed the basis for
hypotheses regarding autonomic responding, as well as for selecting emotion-eliciting stimuli. Fredrickson (1998, 2001) has proposed that positive emotions generally facilitate broadening attention and building resources, but she and others have also begun to
offer functional accounts of several distinct positive emotions
thought to facilitate appropriate responses to specific fitnessrelevant opportunities presented by the environment (e.g.,
Fredrickson, 1998; Griskevicius, Shiota, & Neufeld, 2010; Shaver,
Morgan & Wu, 1996; Shiota, Keltner, Campos, & Hertenstein,
2004; Shiota, Keltner, & Mossman, 2007).
1

Contentment/serenity, while included in several prior studies of ANS
responding in positive emotion, was not included on the present study. We
agree that contentment may be a functionally distinct positive emotion.
However, the postconsummatory aspect emphasized in functional definitions of this construct (e.g., Fredrickson, 1998; Griskevicius et al., 2009)
made us wary of attempting to elicit it via photographs.

 1370

SHIOTA, NEUFELD, YEUNG, MOSER, AND PEREA

Anticipatory enthusiasm addresses the need for food and other
material resources, and should facilitate the physical and cognitive
effort needed to acquire such resources (Griskevicius et al., 2010).
It is important that this definition of anticipatory enthusiasm differentiates it from sexual arousal, which is reasonably well characterized in the ANS literature (Kreibig, 2010). Anticipatory enthusiasm is experienced in response to cues of imminent reward,
and has also been described as “wanting” by Berridge and Robinson (1995). A rich body of literature documents the dopaminergic neural network that appears to mediate the anticipation of
reward (e.g., Depue & Collins, 1999; Knutson, Taylor, Kaufman,
Peterson, & Glover, 2005), dubbed the “Seeking” system by Panksepp (1998). Prior research suggests that the anticipation of reward, as well as the motivation to pursue rewards and achieve
important goals, is associated with increased cardiovascular
arousal (e.g., Fowles, Fisher, & Tranel, 1982; Kreibig et al., 2010).
Prior research also provides strong evidence of sympathetically
mediated skin conductance responses during the anticipation of
aversive stimuli (Bach, Daunizeau, Friston, & Dolan, 2010), and
exposure to sexually arousing material (see Kreibig, 2010 for a
review), although less is known about skin conductance responses
to anticipated nonsexual rewards. As a result, we hypothesized that
anticipatory enthusiasm would be accompanied by broad sympathetic activation.
Attachment love addresses the need for others’ nurturance and
protection, and has been described as the surge of love and trust
experienced in response to an attachment figure (Griskevicius et
al., 2010; Shaver et al., 1996). This emotion is thought to facilitate
the passive acceptance of nurturance and attention, especially
when one is vulnerable, or when a caregiver has returned from an
absence (Shiota, Keltner, & John, 2006). Although parents and
romantic partners have received the most research attention as
attachment figures, humans may form attachments to others perceived as caregivers as well (Ainsworth, 1989). Prior research
suggests that presence of and physical contact with close others
helps alleviate HPA axis activation in times of stress (Carter, 1998;
Hennessy, 1997), and it has been proposed that attachment love
involves vaso-vagal parasympathetic activation (Porges, 1997).
Thus, we hypothesized that attachment love would be accompanied by baseline-to-trial increase in respiratory sinus arrhythmia.
Nurturant love addresses the need to care for the young
(Griskevicius et al., 2010; Shiota et al., 2006). Individuals displaying youth and helplessness are likely to elicit this response; ethologists have defined “cuteness” as a combination of physical characteristics (e.g., large head/body ratio; short, stubby limbs; large
eyes, small nose, and tight clustering of facial features) and behavioral characteristics (e.g., clumsiness, attention-seeking) that
serve as a mammalian cue for nurturance (Hildebrandt & Fitzgerald, 1979; Lorenz, 1971). It has been proposed that nurturant
behavior draws on a broader appetitive motivational system that
triggers sympathetic arousal (Bradley, 2009). Thus, we hypothesized that nurturant love would involve increased sympathetic
activation, facilitating approach, and caregiving. Although increased vagal response might be involved in affectionate contact
associated with nurturance (i.e., cuddling), a later “stage” of nurturant love, we did not expect this effect in participants viewing
photographic images.
Amusement addresses the need to practice crucial physical and
cognitive skills (Smith, 1982), and has been defined as the emotion

experienced in response to opportunities for play (Griskevicius et
al., 2010; Panksepp, 1998). Although rough-and-tumble physical
play is the prototypical form, cognitive play and humor serve
comparable skill-building purposes and may draw on closely related cognitive and physiological mechanisms (Pellegrini & Smith,
2005). In particular, theories of play emphasize the need to practice risky skills under safe circumstances, without time pressure or
genuine threat (Smith, 1982). Prior research on autonomic aspects
of amusement has produced inconsistent findings across a number
of channels (Kreibig, 2010). Based on our theoretical definition of
this construct, however, we predicted that amusement would be
accompanied by signs of cardiovascular arousal (as observed in
previous studies, e.g., Demaree et al., 2004; Guiliani et al., 2008;
Mauss et al., 2005), and increased respiration rate, reflecting
sympathetic activation and/or vagal withdrawal.
Finally, awe addresses our species’ need to acquire and store
information about the outside world. This emotion has been described as a response to novel, vast, complex stimuli not accounted
for by one’s current worldview (Keltner & Haidt, 1999; Shiota et
al., 2007). Behaviorally, awe should facilitate orientation toward
the stimulus and immobility; cognitive activity should reflect
attentive sensory processing and accommodation (Shiota et al.,
2007). A number of researchers have found that perception of and
orienting toward novel features of the environment (“environmental intake”) is associated with prolonged heart rate deceleration
(Bradley, 2009; Graham, 1979; Lacey & Lacey, 1970), which may
be due to increased parasympathetic activation and/or sympathetic
withdrawal. Demaree et al. (2004) have also suggested that some
kinds of intense cognitive effort are accompanied by sympathetic
withdrawal. Thus, we hypothesized that awe would be accompanied by signs of parasympathetic activation and/or sympathetic
withdrawal.

Method
Participants
Participants were 37 undergraduates enrolled in Psychology
courses at Arizona State University. Mean age was 18.80 years
(SD ⫽ 1.32), 76% of participants were female and 24% male, 57%
were European American, 24% Latino/Latina, 11% African or
African American, 5% Asian, and 3% Native American. Participants received $15 and one hour of course credit for their participation, plus an unexpected, additional $10 award during the
Enthusiasm trial. Four additional participants completed the study,
but their physiological data were unusable due to extreme sources
of noise in the physiological signals (e.g., near-constant sneezing
and coughing) or problems with sensor placement.

