Access to this full-text is provided by Springer Nature.
Content available from Scientific Reports
This content is subject to copyright. Terms and conditions apply.
1
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
www.nature.com/scientificreports
Functionally distinct smiles elicit
dierent physiological responses in
an evaluative context
Jared D. Martin1, Heather C. Abercrombie2, Eva Gilboa-Schechtman3 & Paula M. Niedenthal1
When people are being evaluated, their whole body responds. Verbal feedback causes robust activation
in the hypothalamic-pituitary-adrenal (HPA) axis. What about nonverbal evaluative feedback? Recent
discoveries about the social functions of facial expression have documented three morphologically
distinct smiles, which serve the functions of reinforcement, social smoothing, and social challenge. In
the present study, participants saw instances of one of three smile types from an evaluator during a
modied social stress test. We nd evidence in support of the claim that functionally dierent smiles
are sucient to augment or dampen HPA axis activity. We also nd that responses to the meanings of
smiles as evaluative feedback are more dierentiated in individuals with higher baseline high-frequency
heart rate variability (HF-HRV), which is associated with facial expression recognition accuracy. The
dierentiation is especially evident in response to smiles that are more ambiguous in context. Findings
suggest that facial expressions have deep physiological implications and that smiles regulate the social
world in a highly nuanced fashion.
Sweaty palms, a racing heart, a faltering voice. Most people nd the evaluative context of public speaking unpleas-
ant. Indeed, the mere anticipation of social evaluation increases activity across almost all body systems related to
stress, with particularly robust activation in the hypothalamic-pituitary-adrenal (HPA) axis1–4. However, scientic
inquiry has largely been limited to investigations of the manner in which the body responds to verbal evaluative
feedback, of the type “that was/wasn’t good.”5,6 Does the HPA axis respond to purely nonverbal feedback, such as
facial expression? We investigate this question and demonstrate that evaluators’ smiles are sucient to augment
or dampen HPA axis activity – depending upon the distinct meaning of the smile in the social-evaluative con-
text. Furthermore, we nd that physiological responses to smile meaning are most dierentiated in individuals
with higher baseline high-frequency heart rate variability (HF-HRV), which is associated with facial expression
recognition accuracy7–9.
Nonverbal feedback in social-evaluative contexts should be at least as impactful as verbal feedback, since non-
verbal signals are experienced by perceivers to be spontaneous reactions10 and thus, honest reections of internal
evaluations. In the few studies that have investigated the nonverbal communication of evaluative feedback, par-
ticipants whose audience displayed smiles and other nonverbal expressions of positive evaluation showed lower
physiological activity than participants met with frowns and similar cues to negative evaluation11–13. Recent the-
ory and empirical evidence suggests, however, that smiles do not all communicate identical, uniformly positive
messages14–16. Instead, evidence supports the existence of at least three morphologically distinct types of smiles,
each of which serves a dierent social function necessary for successful group living: “reward” smiles reinforce
behavior, “aliation” smiles signal lack of threat and facilitate or maintain social bonds, and “dominance” smiles
assert claims to higher status in social hierarchies15,16. In light of their social functions, each of these three smiles
should carry distinct meanings when displayed by evaluators. e rst aim of the present research was to test the
hypothesis that reward, aliation, and dominance smiles, delivered as evaluative feedback, inuence perceivers’
HPA axis activity in a manner congruent with their distinct social meaning. We expected dominance smiles,
compared to reward smiles, to increase HPA axis activity. In theory, whereas dominance smiles show disdain for
and reward smiles show approval of performance, aliation smiles reassure without being indicative of a specic
evaluation and so were expected to buer HPA axis activity to a lesser degree than reward smiles.
1University of Wisconsin-Madison, Department of Psychology, 1202 W. Johnson St., Madison, WI, 53706, United
States. 2University of Wisconsin-Madison, Department of Psychiatry, 6001 Research Park Blvd, Madison, WI, 53719,
United States. 3Bar-Ilan University, Department of Psychology, Ramat-Gan, 52900, Israel. Correspondence and
requests for materials should be addressed to J.D.M. (email: jdmartin7@wisc.edu)
Received: 12 October 2017
Accepted: 6 February 2018
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
2
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
e second aim of the present research was to account for variability in participants’ capacity to understand
the evaluative meanings of each smile. Reward smiles are more unambiguous in meaning across contexts than
either aliation or dominance smiles14, which are relevant in more limited contexts. at is, reward smiles can
reinforce widely varying behaviors in most any social situation, whereas aliation smiles serve to smooth exist-
ing or potential social bonds and dominance smiles serve to challenge social standing. Individuals who are more
accurate at recognizing facial expressions should exhibit more dierentiated physiological activity in response to
aliation and dominance smiles, indicating greater sensitivity to their social-evaluative meanings. Recent empir-
ical evidence suggests that baseline HF-HRV – an index of parasympathetic nervous system activity at the level of
the heart – is positively associated with facial expression recognition accuracy7–9. We thus tested the hypothesis
that individuals with higher baseline HF-HRV exhibit more dierentiated physiological activity in response to
smiles presented as evaluative feedback, particularly in response to smiles that are more ambiguous in context
(i.e., aliation and dominance smiles).