Procedure
The experimental protocol was approved by ASU’s Office of
Research Integrity and Assurance prior to data collection. Upon
arrival at the laboratory participants completed informed consent
procedures (including a verbal description of the procedures as
well as a printed consent form), and sensors for physiological
measurement were attached. The protocol included nine slideviewing trials. Within each trial, participants viewed (a) an “X” on
the monitor for a 60-s trial baseline, during which they were asked

 ANS RESPONDING IN FIVE POSITIVE EMOTIONS

to clear their minds of thoughts, feelings, and memories; and (b)
six 15-s slides, for a total of 90 seconds. Slides were presented on
a 42” LCD monitor, suspended on the wall eight feet from the
participant. Participants were instructed to watch the slides carefully, and told that some slides might elicit positive, negative, or no
emotions, but were not given any instructions for how to experience or control emotions they might feel. Participants were also
asked to minimize movement and speech during the trials. E-prime
was used to direct stimulus presentation, and also to embed “triggers” identifying the start of baseline and slide epochs of each trial
in the physiology data files. After delivering trial instructions
verbally, the experimenter left the room before each trial. At the
end of each trial, the experimenter reentered the room to administer an emotional experience questionnaire. Participants were
videotaped throughout the lab session with their knowledge and
consent; audio-visual data are not used in the present analyses.
The neutral slides—images of household objects (e.g., office
desk set, pile of telephone books) —were always used in the first
trial. The order of the remaining trials was randomly determined
for each participant. Data from the three negative emotion trials
(sadness, fear, and disgust) are not used in the present analyses.
People respond to money as a fundamental resource, so we used
slides presenting a lottery-like game leading to an unexpected $10
reward to elicit anticipatory enthusiasm. The first slide presented
five target numbers and the reward scheme, and subsequent slides
showed an increasing set of “matching” numbers leading to the
largest possible reward. In lieu of personalized attachment figures,
images of caregiving childhood fictional characters (pretested for
recognizability by this cohort, e.g., Big Bird, Papa Smurf) were
used to elicit attachment love. Nurturant love slides showed baby
animals (we anticipated that actual baby humans might elicit some
anxiety in this college-age sample). Amusement slides showed
“Far Side” cartoons. Awe slides showed panoramic views, following prior research documenting this prototypical awe elicitor
(Shiota et al., 2007). All stimulus images were bitmap files approximately 300K pixels in size (e.g., 640 ⫻ 480). All images
were brightly colored, and images for most trials were also fairly
complex, showing a foreground figure against a background, or a
complex background only (awe). Enthusiasm slides were somewhat less visually complex, as they primarily showed text, but
these were also brightly colored.

Measures
Physiological data were collected using sensors and amplifying
hardware supplied by Mindware, Inc., Biopac, and Medwave.
Continuous ANS data were recorded using Biopac’s Acquisition
software throughout the laboratory session. Sampling frequency
for all variables was 1000 Hz. In order to remove error and artifact
(movement, sneezing, coughing, brief sensor displacement, etc.),
all data were screened visually by the first author, blind to emotion
condition, prior to further analysis. Six physiological variables
were assessed:
Cardiac Interbeat Interval (IBI, in ms) has an inverse relationship with ␤-adrenergic sympathetic influence, and a positive relationship with vagal parasympathetic influence. Cardiac IBI was
measured using a 3-lead configuration with disposable electrodes
on the left rib cage and right clavicle, with the ground on the right
rib cage. Signals were amplified using Mindware, Inc.’s Imped-

1371

ance Cardiography unit. Data were reduced into target epoch
means using Mindware, Inc’s Impedance Cardiography software
module (version 2.51), which defines IBI as the time elapsed in
milliseconds between R-peaks of the ECG signal, and averages
these R-peaks across the target epoch. Maximum heart rate was set
at 200 beats/min. (minimum IBI ⫽ 300 ms.) and minimum heart
rate at 40 beats/min. (maximum IBI ⫽ 1500 ms.). The few IBIs
outside this range were discarded when calculating epoch means.
Cardiac Pre-Ejection Period (PEP, in ms), the time elapsed
between the beginning of ventricular contraction and the opening
of the aortic valve, shortens with increasing ␤-adrenergic sympathetic influence. Preejection period was measured using four leads
with disposable electrodes placed at the base of the neck, the center
of the clavicle, the xyphoid process, and on the spine approximately one inch inferior to the sensor on the xyphoid process (in
addition to the ECG leads described above). Signals were amplified using Mindware, Inc.’s Impedance Cardiography unit. Data
were reduced into target epoch means using Mindware, Inc’s
Impedance Cardiography software module (version 2.51), which
creates a composite heartbeat across all valid beats in the target
epoch, and calculates the time elapsed in milliseconds between the
Q-point of the ECG signal (defined as the onset of the R-wave, or
the point of maximum positive slope during the 65 milliseconds
preceding the R-peak) and the B-point of the first derivative of
the impedance signal (defined as 56% of the time elapsed from the
R-peak of the ECG signal and the maximum value of dz/dt, the
first derivative of the impedance signal) on this composite beat
(Berntson, Lozano, Chen, & Cacioppo, 2004).
Skin Conductance Responses (SCRs) reflect brief bursts of
sweat gland activity resulting from cholinergic sympathetic receptor stimulation. In the present study, SCRs were measured by
passing a constant 0.5 V voltage between two electrodes attached
to the palmar surfaces of the distal phalanges of the index and ring
fingers of the nondominant hand. Signals were filtered and amplified by Biopac’s GSR100C unit, with low-pass filtering at 10 Hz,
high-pass filtering at DC, and a gain of 5 microSiemens. Valid
SCRs were defined as increases of at least .05 microSiemens
occurring during the slide set, and typically occurred immediately
after a new slide was displayed on the monitor. The total number
of valid SCRs during each slide viewing epoch were counted; skin
conductance responses occurring less than 1 second after the first
slide appeared, or immediately following a deep breath, were not
included in the counts.2
Respiration Rate (in breaths per minute) generally increases
with sympathetic influence and decreases with parasympathetic
influence, although it is also under voluntary control. In the present
study respiration rate was measured using an elastic belt with a
tension-sensitive crystal; signals were filtered and amplified by
Biopac’s RSP100C unit with low-pass filtration set to 10Hz,
high-pass at 0.5 Hz and DC, and a gain of 10. Epoch-level
respiration rates were calculated using Mindware, Inc.’s HRV 2.51
analysis software.
Respiratory Sinus Arrhythmia (RSA) is the variability in IBI
associated with the phases of breathing, increasing with greater
2
Analyses examining baseline-to-trial change in tonic skin conductance
level, rather than counts of skin conductance responses during the slides,
produced the same results as in the current analyses.

 1372

SHIOTA, NEUFELD, YEUNG, MOSER, AND PEREA

vagal parasympathetic influence on the heart. In the present study,
RSA was derived from the IBI series over the course of each
epoch, using Mindware, Inc.’s HRV 2.51 analysis software. This
program calculates RSA by subjecting the IBI series for each
epoch to Fast Fourier Transform, and applying a Hamming window for the .12–.40 Hz frequency range of the resulting spectral
distribution, which offers a reliable estimate of the extent of
parasympathetic influence on the heart (Berntson, Cacioppo, Quigley, & Fabro, 1994). Spectral distributions of the respiration signals were also examined to ensure that integral power peaked
within the .12–.40 Hz frequency range corresponding to the expected respiration rates for using RSA as a marker of parasympathetic activation. Trials with respiration rates outside the expected
range were removed from analysis.
There is controversy in the literature over whether or not
changes in RSA should be examined only after controlling for
corresponding changes in respiration rate, which exerts an effect
on RSA independent of vagal influence (e.g., Allen, Chambers, &
Towers, 2007; Butler, Wilhelm, & Gross, 2006; Porges, 2007). For
this reason, analyses of baseline-to-trial change in RSA were
performed both directly on the change scores, and again controlling for the corresponding change in respiration rate. In the latter
analyses, baseline-to-trial changes in RSA for each trial across the
sample were regressed onto baseline-to-trial changes in respiration
rate; unstandardized residuals were saved and used in a second set
of hypothesis testing analyses.
Mean Arterial Pressure (MAP, in mmHg) increases with greater
␣-adrenergic influence, and is also affected both directly and via
complex feedback loops by ␤-adrenergic influence and vagal parasympathetic influence. We measured MAP using Medwave’s Vasotrac APM205A, which uses an elastic band on the wrist of the
dominant arm, adjusts tension at 12-s intervals, and detects effects on
pulse magnitude using a sensor at the pulse point of the wrist.
Participants were asked to rest their dominant arm on the chair with
the wrist facing up, in order to avoid putting pressure on the sensor.
All six measures were collected throughout the laboratory session.
For all measures except SCRs, mean values were calculated for the
baseline (while the “X” was on the screen) and slide-viewing epochs
of each trial, and trial-level baseline-to-trial change scores were calculated from these means. These change scores were taken to represent reactivity to the slides, and used as the dependent variables in
further analyses. Skin Conductance Responses were simply counted
during the slide-viewing epochs of each trial.
Self-report measures of emotion. In addition to the ANS
measures, participants’ ratings of their emotional experience were
collected at the end of each trial. Immediately after the last slide in
each trial the experimenter entered the room, and asked participants to rate the intensity of their experience of 10 emotions, as felt
while viewing the slides. Ratings were on a scale from 0 (did not
feel that emotion at all) to 8 (strongest experience of that emotion
ever). The 10 rated emotions were: Amusement, Anger, Awe,
Contentment, Disgust, Enthusiasm/Excitement, Fear, Love/
Attachment, Sadness, and Tenderness/Compassion.

Analyses
In the first stage of hypothesis testing, Multivariate Analysis of
Variance in SPSS was used to examine the main effect of emotion
condition on the set of ANS variables (excluding MAP). This analysis

asks whether the several emotion conditions can be differentiated with
respect to a linear composite of the ANS variables, but does not
address differences between emotion conditions in terms of profile
across ANS variables. This analysis included only the 28 participants
providing complete IBI, PEP, SCL, respiration rate, and RSA data for
all six trials. Mean arterial pressure was excluded from this analysis
because eliminating participants with missing data for at least one of
the six trials would have reduced the sample for the test to 11.3
A 6 [Emotion] ⫻ 5 [Physiological Measure] Repeated Measures
Analysis of Variance was then used to confirm the omnibus effect
of emotion condition on the set of physiological variables (main
effect of Emotion condition), but also to assess the extent to which
the emotion conditions led to qualitatively different profiles of
ANS responding (Emotion condition ⫻ Physiological Measure
interaction). Profiles across ANS variables are at least as important as
linear composites of such variables in documenting possible differences among emotion conditions (e.g., Kreibig, Wilhelm, Roth, &
Gross, 2007; Stemmler, 1989). The Greenhouse-Geisser correction of
p was always applied to compensate for violations of sphericity.
Next, overall differences among the Neutral and five positive
emotion conditions in baseline-to-trial change scores for each ANS
measure (for SCRs, counts during the trial) were assessed using a
series of omnibus one-way repeated measures analyses of variance
(ANOVAs). Again, the Greenhouse-Geisser correction of p was
applied in each test to compensate for any violations of sphericity
in the data. Each of these analyses included all participants with
acceptable data on the target physiological variable for every
emotion condition, including some not represented in the multivariate analyses, in order to maximize statistical power (see Table
2 for Ns for each analysis). Omnibus ANOVAs were followed by
t tests examining planned pairwise contrasts between the Neutral
condition and each of the five positive emotion conditions, as well
as exploratory t tests contrasting positive emotion conditions with
each other. With the exception of MAP analyses, these t tests used
only the participants included in the omnibus one-way ANOVAs.4
Unusability of MAP data in some trials limited the sample with
data for all trials to only 11 participants. Thus, t tests for MAP used
all participants with data needed for a given pairwise test, in order
to maximize statistical power.

Results
Manipulation Check
Mean ratings of experienced Enthusiasm/Excitement, Love/
Attachment, Tenderness/Compassion, Amusement, and Awe
while viewing the slide sets are reported in Table 1. Mean reports
3
Unfortunately, MAP sensor recalibration (which takes place at regular
intervals using the Vasotrac system) led to substantial loss of data during
the baseline and/or slide-viewing epoch of at least one trial for many
participants. As a result, only 11 participants had the necessary baseline
and slide-epoch MAP data to calculate a change score for every trial,
necessary for inclusion in the MANOVA.
4
The pattern of findings observed using this approach is also produced
by using all eligible subjects in the pairwise t tests (as for MAP), and when
pairwise contrasts are limited to those reporting at least moderate levels of
the target positive emotion in each contrast. The current approach was
selected by virtue of simple graphical presentation.

 Note. All t-values contrasting ratings of the target emotion with ratings of the same emotion in other trials (e.g., ratings of Enthusiasm/Excitement in the Enthusiasm versus Neutral trial) are significant
at the p ⬍ .01 level.

3.64 (2.70) tvEnth ⫽ ⫺3.59
2.04 (2.46) tvAL ⫽ ⫺2.76
2.18 (2.47) tvNL ⫽ ⫺8.02
1.89 (2.33) tvAmu ⫽ ⫺9.90
5.86 (1.98)
4.43 (2.70) tvEnth ⫽ ⫺2.33 3.64 (2.28) tvEnth ⫽ ⫺3.87
5.61 (1.77) tvAL ⫽ 5.71
.43 (1.23) tvAL ⫽ ⫺6.50
5.93 (1.70)
.89 (1.69) tvNL ⫽ ⫺14.06
4.57 (2.08) tvAmu ⫽ ⫺4.81 6.14 (1.35)
3.25 (2.52) tvAwe ⫽ ⫺4.89 1.61 (1.87) tvAwe ⫽ ⫺8.88
3.25 (2.19) tvEnth ⫽ ⫺5.52
3.21 (2.18)
2.96 (2.30) tvNL ⫽ ⫺6.59
4.29 (1.84) tvAmu ⫽ ⫺5.80
1.18 (1.59) tvAwe ⫽ ⫺9.28
Emotional Experience
Rated
Enthusiasm/Excitement
.93 (1.74) tvEnth ⫽ ⫺10.32 5.54 (1.90)
Love/Attachment
.11 (.57) tvAL ⫽ ⫺7.41
.36 (1.19) tvAL ⫽ ⫺6.56
Tenderness/Compassion .32 (.90) tvNL ⫽ ⫺14.93
.32 (1.19) tvNL ⫽ ⫺16.56
Amusement
1.71 (2.14) tvAmu ⫽ ⫺10.99 4.50 (2.35) tvAmu ⫽ ⫺4.30
Awe
.46 (1.20) tvAwe ⫽ ⫺12.97 2.39 (2.60) tvAwe ⫽ ⫺6.50