In an adaptation of the classic Trier Social Stress Test17, male participants (N = 90) extemporaneously
addressed three topics about themselves in front of a male evaluator who watched them over a web camera. e
evaluator was, in fact, one of two confederates working for the research team; approximately half of the partici-
pants (Ns = 47 and 43) interacted with each confederate. To increase believability, the evaluator briey appeared
live on the computer screen, then turned o his web camera for the remainder of the session. Aer responding
to each of the three topics, participants saw videos of their evaluator’s facial expressions, which they believed
represented spontaneous reactions to their performance that had been extracted by facial recognition soware.
e videos were, in fact, pre-recorded. Participants were assigned to one of three smile conditions such that they
saw either one reward, aliation, or dominance smile aer each of their responses to the three topics. Along with
one smile video, participants also saw a control video that showed the evaluator making a neutral response such
as face scratching or eye blinks. us, each participant was exposed to six videos of their evaluator in total (three
dierent instances of one type of smile, three neutral videos) with one smile and one neutral video presented
aer each of the three responses. Smile videos were constructed to meet a priori specications of morphological
activity associated with each of the three smile types (for further details, see Supplementary Materials)14. us,
smile type was the only feature of the evaluative feedback that varied between conditions. Physiological activity, in
both the HPA axis and cardiovascular system, was assessed throughout the study; salivary cortisol was measured
at seven time points, and a continuous electrocardiograph was collected before, during, and aer the speech task.
Results
Raw means for total salivary cortisol level by smile condition were consistent with the prediction that reward, al-
iation, and dominance smiles inuence perceivers’ physiological activity in distinctive ways, such that the receipt
of dominance smiles is associated with higher HPA axis activity relative to thereceipt of reward smiles. (AUCi
– nmol/l: dominance: M = 19.4, SD = 24.74; aliation: M = 2.43, SD = 22.3; reward: M = 1.21, SD = 21.54). We
used dummy-coded condition contrasts to directly compare total salivary cortisol responses between smile types
(“AUCi”, see Methods, below). Compared to reward smiles, dominance smiles induced a greater overall salivary
cortisol response (b = 18.18, t (86) = 2.93, p = 0.004, CI(95%) = [5.83, 30.54], Δr2 = 0.087). Similarly, even though
aliation smiles do not signal a clear social evaluation in this context, compared to such smiles, dominance
smiles induced a greater overall salivary cortisol response (b = 16.97, t (86) = 2.92, p = 0.004, CI(95%) = [5.41,
28.53], Δr2 = 0.086).
Further corroborating the nding that dominance smiles induce greater HPA axis activity, participants receiv-
ing reward or aliation smiles returned to their individual cortisol baseline by 30-minutes post-speech, whereas
those who received dominance smiles continued to have signicantly higher cortisol levels than their individ-
ual baseline. In this ancillary analysis, cortisol values at 20- and 30-minutes post-speech were predicted from
dummy-coded condition contrasts and average baseline cortisol level. Intercepts for each smile condition at both
20-minutes post-speech (reward, b = 0.23, t (86) = 1.49, p = 0.14, CI(95%) = [−0.08, 0.53]; aliation, b = 0.25, t
(86) = 1.79, p = 0.08, CI(95%) = [−0.03, 0.53]; dominance, b = 0.57, t (86) = 4.47, p < 0.0001, CI(95%) = [0.31,
0.82]) as well as 30-minutes post-speech (reward, b = −0.02, t (86) = −0.13, p = 0.9, CI(95%) = [−0.32, 0.28];
aliation, b = 0.03, t (86) = 0.2, p = 0.84, CI(95%) = [−0.25, 0.31]; dominance, b = 0.36, t (86) = 2.82, p = 0.006,
CI(95%) = [0.11, 0.61]) show that mean salivary cortisol values for the dominance group continued to be signif-
icantly greater than zero up to 30-minutes post-speech, which was not the case for the other groups. is shows
that individuals receiving dominance smiles take signicantly longer to return to their individual cortisol base-
line, thus corroborating ndings from the AUCi analysis.