Nurturant love
Attachment love
Enthusiasm
Neutral

Emotion condition

Table 1
Self-Reported Emotional Experience, Means and Standard Deviations by Emotion Condition

Amusement

Awe

ANS RESPONDING IN FIVE POSITIVE EMOTIONS

1373

of all five positive emotions were less than 2.0 in the Neutral
condition. The Anticipatory Enthusiasm slides elicited stronger
reports of Enthusiasm (M ⫽ 5.54, SD ⫽ 2.04) than of any other
positive emotion, and reported Enthusiasm was significantly
higher in this condition than in any other emotion condition (see
Table 1 for t-values associated with these differences).
The Attachment Love slides elicited moderately strong reports
of Love/Attachment (M ⫽ 3.22, SD ⫽ 2.36), and reports of
Love/Attachment were significantly higher in this condition than
in the Neutral, Anticipatory Enthusiasm, Amusement, and Awe
conditions. Reports of Love/Attachment were significantly higher
in the Nurturant Love condition than in the Attachment Love
condition; however, this is consistent with the colloquial use of the
term “love” and was not taken to indicate invalidity of either of the
“love” stimuli. Reports of Amusement were also moderately high
in the Attachment Love condition, though significantly lower than
in the Amusement condition.
The Nurturant Love slides elicited stronger reports of Tenderness/Compassion (M ⫽ 5.49, SD ⫽ 2.08) than of any other
positive emotion, and reported Tenderness/Compassion was significantly higher in this condition than in any other emotion
condition. The Amusement slides elicited stronger reports of
Amusement (M ⫽ 6.16, SD ⫽ 1.40) than of any other positive
emotion, and reported Amusement was significantly higher in this
condition than in any other emotion condition. Finally, the Awe
slides elicited stronger reports of Awe (M ⫽ 5.62, SD ⫽ 2.028)
than of any other positive emotion, and reported Awe was significantly higher in this condition than in any other emotion condition. Thus, the manipulation check suggests that the target emotions were successfully elicited in each emotion condition, with the
Attachment Love slides eliciting mild levels of amusement as well.

Hypothesis Testing
Results are summarized numerically in Table 2, and graphically
in Figure 1. Multivariate Analysis of Variance in SPSS indicated
that the main effect of Emotion Condition on the composite of IBI,
PEP, SCR, Respiration Rate, and RSA was significant, F(25,
675) ⫽ 2.64, p ⬍ .001, indicating overall differences among the
emotion conditions. The 6 [Emotion Condition] ⫻ 5 [ANS Variable] Repeated Measures ANOVA confirmed the main effect of
Emotion Condition on the set of ANS variables, Epsilon ⫽ .851,
F(5, 135) ⫽ 2.75, p ⫽ .028, and also detected a significant
interaction between Emotion Condition and ANS Variable, Epsilon ⫽ .215, F(20, 540) ⫽ 2.93, p ⫽ .021, indicating qualitative
differences among Emotion Conditions in the profiles of responding across ANS variables.5 These effects remained significant
when the analysis was performed using RSA change scores controlling for change in Respiration Rate: for main effect of Emotion
Condition Epsilon ⫽ .852, F(5, 135) ⫽ 2.72, p ⫽ .030; for
Emotion Condition ⫻ ANS Variable interaction Epsilon ⫽ .214,
F(20, 540) ⫽ 2.95, p ⫽ .021.
Turning to the one-way ANOVAs for each physiological measure, omnibus differences among the six emotion conditions were
The Emotion Condition ⫻ Physiological Measure interaction is also
significant in analyses using residual variability in trial-epoch means after
controlling for baseline-epoch means as the index of reactivity, rather than
baseline-to-trial change scores, F(20, 540) ⫽ 3.356, p ⫽ .001.
5

 SHIOTA, NEUFELD, YEUNG, MOSER, AND PEREA

1374

Table 2
Baseline and Trial Epoch Means and Standard Errors for ANS Measures, by Emotion Condition

IBI (n ⫽ 37)
Baseline Mean (SE)
Trial Mean (SE)
t for BT ⌬ vs. Neutral
PEP (n ⫽ 35)
Baseline Mean (SE)
Trial Mean (SE)
t for BT ⌬ vs. Neutral
SCRs (n ⫽ 35)
Trial Mean (SE)
t vs. Neutral
Respiration rate (n ⫽ 32)
Baseline Mean (SE)
Trial Mean (SE)
t for BT ⌬ vs. Neutral
RSA (n ⫽ 29)
Baseline Mean (SE)
Trial Mean (SE)
t for BT ⌬ vs. Neutral
MAP
N for emotion
Baseline Mean (SE)
Trial Mean (SE)
t for BT ⌬ vs. Neutral

Neutral

Enthusiasm

Attachment love

Nurturant love

Amusement

Awe

833.5 (23.4)
851.3 (21.8)

859.0 (21.5)
859.1 (23.0)
⫺1.85⫹

852.7 (22.1)
848.1 (22.6)
⫺2.88ⴱⴱ

856.6 (24.1)
856.3 (23.5)
⫺2.14ⴱ

859.9 (23.4)
874.0 (22.8)
⫺.45

847.8 (23.0)
862.1 (22.1)
⫺.44

101.1 (1.4)
101.5 (1.3)

102.9 (1.1)
102.5 (1.3)
⫺1.36

103.1 (1.1)
103.1 (1.2)
⫺.68

102.8 (1.2)
103.4 (1.2)
.65

103.1 (1.1)
102.9 (1.2)
⫺.92

102.2 (1.1)
103.6 (1.1)
2.86ⴱⴱ

1.71 (.42)

3.74 (.64)
4.44ⴱⴱ

1.89 (.39)
.51

1.46 (.38)
⫺1.04

1.94 (.36)
.81

1.29 (.32)
⫺1.23

14.37 (.53)
14.59 (.49)

13.91 (.61)
15.22 (.48)
1.77⫹

13.91 (.59)
14.94 (.60)
1.69

13.75 (.58)
15.34 (.38)
2.60ⴱ

13.78 (.56)
15.69 (.46)
3.51ⴱⴱ

13.34 (.62)
14.94 (.51)
3.02ⴱⴱ

6.81 (.22)
6.90 (.22)

7.12 (.24)
6.67 (.23)
⫺2.55ⴱ

7.11 (.22)
6.79 (.24)
⫺2.74ⴱ

6.99 (.24)
6.80 (.25)
⫺1.85⫹

6.93 (.20)
6.95 (.20)
⫺.41

7.11 (.22)
6.71 (.24)
⫺2.34ⴱ

23
86.87 (2.8)
86.19 (2.8)

21
84.20 (2.6)
85.21 (2.6)
1.76⫹(df ⫽ 20)

20
84.46 (2.8)
82.98 (2.8)
⫺.12 (df ⫽ 19)

18
84.52 (2.6)
84.97 (2.6)
.82 (df ⫽ 17)

17
86.34 (3.6)
82.83 (3.0)
⫺.35 (df ⫽ 16)

19
80.88 (2.4)
80.27 (2.6)
⫺.11 (df ⫽ 18)

Note. IBI ⫽ Cardiac Interbeat Interval in ms; PEP ⫽ Cardiac Pre-Ejection Period in ms; SCRs ⫽ number of valid Skin Conductance Responses during
trial only; RSA ⫽ Respiratory Sinus Arrhythmia in ms2; MAP ⫽ Mean Arterial Pressure in mmHg.
⫹
p ⬍ .10. ⴱ p ⬍ .05. ⴱⴱ p ⬍ .01.

statistically significant for change in IBI [Epsilon ⫽ .822, F(5,
180) ⫽ 2.68, p ⫽ .032], number of SCRs [Epsilon ⫽ .614, F(5,
170) ⫽ 10.98, p ⬍ .001], and change in Respiration Rate [Epsilon ⫽ .840, F(5, 155) ⫽ 2.62, p ⫽ .035], and approached significance for change in PEP [Epsilon ⫽ .456, F(5, 170) ⫽ 2.87, p ⫽
.056]. The six conditions also differed significantly on change in
RSA [Epsilon ⫽ .775, F(5, 140) ⫽ 2.88, p ⫽ .027], although this
omnibus effect ceased to be significant after controlling for change
in Respiration Rate [Epsilon ⫽ .694, F(5, 140) ⫽ .36, ns]. Omnibus differences among emotion conditions were not significant
for change in MAP [Epsilon ⫽ .661, F(5, 50) ⫽ 1.04, ns]. Relative
to baseline, Neutral slides led to modest and nonsignificant increases in IBI and PEP and decrease in MAP, a small number of
SCRs, and no effect on Respiration Rate or RSA— effects consistent with the orienting response.
Compared with responses to the Neutral stimuli, Anticipatory
Enthusiasm led to a significantly greater number of SCRs, t ⫽
4.44, p ⬍ .001. Number of SCRs during the Enthusiasm slides was
also significantly greater than during the Attachment Love, t ⫽
3.72, p ⫽ .001, Nurturant Love, t ⫽ 4.49, p ⬍ .001, Amusement,
t ⫽ 4.05, p ⬍ .001, and Awe slides, t ⫽ 4.44, p ⬍ .001. Differences of the Anticipatory Enthusiasm from the Neutral condition
on three other physiological measures also approached significance: smaller increase in IBI, t ⫽ ⫺1.85, p ⫽ .072, greater
increase in respiration rate, t ⫽ 1.77, p ⫽ .087, and increase in
MAP, t ⫽ 1.76, p ⫽ .094. The increase in MAP also distinguished
Anticipatory Enthusiasm from Amusement, t ⫽ 2.16, p ⫽ .048.
Anticipatory Enthusiasm led to a decrease in RSA that differed
significantly from change in RSA during the Neutral slides (t ⫽
⫺2.55. p ⫽ .017). However, this effect was no longer significant

after controlling for baseline-to-trial change in Respiration Rate.
Similarly, change in RSA differed significantly between the Enthusiasm and Amusement trials, t ⫽ 2.74, p ⫽ .011, although this
effect was no longer significant after controlling for changes in
respiration.
Attachment Love led to a decrease in IBI that differed significantly from the Neutral condition, t ⫽ ⫺2.88, p ⫽ .007, as well as
from the Amusement condition, t ⫽ ⫺2.49, p ⫽ .017 and the Awe
condition, t ⫽ ⫺2.22, p ⫽ .033. Change in PEP during Attachment
Love did not differentiate it from the Neutral condition, but did
differ significantly from that in Awe, t ⫽ ⫺2.15, p ⫽ .039.
Number of valid SCRs observed during Attachment Love was
significantly lower than during Enthusiasm, as reported above.
Attachment Love also led to a decrease in RSA that differed
significantly from change in RSA during the Neutral slides (t ⫽
⫺2.74. p ⫽ .011). However, this effect was no longer significant
after controlling for baseline-to-trial change in Respiration Rate.
No other differences between Attachment Love and other emotion
conditions were significant.
Nurturant Love led to a decrease in IBI, t ⫽ ⫺2.14, p ⫽ .040
and increase in Respiration Rate, t ⫽ 2.60, p ⫽ .014 that differed
significantly from the Neutral condition. Change in PEP during
Nurturant Love did not differentiate it from the Neutral condition,
but did differ significantly from that in Awe, t ⫽ ⫺2.19, p ⫽ .036.
Number of valid SCRs observed during Nurturant Love was significantly lower than during Enthusiasm, as reported above. Nurturant Love also led to a marginally significant decrease in RSA
compared with change in RSA during the Neutral slides (t ⫽
⫺1.85. p ⫽ .075). However, this effect disappeared after controlling for baseline-to-trial change in Respiration Rate. No other

 ANS RESPONDING IN FIVE POSITIVE EMOTIONS

Figure 1. Baseline-to-Trial Changes in ANS Variables, by Emotion Condition Note: Error bars represent
standard errors around the observed mean.

1375

 SHIOTA, NEUFELD, YEUNG, MOSER, AND PEREA

1376

differences between Nurturant Love and other emotion conditions
were significant.
The baseline-to-trial increase in IBI did not significantly distinguish Amusement from the Neutral condition, but did distinguish
it from Attachment Love, as reported above. Change in PEP during
Amusement also did not differentiate it from the Neutral condition,
but did differ significantly from that in Awe, t ⫽ ⫺2.65, p ⫽ .012.
Number of valid SCRs observed during Amusement did not differ
significantly from Neutral, but was greater than during Awe, t ⫽
2.09, p ⫽ .044, and less than Anticipatory Enthusiasm (as reported
above). Amusement led to a significantly greater increase in Respiration Rate than the Neutral condition, t ⫽ 3.51, p ⫽ .005. Also
as reported above, change in RSA differed significantly between
the Enthusiasm and Amusement trials, although this effect was no
longer significant after controlling for changes in respiration. The
decrease in MAP observed during Amusement did not distinguish
it significantly from the Neutral condition, but did distinguish it
significantly from Anticipatory Enthusiasm, as reported above. No
other differences between Amusement and other emotion conditions were significant.
Awe did not lead to a change in IBI that differed significantly
from the Neutral condition, although it did differ significantly from
Attachment Love on this measure as described above. Awe led to
a lengthening of PEP that differed significantly from the Neutral
condition, t ⫽ 2.86, p .007, as well as from each of the other four
positive emotion conditions, as described above. Awe involved
fewer SCRs than did Anticipatory Enthusiasm and Amusement, as
reported above. Awe led to an increase in Respiration Rate that
distinguished it from the Neutral condition, t ⫽ 3.02, p ⫽ .005, but
not the other positive emotions. Awe also led to a significant
decrease in RSA compared with the Neutral slides (t ⫽ ⫺2.34. p ⫽
.075). However, this effect disappeared after controlling for
change in Respiration Rate. No other differences between Awe and
other emotion conditions were significant.