We also tested the prediction that baseline HF-HRV is positively associated with dierentiation in HPA axis
activity in response to aliation and dominance smiles. We again created dummy-codes for smile feedback con-
dition in order to compare the eect of smile feedback between conditions. We entered these dummy-codes
along with a continuous measure of baseline HF-HRV and the dummy-code by HF-HRV interaction terms into
a linear regression model. Results from this analysis conrmed our expectations. With those who received dom-
inance smiles as the dummy-coded reference group, baseline HF-HRV was positively correlated with salivary
cortisol responses (b = 9.72, t (84) = 2.02, p = 0.046, CI(95%) = [0.14, 19.29], Δr2 = 0.04). Comparing the linear
association between baseline HF-HRV and salivary cortisol responses between individuals who received domi-
nance versus aliation smiles, the comparison of simple slopes was signicant (b = 14.63, t (84) = 2.40, p = 0.018,
CI(95%) = [2.54, 26.83], Δr2 = 0.057), indicating that the relationship between baseline HF-HRV and salivary
cortisol is more positive for those receiving dominance compared to aliation smiles as evaluative feedback
(Fig.1).
Convergent evidence that baseline HF-HRV is positively associated with greater dierentiation of phys-
iological responses to affiliation and dominance smiles comes from cardiovascular data. Again, we created
dummy-coded condition contrasts and entered them into a linear regression model along with baseline HF-HRV
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
3
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
and the HF-HRV by smile feedback dummy-code interaction terms. With the dummy-coded condition contrasts
referenced on participants who received aliation smiles, baseline HF-HRV was negatively associated with heart
rate during the speech task (b = −5.84, t (84) = −2.85, p = 0.006, CI(95%) = [−9.91, −1.76], Δr2 = 0.086). A
comparison of the linear association between baseline HF-HRV and heart rate between individuals who received
aliation versus dominance smiles reveals that the relationship between baseline HF-HRV and heart rate was
more negative for those exposed to aliation compared to dominance smiles as evaluative feedback (b = −7.28,
t (84) = 2.19, p = 0.032, CI(95%) = [−13.92, −0.66], Δr2 = 0.051).
Since indicators of HF-HRV reactivity (i.e., baseline – task) have also been associated with facial expres-
sion recognition accuracy18, we conducted ancillary analyses of the association between reactivity measures of
HF-HRV and perceivers’ physiological activity in response to smile feedback. Specically, we calculated a dif-
ference score between task level HF-HRV and baseline by subtracting task HF-HRV, assessed during the rst
2.5 minutes of the speech, from baseline HF-HRV. Higher values on this dierence score reect greater with-
drawal of vagal inuence over the heart during task, compared to baseline. We entered the dierence score into
a model similar to the one employed in the cortisol and heart rate analyses, entering dummy-coded condition
contrasts along with the continuous HF-HRV dierence score and the condition by dierence score interactions.
Table1 reports these ndings. HF-HRV reactivity was positively associated with physiological responses to social
evaluation, regardless of the type of smile feedback a participant received (i.e., HF-HRV reactivity did not interact
with experimental condition: see Table1). ese ndings strengthen the conclusion that higher baseline HF-HRV
is associated with more dierentiated physiological responses to smiles as evaluative feedback.
Analyses were conducted in the “R” statistical environment19. All results reported in the present work remain
signicant when adding a dichotomous predictor to account for which of the two confederates a participant inter-
acted with, as well as when statistically accounting for factors known to inuence levels of physiological activity
(caeine use, depression severity).
Discussion
In the present study, we observed that smiles with dierent social functions16, when delivered as evaluative feed-
back in a stressful social context, exert distinct inuences on perceivers’ physiological activity. Dominance smiles
Figure 1. Physiological Response to Smiles as Evaluative Feedback Depends on Baseline High Frequency Heart
Rate Variability. Salivary Cortisol: Total salivary cortisol (nmol/l: dominance: M = 19.4, SD = 24.74, N = 27;
aliation: M = 2.43, SD = 22.3, N = 36; reward: M = 1.21, SD = 21.54, N = 27) in response to social evaluation
was greater for those receiving dominance smiles as evaluative feedback relative to the two other types of smiles.
e dierence in total salivary cortisol response between the aliation and dominance groups increased as HF-
HRV increased. Heart Rate: Heart rate assessed during the speech task (bpm: dominance: M = 83.51, SD = 9.84,
N = 27; aliation: M = 82.13, SD = 14.43, N = 36; reward: M = 85.93, SD = 12.65, N = 27) was not signicantly
dierent between conditions. e dierence in heart rate between the dominance and aliation groups
increased as HF-HRV increased. Dotted lines between the +/− 1 SD bars indicate a statistically signicant
simple slope for baseline HF-HRV in that feedback condition; solid lines are not signicant below the 0.05 level.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
4
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
(compared to reward or aliation smiles) were associated with increases in heart rate and salivary cortisol that
mirror the inuences of negative verbal feedback20. In contrast, reward and aliation smiles exerted inuences
similar to the eects of displays of friendliness21 and positive social evaluation12, such that, compared to domi-
nance smiles, they buered physiological activity. e ndings thus provide further evidence for the view that
smiles do not constitute a homogenous category of “positive” nonverbal feedback. e ndings also contribute
to abody of literature that proposes a role for cortisol in adaptive responding to the social environment22. In
particular, cortisol appears to support the detection of social threat and coordinate biological activity needed to
adequately respond to the threat.