Discussion
The current study offers evidence of differences in autonomic
nervous system responding associated with multiple positive emotions. We detected significant overall differentiation among six
emotion conditions—five positive, one neutral—not only in terms
of a main effect on the set of physiological measures used in the
study, but also in the profile of responding across measures. This
suggests that ANS responses associated with the emotions in this
study differed not only as a matter of degree, but also as a matter
of kind. Moreover, we observed omnibus univariate differences
among the emotion conditions on four ANS variables (IBI, PEP,
SCRs, and Respiration Rate), and also differentiated each of the
five positive emotions from the neutral control on at least one
variable. This differentiation at the level of individual physiological measures is compelling, even compared to the many studies
successfully contrasting multiple negative emotions using multivariate approaches. We believe that the degree of differentiation
observed in the present study reflects our strict definition of the
positive emotion constructs in theoretical terms, specifying the
distinct fitness-enhancing function of each, and using that function
to guide stimulus selection.
Physiological changes during anticipatory enthusiasm, thought
to facilitate active pursuit of material rewards, were consistent with

the hypothesized broad increase in sympathetic activation. In contrast, the lengthening of PEP associated with Awe was consistent
with the hypothesized withdrawal of sympathetic influence—at
least the ␤-adrenergic component—and directions of change in
other variables (with the exception of respiration rate) were also
consistent with this interpretation. The relatively shortened IBI
associated with Attachment Love, combined with nonsignificant
decreases in PEP and MAP, somewhat resembles the “challenge
pattern” identified by Tomaka, Blascovich, Kibler, and Ernst
(1997). This may suggest an increase in ␤-adrenergic, but not
␣-adrenergic, aspects of sympathetic influence. This effect differs
from the increase in vagal parasympathetic activation we had
predicted, but might well facilitate effortful approach toward a
still-distant attachment figure. Nurturant Love was associated with
relatively shortened IBI and increased respiratory rate, but no sign
of reduced arterial pressure, and other variables offered few clues
to the mechanism behind this increase in “arousal.” Finally,
amusement did not lead to the hypothesized increase in cardiovascular arousal. While this finding is inconsistent with effects observed in a number of other studies, it is consistent with studies of
the “undoing” effect of amusement and other positive emotions
(e.g., Fredickson et al., 2000). Methodological differences may
well explain this inconsistency, a subject to be explored in future
research.
A great deal more work is still needed to fully characterize the
ANS aspects of particular positive emotions. The present study
examined 90-s “snapshots” of autonomic responding associated
with one stage of each emotion, as elicited by theoretically relevant
photographs. In some cases, the physiological aspects of particular
emotions may be best characterized by sequences of responding
rather than a single stage, and our study would not have detected
such effects. For example, the enhanced vagal parasympathetic
activation we predicted for attachment love might be observed
during actual reunion with an attachment figure (especially physical contact or “cuddling”), whereas detection of and approach
toward such a figure involves the increased ␤-adrenergic influence
observed in this study. It is also possible that the nature of ANS
responding in a particular emotion depends to some degree on the
way that emotion is elicited. The present study would not be able
to detect such moderator effects.
Nonetheless, the present study offers strong added support for
autonomic differentiation among some positive emotions, and thus
adds to the evidence for the existence of multiple functionally
distinct positive emotion constructs. We do not claim that the five
emotions in the present study are the only positive emotions, and
in fact studies using other methodologies offer some evidence for
additional constructs (e.g., Griskevicius, Shiota, & Neufeld, 2010;
Shiota et al., 2006; Tracy & Robins, 2004). It is also plausible that,
as some have argued, positive emotion space is not organized into
“discrete” categories at all, but rather reflects interaction among
several continuous dimensions (e.g., Feldman Barrett & Russell,
1998; Scherer, 2001). These data do, however, add to the growing
body of literature showing that positive emotion is not a single,
unidimensional phenomenon. Furthermore, the data are not well
explained by models of emotion that limit differentiation to highversus low-arousal states. Most of the positive emotions in this
study (except awe) elicited some increase in “arousal,” but more
specific physiological changes varied from emotion to emotion in
ways that likely reflect different underlying autonomic mecha-

 ANS RESPONDING IN FIVE POSITIVE EMOTIONS

nisms. Prior studies have found that different experimentally elicited positive emotions have distinct effects of the processing of
persuasive messages (Griskevicius, Shiota, & Neufeld, 2010); and
on the attractiveness of specific consumer products (Griskevicius,
Shiota & Nowlis, 2010). Other studies have found that different
positive emotion dispositions show different patterns of correlation
with core aspects of personality (Shiota et al., 2006). The present
study documents qualitative differences among five positive emotions at the physiological level as well.
The present study does have a number of limitations. As noted
earlier, our methods only captured brief snapshots of each positive
emotion state—the point at which the eliciting stimulus is detected
visually. Also, we elicited the target emotions using the same
visual images for all participants. This inevitably leads to some
loss in personal impact— especially for attachment love stimuli.
Further, studies using event-related potential and startle response
outcomes often find that simple figure-ground images convey
hedonic content more effectively than complex scenes (e.g., Bradley, Hamby, Löw, & Lang, 2007). The inclusion of awe as one of
the target emotions in the present study necessitated more complex
images, but this might have diluted hedonic impact. Additional
studies of these phenomena should use other elicitation methods,
including film clips (which are likely to be more potent), and
relived experiences (which will be more personal). Finally, losses
of data in some ANS variables (particularly MAP) mean that null
effects should be treated with the usual caution.
Despite these limitations, the present findings suggest that positive emotions still deserve greater attention than they have received from emotion research. Recent years have seen great advances in research on the global functions and characteristics of
positive emotion (e.g., Fredrickson, 2001; Fredrickson & Branigan, 2005; Johnson & Fredrickson, 2005; Tugade & Fredrickson,
2004). Still, we have investigated the spectrum of positive emotions far less carefully than that of negative emotions. Distinct
positive emotion constructs can be defined in functional/
evolutionary terms, and these theoretical definitions are just as
useful in generating empirical hypotheses as is true of the negative
emotions. In failing to examine the depth and complexity of
positive emotions, we lose a great opportunity to understand the
depth and complexity of our own nature.