e present research also contributes to a growing literature on individual dierences in sensitivity to the
meaning of facial expression. Specically, baseline HF-HRV was negatively related to participants’ physiologi-
cal activity when they received aliation smiles, and was positively related to their physiological activity when
they received dominance smiles as evaluative feedback. e observed relationship between HF-HRV and phys-
iological responsiveness to specic smile meaning is consistent with a claim of the Neurovisceral Integration
Model according to which HF-HRV indexes an individual’s capacity to respond appropriately to social stimuli23.
Conversely, individuals with particularly low baseline HF-HRV may be at risk of poor socio-emotional outcomes
due to decits in both understanding as well as responding to the social signals of others. A number of pathologi-
cal and pre-disease body states are related to lower HF-HRV, including heightened inammation and obesity24,25.
As such, these body states may fundamentally change an individual’s ability to respond physiologically to the
social world, and in light of the present ndings, may be associated with decits in understanding the expressions
of others. A compelling avenue for future research therefore is to investigate how individuals with relatively low
HF-HRV recognize and respond to the expressions of others.
Future research should also consider how aective disorders associated with lower HF-HRV, such as anxiety26
and depression27, relate to cognitive and physiological responses to social stimuli. Given the social stress proce-
dure implemented in the present work, an especially fruitful avenue for future research on the social functions
of smiles lies in the relationship between smile processing and social anxiety. Social anxiety disorder involves the
fear of embarrassing oneself in front of others in performance or evaluative situations28. Although research con-
sistently documents that individuals with social anxiety disorder exhibit heightened negative aect both in antici-
pation of and in response to social evaluation29,30, ndings are mixed31,32 regarding how these individuals respond
to positive and supportive social signals. Social anxiety was not accounted for in the present study. Although
randomization into condition likely mitigated any eect of social anxiety on the present ndings, future work is
needed to investigate how individuals with social anxiety respond to facial expressions as evaluative feedback. A
crucial comparison in this work will be to assess dierences in aective and physiological responses between the
relatively unambiguous reward smiles and the less clear aliation smiles.
One limitation of the present research is that the experimental design contained no “neutral” feedback con-
dition. It would be hard to create such an experience, since neutral feedback is interpreted in highly varied ways
across individuals. However, methods to create such a control condition could be imagined for future research.
Along similar lines, future research could also directly compare the eects of receiving dominance smiles with the
eects of receiving more overt signals of negative evaluation such as disgust, contempt, and anger. By employing
the present paradigm to test the physiological eects of facial expressions beyond the smiles tested here, research-
ers could not only more clearly describe the unique eects of receiving social functional smiles but also under-
stand the eects of facial expressions on perceivers’ physiological activity more generally.
e present work includes other limitations that warrant comment, two of which concern the participant sam-
ple. First, the sample size in this study may have been relatively small given the size of the eects detected. Since,
to our knowledge, no work has explored the eect of receiving social functional smiles as social evaluation, a pri-
ori power analyses were dicult to conduct. We estimated required sample size from extant studies with methods
and aims as similar to the present research as possible12. Following these guidelines, 90–100 participants is typical
for a study of this nature involving three between-subjects conditions. Post-hoc power analyses using G*Power33
indicate that achieved power was modest for the critical comparisons within the HPA axis (dominance vs. alia-
tion: β = 0.72; dominance vs. reward: β = 0.81). In light of the achieved power, we recommend that future studies
of this sort rely on no fewer than 35–40 participants per between-subjects condition to ensure adequate power.
(N = 90) b (s.e.) vs. Aliation vs. Dominance
Reward
Cortisol (AUCi)7.04 (4.39) 0.95 (5.60) 1.30 (5.83)
Heart R ate 6.44** (2.31) −0.32 (2.94) −1.97 (3.06)
Aliation
Cortisol (AUCi) 7.99* (3.46) —0.36 (5.16)
Heart R ate 6.12** (1.82) —−1.65 (2.71)
Dominance
Cortisol (AUCi) 8.35* (00.0) —
Heart R ate 4.47* (2.01) —
Table 1. Associations Between HF-HRV Reactivity and Physiological Responses to Social Evaluation.
Unstandardized regression coecients quantifying the association between HF-HRV reactivity (baseline – task)
and physiological responses are reported in the le-hand column, for each group. Dierences between groups
are reported in the middle and right-hand columns. *p < 0.05; **p < 0.01; ***p < 0.001.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
5
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
Second, the present sample was restricted to men. Although we limited the sample to males for reasons that
were established and justied before the research was conducted (see Methods, below), generalizations from our
ndings are necessarily limited. Since sex eects are observed in research on the perception of facial expression
of emotion34,35, future work should determine what portion (if any) of the currently documented eects are con-
tingent upon the sex of the smiler or perceiver. Some work suggests that men respond more to threats of physical
aggression whereas women respond more to condescending behaviors and social aggression36. us, men and
women may respond to the same type of smile in dierent ways and may also use each of the smile types with
varying degrees of frequency37.
e present research demonstrates that functionally dierent smiles are sucient to augment or dampen HPA
axis activity in accordance with the social functional meaning associated with each smile. Furthermore, physi-
ological responses to each functionally distinct smile are most dierentiated in individuals with higher baseline
HF-HRV, suggesting that HF-HRV is a useful individual dierence moderator of the ability to respond to the
social environment in a exible and nuanced fashion. We may therefore conclude that smiles coordinate the phys-
iological activity that supports interpersonal encounters to a previously undocumented degree, and that facial
expressions help regulate the social world, in part, through their impact on the physiological activity of perceivers.
Methods
Participants. Ninety-two male undergraduates at a large university in the Midwest participated in exchange
for credit in an introductory Psychology course. Participants provided written consent, indicating full under-
standing of the requirements for participation. e research protocol was reviewed and approved under the
University of Wisconsin – Madison Institutional Review Board. All research was conducted in accordance with
institutional guidelines and regulations.
Due to the robust sex dierences in cortisol responses to laboratory stressors38 and the variability in corti-
sol responses introduced by oral contraceptive use39, only males were invited to participate in this study. Since
females generally outperform males in the accurate recognition of positive facial expressions40, the expectation
was that the present study would underestimate the eects of smiles on physiological responses. Pre-inclusion cri-
teria limited participation to U.S.-born, English-speaking males without a diagnosed heart condition and not cur-
rently taking medications that alter hormone levels. Participants were instructed to refrain from exercise on the
day of the study and to avoid alcohol and caeine consumption within twenty-four hours of their participation.
Due to a network failure, data collection from one participant was terminated before experimental manipulation.
Furthermore, data from a second participant were excluded from analysis due to the presence of an abnormal
heart rhythm resembling premature beats41 which made it dicult to score the data and conicted with the
pre-inclusion criterion of cardiac health. e exclusion of all data from these two participants le a nal sample
of ninety participants (dominance: N = 27; aliation: N = 36; reward N = 27).
Data collection was limited to the number of participants that could be involved during one academic semes-
ter, not to exceed 120 participants (40 per condition). Participants were randomly assigned to one of three smile
experimental conditions as well as one of two confederates. Given that some participants did not show up for
their assigned experimental sessions, a certain amount of imbalance between the number of participants in each
condition and assigned to each confederate is to be expected.
At the conclusion of the study, participants underwent a funneled debrieng. First, they were asked if they
thought they knew what the study was about. In this general interview, no participants brought up suspicions
about deception. We then asked participants if they found anything strange about the study or were suspicious of
anything. In a logistic regression model with experimental condition and confederate as predictors of a dichoto-
mous (“yes”/“no”) suspicion outcome, no signicant dierences were detected by experimental condition (all ps
> 0.2) or by confederate (p = 0.93). Participants who thought the study was slightly strange indicated that they
were unsure how the ller video was related to the study, were unaccustomed to providing saliva samples, or were
uncomfortable giving speeches.
Procedure. In order to reduce variation due to diurnal changes in cortisol levels, experimental sessions took
place in the aernoon. Upon arriving at the experimental laboratory, the participant encountered another male
“participant” who was actually a confederate—one of the two whose stimuli were validated in a separate study (see
Supplemental Material for further information). e experimenter then entered from a nearby room and told the
two men that he had randomly assigned them to dierent tasks in the study: the participant was always assigned
to “give the speech” and the confederate was always assigned to “judge the speech.” Aer the participant and
confederate had provided informed consent, the participant supplied the rst of seven saliva samples (collection
method described below) and completed an online questionnaire assessing his compliance with pre-restriction
criteria (alcohol and caeine use) as well as other medical information (prescription and recreational drug use).
e men were told that only the person giving the speech (always the participant and never the confederate) had
to answer the online questionnaires and provide the saliva samples because it directly pertained to giving the
speech; the confederate waited with the participant and the experimenter while the participant responded to the
surveys and provided the saliva sample.