References
Ainsworth, M. D. A. (1989). Attachments beyond infancy. American
Psychologist, 44, 709 –716.
Allen, J. J. B., Chambers, A. S., & Towers, D. N. (2007). The many metrics
of cardiac chronotropy: A pragmatic primer and a brief comparison of
metrics. Biological Psychology, 74, 243–262.
Bach, D. R., Daunizeau, J., Friston, K. J., & Dolan, R. J. (2010). Dynamic
causal modeling of anticipatory skin conductance responses. Biological
Psychology, 85, 163–170.
Berntson, G. G., Cacioppo, J. T., Quigley, K. S., & Fabro, V. T. (1994).
Autonomic space and psychophysiological response. Psychophysiology,
31, 44 – 61.
Berntson, G. G., Lozano, D. L., Chen, Y.-J., & Cacioppo, J. T. (2004).
Where to Q in PEP. Psychophysiology, 41, 333–337.
Berridge, K. C., & Robinson, T. E. (1995). The mind of an addicted brain:
Neural sensitization of wanting versus liking. Current Directions in
Psychological Science, 4, 71–76.
Bradley, M. M. (2009). Natural selective attention: Orienting and emotion.
Psychophysiology, 46, 1–11.

1377

Bradley, M. M., Hamby, S., Löw, A., & Lang, P. J. (2007). Brain potentials
in perception: Picture complexity and emotional arousal. Psychophysiology, 44, 364 –373.
Butler, E. A., Wilhelm, F. H., & Gross, J. J. (2006). Respiratory sinus
arrhythmia, emotion, and emotion regulation during social interaction.
Psychophysiology, 43, 612– 622.
Cacioppo, J. T., Berntson, G. G., Larsen, J. T., Pohlmann, K. M., & Ito,
T. A. (2000). The psychophysiology of emotion. In M. Lewis & J. M.
Haviland-Jones (Eds.), Handbook of emotions (2nd ed., pp. 173–191).
New York: Guilford Press.
Cannon, W. B. (1915). Bodily changes in pain, hunger, fear, and rage: An
account of recent researches into the function of emotional excitement.
New York: Appleton & Co.
Carter, C. S. (1998). Neuroendocrine perspectives on social attachment and
love. Psychoneuroendocrinology, 23, 779 – 818.
Christie, I. C., & Friedman, B. H. (2003). Autonomic specificity of discrete
emotion and dimensions of affective space: A multivariate approach.
International Journal of Psychophysiology, 51, 143–153.
Collet, C., Vernet-Maury, E., Delhomme, G., & Dittmar, A. (1997). Autonomic nervous system pattern specificity to basic emotions. Journal of
the Autonomic Nervous System, 62, 45–57.
Damasio, A. (1999). The feeling of what happens: Body and emotion in the
making of consciousness. Fort Worth, TX: Harcourt College.
Demaree, H. A., Schmeichel, B. J., Robinson, J. L., & Everhart, D. E. (2004).
Behavioural, affective, and physiological effects of negative and positive
emotional exaggeration. Cognition and Emotion, 18, 1079 –1097.
Depue, R. A., & Collins, P. (1999). Neurobiology of the structure of
personality: Dopamine, facilitation of incentive motivation, and extraversion. Behavioral and Brain Sciences, 22, 491–569.
Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6, 169 –200.
Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic nervous
system activity distinguishes among emotions. Science, 221, 1208 –1210.
Feldman Barrett, L., & Russell, J. A. (1998). Independence and bipolarity
in the structure of current affect. Journal of Personality and Social
Psychology, 74, 967–984.
Fowles, D. C., Fisher, A. E., & Tranel, D. T. (1982). The heart beats to
reward: The effect of monetary incentive on heart rate. Psychophysiology, 19, 506 –513.
Fredrickson, B. L. (1998). What good are positive emotions? Review of
General Psychology, 2, 173–186.
Fredrickson, B. L. (2001). The role of positive emotions in positive
psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56, 218 –226.
Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden the
scope of attention and thought-action repertoires. Cognition and Emotion, 19, 313–332.
Fredrickson, B. L., & Levenson, R. W. (1998). Positive emotions speed
recovery from the cardiovascular sequelae of negative emotions. Cognition and Emotion, 12, 191–220.
Fredrickson, B. L., Mancuso, R. A., Branigan, C., & Tugade, M. M.
(2000). The undoing effect of positive emotions. Motivation and Emotion, 24, 237–258.
Friedman, B. H. (2010). Feelings and the body: The Jamesian perspective on
autonomic specificity of emotion. Biological Psychology, 84, 383–393.
Frijda, N. H. (1986). The emotions. New York: Cambridge University
Press.
Graham, F. K. (1979). Distinguishing among orienting, defense, and startle
reflexes. In H. D. Kimmel, E. H. van Olst, & J. F. Orlebeke (Eds.), The
orienting reflex in humans (pp. 137–167). Hillsdale, NJ: Erlbaum.
Griskevicius, V., Shiota, M. N., & Neufeld, S. L. (2010). Influence of
different positive emotions on persuasion processing: A functional evolutionary approach. Emotion. In press.
Griskevicius, V., Shiota, M. N., & Nowlis, S. M. (2010). The many shades