The experimenter then demonstrated a facial expression recognition software, the Computer Emotion
Recognition Toolbox42. e participant and confederate were informed that the program could extract mean-
ingful facial expressions from live video feed. e experimenter used the computer’s built-in web camera and the
participant’s face as the live feed in a demonstration in order to increase believability in the soware’s (actual)
capabilities.
Next, the participant and confederate were separated. Leaving the confederate behind in the initial room,
the experimenter mentioned that a second experimenter would arrive shortly to provide the confederate with
further instructions and tasks. e experimenter escorted the participant to a psychophysiology lab in the same
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
6
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
building. ere, the experimenter attached sensors to the participant’s chest and explained that they would be
used to measure aspects of his cardiovascular reactivity. e experimenter sat with the participant in the experi-
mental room while a research assistant in an adjacent control room recorded a 3-minute baseline measure of the
participant’s heart activity.
Aer completion of this baseline recording, the experimenter informed the participant that he would deliver
his speech to the “other participant” via Skype. He would not see the evaluator as his web camera would be turned
o in order to avoid distraction. e format of the speech involved answering three questions in sequence, with
two minutes to respond to each question. e participant was also told that at the conclusion of every two-minute
speech period, he would see videos of several of the evaluator’s facial expressions. e participant was told that
the videos had been “randomly extracted by the facial expression soware” while the evaluator watched the par-
ticipant’s speech. is led the participant to believe that the videos conveyed authentic evaluative responses by the
confederate. e facial expressions displayed in these videos constituted the experimental manipulation.
When the participant indicated that he had understood the speech task, the experimenter gave him a sheet
of paper with the three questions he was required to answer, and then sat with the participant for three min-
utes while the participant prepared his speech. A continuous cardiovascular recording was taken during this
three-minute “anticipation” period. Aer the time was over, the second saliva sample was taken.
Next, the experimenter launched Skype. In order to enhance believability, the confederate appeared live on
the participant’s screen and waved “hello.” e experimenter asked the evaluator to turn o his camera “so as
not to distract” the participant during the speech. With the participant no longer able to see the evaluator, the
experimenter asked the participant to begin his speech. Participants responded to each question in order, with
two minutes for each question: (1) What makes you happy? (2) What do you like most and least to eat?, and (3)
What is your favorite part of living in Madison, Wisconsin? ese questions were designed to be personal and to
contain enough positive material to make the smiles sent by the evaluator appear plausible.
At the conclusion of each two-minute speech period, the experimenter stopped the participant and showed
him two videos of the evaluator that were “randomly extracted” when he was listening to the participant’s speech.
From the other room, the confederate dropped each video into a network-basedfolder in order to simulate the
facial expression recognition program extracting and sending the videos in real time. Each participant saw 6 vid-
eos in total, 2 aer each question. 3 of the 6 total videos were smiles and the other 3 were a set of dierent neutral
videos—evaluators faced the camera with a neutral expression and madeoccasional small, non-evaluative move-
ments such as face scratching. e smile videos were three dierent short video clips of the same confederate
making only one smile type (dominance, reward, or aliation). Participants were thus exposed to three dierent
examples of the same type of smile. In sum, smile type was manipulated between-subjects with the 3 neutral
videos retained across participants. At the conclusion of the speech task, the third saliva sample was taken. e
speech task lasted between 7–8 minutes, during which a continuous cardiovascular recording was taken.
Immediately upon concluding his nal speech, the participant was directed to reect on his performance,
focusing on how he felt and what his evaluator thought. A continuous cardiovascular recording was taken during
this ve-minute reection period. Aer the reection/recovery period, the participant was detached from the
sensors and led to the nal room by a second experimenter who was blind to the participant’s video feedback
condition.
In the last room, the participant watched a ller video available on YouTube from the series “e Life of Birds”
(available at https://www.youtube.com/playlist?list=PLB1F251E81DE15E9B) which provided a neutral experi-
ence during which cortisol recovery was assessed. e video was the same for all participants. e remaining four
saliva samples were taken at 10-minute intervals from the cessation of the speech task. At the conclusion of the
ller video, and aer completing verbal questions assessing deception suspicion, the participant was debriefed
and dismissed.
Physiological Measures: Collection and Analysis. Cortisol. Saliva samples were obtained with cot-
ton salivettes (Fisher Scientic Company, LLC). Participants were instructed to let the cotton salivette touch all
parts of their mouth (under their tongue, between their teeth and their cheeks) without chewing on it. Saliva
collection was strictly timed for two minutes, aer which the sample was returned to its plastic casing. Samples
were frozen aer collection and stored at −20 C. At the conclusion of the study, samples were express shipped to
Dresden, Germany where they were single-assayed at the lab of Dr. Clemens Kirschbaum (T.U. Dresden). Samples
were assayed using the chemi-luminescence assay, which has a high sensitivity of 0.16 ng/mL (IBL-International,
Hamburg, Germany) and intra and interassay CVs of <10%. In total, saliva samples were collected at seven
time points during the study and assayed for unbound cortisol. Due to skewness, all cortisol values were rst
log-transformed. Salivary alpha amylase was assayed but is not reported in these analyses.
In order to test our hypotheses with regard to the HPA axis, Area Under the Curve with respect to increase
(AUCi)43 values were calculated for the cortisol response of each participant. AUCi scores index total cortisol
response over a given period of time, referenced to each individual’s baseline cortisol level. We averaged the two
cortisol values collected before experimental manipulation (receipt of smile feedback) as a pre-speech baseline.
Heart Rate. Continuous EKG recordings were sampled at 1000 Hz via one of the bipolar inputs available on
the SynAmps2 Headbox (Compumedics Neuroscan Ltd., U.S.A). Ag/AgCl spot electrodes were placed in a
thoracic-modied lead-II conguration to maximize detection of R-spikes while minimizing movement arti-
facts. We calculated mean heart rate values separately for the baseline period, the “anticipation” period, the speech
period, and the post-speech period.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
7
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
Data Processing. EKG data were rst scored oine using OpenANSLAB44, manually inspected for artifacts, and
the resultant inter-beat-interval series were extracted and saved. CMETx soware (available at http://apsychos-
erver.psych.arizona.edu) was then used on the extracted inter-beat-interval to quantify HF-HRV.
References
1. Dicerson, S. S., Myce, P. J. & Zaldivar, F. Negative social evaluation, but not mere social presence, elicits cortisol responses to a
laboratory stressor tas. Health Psychol. 27, 116–21 (2008).
2. Dicerson, S. S., Gruenewald, T. L. & emeny, M. E. When the social self Is threatened: shame, physiology, and health. J. Pers. 72,
1191–1216 (2004).
3. Bosch, J. A. et al. A general enhancement of autonomic and cortisol responses during social evaluative threat. Psychosom. Med. 71,
877–885 (2009).
4. Blachart, G. C., Ecel, L. A. & Tice, D. M. Salivary cortisol in response to acute social rejection and acceptance by peers. Biol.
Psychol. 75, 267–276 (2007).
5. Fallon, M. A., Careaga, J. S., Sbarra, D. A. & Connor, M. O. Utility of a virtual trier social stress test: initial ndings and benchmaring
comparisons. Psychosom. Med. 78, 835–840 (2016).
6. Eisenberger, N. I., Inagai, T. ., Muscatell, . A., Haltom, . E. B. & Leary, M. . e neural sociometer: brain mechanisms
underlying state self-esteem. J. Cogn. Neurosci. 23, 3448–3455 (2011).
7. Quintana, D. S., Guastella, A. J., Outhred, T., Hicie, I. B. & emp, A. H. Heart rate variability is associated with emotion recognition:
direct evidence for a relationship between the autonomic nervous system and social cognition. Int. J. Psychophysiol. 86, 168–172
(2012).
8. Lische, A., Leme, D., Neubert, J., Hamm, A. O. & Lotze, M. Inter-individual dierences in heart rate variability are associated with
inter-individual dierences in mind- reading. Sci. ep. 7, 11557, https://doi.org/10.1038/s41598-017-11290-1 (2017).
9. Sellaro, ., de Gelder, B., Finisguerra, A. & Colzato, L. S. Transcutaneous vagus nerve stimulation (tVNS) enhances recognition of
emotions in faces but not bodies. Cortex 23, 213–223 (2017).
10. Burgoon, J. ., Guerrero, L. ., & Manusov, V. In e SAGE handboo of interpersonal communication (SAGE, 2011).
11. assam, . S., oslov, . & Mendes, W. B. Decisions under distress: stress proles inuence anchoring and adjustment. Psychol. Sci.
20, 1394–1399 (2009).
12. Ainola, M. & Mendes, W. B. e dar side of creativity: biological vulnerability and negative emotions lead to greater artistic
creativity. Personal. Soc. Psychol. Bull. 34, 1677–1686 (2008).
13. Taylor, S. E. et al. Eects of a supportive or an unsupportive audience on biological and psychological responses to stress. J. Personal.
Soc. Psychol. 98, 47–56 (2010).
14. ychlowsa, M. et al. Functional smiles: tools for love, sympathy, and war. Psychol. Sci. 28, 1259–1270 (2017).
15. Martin, J. D., Wood, A., ychlowsa, M. & Niedenthal, P. M. Smiles as multipurpose social signals. Trends Cogn. Sci. 21, 861–877
(2017).
16. Niedenthal, P. M., Maringer, M. & Hess, U. e simulation of smiles (SIMS) model: embodied simulation and the meaning of facial
expression. Behav. Brain Sci. 33, 417–430 (2010).