 1378

SHIOTA, NEUFELD, YEUNG, MOSER, AND PEREA

of rose-colored glasses: Discrete positive emotions and product perception. Journal of Consumer Research, 37, 238 –250.
Guiliani, N. R., McRae, K., & Gross, J. J. (2008). The up- and downregulation of amusement: Experiential, behavioral, and autonomic consequences. Emotion, 8, 714 –719.
Hennessy, M. B. (1997). Hypothalamic-pituitary-adrenal responses to brief
social separation. Neuroscience and Biobehavioral Reviews, 21, 11–29.
Hildenbrandt, K. A., & Fitzgerald, H. E. (1979). Adults’ perceptions of
infant sex and cuteness. Sex Roles, 5, 471– 481.
James, W. (1884). What is an emotion? Mind, 9, 188 –205.
Johnson, K. J., & Fredrickson, B. L. (2005). “We all look the same to me”:
Positive emotions eliminate the own-race bias in face recognition. Psychological Science, 16, 875– 881.
Keltner, D., & Haidt, J. (2003). Approaching awe, a moral, spiritual, and
aesthetic emotion. Cognition and Emotion, 17, 297–314.
Knutson, B., Taylor, J., Kaufman, M., Peterson, R., & Glover, G. (2005).
Distributed neural representation of expected value. Journal of Neuroscience, 25, 4806 – 4812.
Kreibig, S. D. (2010). Autonomic nervous system activity in emotion: A
review. Biological Psychology, 84, 394 – 421.
Kreibig, S. D., Gendolla, G. H. E., & Scherer, K. R. (2010). Psychophysiological effects of emotional responding to goal attainment. Biological
Psychology, 84, 474 – 487.
Kreibig, S. D., Wilhelm, F. H., Roth, W. T., & Gross, J. J. (2007).
Cardiovascular, electrodermal, and respiratory response patterns to fearand sadness-inducing films. Psychophysiology, 44, 787– 806.
Lacey, J. I., & Lacey, B. C. (1970). Some autonomic-central nervous
system relationships. In P. Black (Ed.), Physiological correlates of
emotion (pp. 205–227). New York: Academic Press.
Lang, P. J. (1985). The cognitive psychophysiology of emotion: Fear and
anxiety. In A. H. Tuma & D. Maser (Eds.), Anxiety and the anxiety
disorders (pp. 131–170). Hillsdale, NJ: Erlbaum.
Levenson, R. W. (1992). Autonomic nervous system differences among
emotions. Psychological Science, 3, 23–27.
Levenson, R. W., Ekman, P., Heider, K., & Friesen, W. V. (1992). Emotion
and autonomic nervous system activity in the Minangkabau of West
Sumatra. Journal of Personality and Social Psychology, 62, 972–988.
Lorenz, K. (1971). Part and parcel in animal and human societies. In K.
Lorenz, Studies in animal and human behavior (Vol. 2, pp. 115–195).
Cambridge, MA: Harvard University Press.
Matsunaga, M., Isowa, T., Kimura, K., Miyakoshi, M., Kanayama, N.,
Murakami, H., . . . Ohira, H. (2009). Associations among positive mood,
brain, and cardiovascular activities in an affectively positive situation.
Brain Research, 1263, 93–103.
Mauss, I. B., Levenson, R. W., McCarter, L., Wilhelm, F. H., & Gross, J. J.
(2005). The tie that binds? Coherence among emotion experience, behavior, and physiology. Emotion, 5, 175–190.
McCraty, R., Atkinson, M., Tiller, W. A., Rein, G., & Watkins, A. D. (1995).
The effects of emotions on short-term power spectrum analysis of heart rate
variability. American Journal of Cardiology, 76, 1089 –1093.
Neumann, S. A., & Waldstein, S. R. (2001). Similar patterns of cardiovascular
response during emotional activation as a function of affective valence and
arousal and gender. Journal of Psychosomatic Research, 50, 245–253.
Oveis, C., Cohen, A. B., Gruber, J., Shiota, M. N., Haidt, J., & Keltner, D.
(2009). Resting respiratory sinus arrhythmia is associated with tonic
positive emotionality. Emotion, 9, 265–270.
Panksepp, J. (1998). Affective neuroscience: The foundations of human and
animal emotions. New York: Oxford University Press.
Pellegrini, A. D., & Smith, P. K. (Eds.). (2005). The nature of play: Play
in great apes and humans. New York: Guilford Press.
Porges, S. W. (1997). Emotion: An evolutionary by-product of the neural
regulation of the autonomic nervous system. In C. S. Carter, I. I.
Lederhendler, & B. Kirkpatrick (Eds.), The integrative neurobiology of

affiliation. Annals of the New York Academy of Sciences, 807, 62–77.
New York: New York Academy of Sciences.
Porges, S. W. (2007). The polyvagal perspective. Biological Psychology,
74, 116 –143.
Rozin, P., Haidt, J., & McCauley, C. R. (1999). Disgust: The body and soul
emotion. In T. Dalgliesh & M. J. Power (Eds.), Handbook of cognition
and emotion (pp. 429 – 445). New York: Wiley.
Scherer, K. R. (2001). Appraisal considered as a process of multilevel
sequential checking. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.),
Appraisal processes in emotion: Theory, methods, research. Series in
affective science (pp. 92–120). New York: Oxford University Press.
Shaver, P. R., Morgan, H. J., & Wu, S. (1996). Is love a basic emotion?
Personal Relationships, 3, 81–96.
Shiota, M. N., Keltner, D., Campos, B., & Hertenstein, M. (2004). Positive
emotion and the regulation of interpersonal relationships. In P. Phillipot
& R. Feldman (Eds.), Emotion regulation (pp. 127–156). Mahwah, NJ:
Erlbaum.
Shiota, M. N., Keltner, D., & John, O. P. (2006). Positive emotions
differentially associated with Big Five personality and attachment relationships. Journal of Positive Psychology, 1, 61–71.
Shiota, M. N., Keltner, D., & Mossman, A. (2007). The nature of awe:
Elicitors, appraisals, and effects on self-concept. Cognition and Emotion, 21, 944 –963.
Sinha, R., Lovallo, W. R., & Parsons, O. A. (1996). Cardiovascular
differentiation of emotions. Psychosomatic Medicine, 54, 424 – 435.
Smith, P. K. (1982). Does play matter? Functional and evolutionary aspects
of animal and human play. Behavioral and Brain Sciences, 5, 139 –184.
Sokolov, E. N. (1990). The orienting response, and future directions of its
development. Integrative Psychological and Behavioral Science, 25,
142–150.
Stekelenburg, J. J., & van Boxtel, A. (2002). Pericranial muscular, respiratory, and heart rate components of the orienting response. Psychophysiology, 39, 707–722.
Stemmler, G. (1989). The autonomic differentiation of emotions revisited:
Convergent and discriminant validation. Psychophysiology, 26, 617– 632.
Stemmler, G., Grossman, P., Schmid, H., & Foerster, F. (1991). A model
of cardiovascular activation components for studies using autonomic
receptor antagonists. Psychophysiology, 28, 367–382.
Stephens, C. L., Christie, I. C., & Friedman, B. H. (2010). Autonomic
specificity of basic emotions: Evidence from pattern classification and
cluster analysis. Biological Psychology, 84, 463– 473.
Tomaka, J., Blascovich, J., Kibler, J., & Ernst, J. M. (1997). Cognitive and
physiological antecedents of threat and challenge appraisal. Journal of
Personality and Social Psychology, 73, 63–72.
Tooby, J., & Cosmides, L. (2008). The evolutionary psychology of the
emotions and their relationship to internal regulatory variables. In M.
Lewis, J. M. Haviland-Jones, & L. F Barrett (Eds.), Handbook of
emotions (2nd ed., pp. 114 –137). New York: Guilford Press.
Tracy, J. L., & Robins, R. W. (2004). Show your pride: Evidence for a
discrete emotion expression. Psychological Science, 15, 194 –197.
Tsai, J. L., Chentsova-Dutton, Y., Freire-Bebeau, L., & Przymus, D. E.
(2002). Emotional expression and physiology in European Americans
and Hmong Americans. Emotion, 2, 380 –397.
Tugade, M. M., & Fredrickson, B. L. (2004). Resilient individuals use
positive emotions to bounce back from negative emotional experiences.
Journal of Personality and Social Psychology, 86, 320 –333.
Vrana, S. R. (1993). The psychophysiology of disgust: Differentiating negative
emotional contexts with facial EMG. Psychophysiology, 30, 279 –286.

Received August 3, 2010
Revision received February 18, 2011
Accepted February 23, 2011 䡲