17. irschbaum, C., arl-Martin, P. & Hellhammer, D. e ‘Trier Social Stress Test’ - a tool for investigating psychobiological stress
responses in a laboratory setting. Neuropsychobiology 28, 76–81 (1993).
18. Muhtadie, L., oslov, ., Ainola, M. & Mendes, W. B. Vagal exibility: a physiological predictor of social sensitivity. J. Personal. Soc.
Pscyhology 109, 106 (2014).
19. Core Team. : A language and environment for statistical computing. Foundation for Statistical Computing, Vienna, Austria.
http://www.-project.org/ (2014).
20. udiela, B. M. & Wüst, S. Human models in acute and chronic stress: assessing determinants of individual hypothalamus-pituitary-
adrenal axis activity and reactivity. Stress 13, 1–14 (2010).
21. Wiemers, U. S., Schoofs, D. & Wolf, O. T. A friendly version of the trier social stress test does not activate the HPA axis in healthy
men and women. Stress 16, 254–260 (2013).
22. Dicerson, S. S., Gruenewald, T. L. & emeny, M. E. Psychobiological responses to social self threat: functional or detrimental? Self
Identity 8, 270–285 (2009).
23. Smith, ., ayer, J. F., halsa, S. S. & Lane, . D. e hierarchical basis of neurovisceral integration. Neurosci. Biobehav. ev. 75,
274–296 (2017).
24. C ooper, T. M. et al. Heart rate variability predicts levels of inammatory marers: evidence for the vagal anti-inammatory pathway.
Brain. Behav. Immun. 49, 94–100 (2015).
25. Laborde, S., Mosley, E. & Thayer, J. F. Heart rate variability and cardiac vagal tone in psychophysiological research –
recommendations for experiment planning, data analysis, and data reporting. Front. Psychol. 8, 1–18 (2017).
26. Chalmers, J. A., Quintana, D. S., Abbott, M. J. & emp, A. H. Anxiety disorders are associated with reduced heart rate variability: a
meta-analysis. Front. Psychiatry 5, 1–11 (2014).
27. C arnevali, L., ayer, J. F., Brosschot, J. F. & Ottaviani, C. Heart rate variability mediates the lin between rumination and depressive
symptoms: A longitudinal study. Int. J. Psychophysiol, https://doi.org/10.1016/j.ijpsycho.2017.11.002 (2017).
28. Morrison, A. S. & Heimberg, . G. Social anxiety and social anxiety disorder. Annu. ev. Clin. Psychol. 9, 249–274 (2013).
29. apee, . M. & Lim, L. Discrepancy between self- and observer ratings of performance in social phobics. J. Abnorm. Psychol. 101,
728–731 (1992).
30. oelofs, . et al. Hypothalamus-pituitary-adrenal axis hyperresponsiveness is associated with increased social avoidance behavior
in social phobia. Biol. Psychiatry 65, 336–343 (2009).
31. Bucner, J. D., DeWall, C. N., Schmidt, N. B. & Maner, J. . A tale of two threats: social anxiety and attention to social threat as a
function of social exclusion and non-exclusion threats. Cognit. er. es. 34, 449–455 (2010).
32. Maner, J. ., DeWall, C. N., Baumeister, . F. & Schaller, M. Does social exclusion motivate interpersonal reconnection? esolving
the ‘porcupine problem’. J. Pers. Soc. Psychol. 92, 42–55 (2007).
33. Faul, F., Erdfelder, E., Buchner, A. & Lang, A.-G. Statistical power analyses using G*Power 3.1: Tests for correlation and regression
analyses. Behav. es. Methods 41, 1149–1160 (2009).
34. Hampson, E., van Anders, S. M. & Mullin, L. I. A female advantage in the recognition of emotional facial expressions: test of an
evolutionary hypothesis. Evol. Hum. Behav. 27, 401–416 (2006).
35. Vassallo, S., Cooper, S. L. & Douglas, J. M. Visual scanning in the recognition of facial aect: Is there an observer sex dierence? J.
Vis. 9, 1–10 (2009).
36. Bettencourt, B. A. & Miller, N. Gender dierences in aggression as a function of provocation: a meta-analysis. Psychol. Bull. 119,
422–447 (1996).
37. Fischer, A. & LaFrance, M. What drives the smile and the tear: why women are more emotionally expressive than men. Em ot. ev.
7, 22–29 (2014).
38. irschbaum, C., Wüst, S. & Hellhammer, D. Consistent sex dierences in cortisol responses to psychological stress. Psychosom. Med.
54, 648–657 (1992).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
8
Scientific REPORts | (2018) 8:3558 | DOI:10.1038/s41598-018-21536-1
39. ohleder, N., Wolf, J. M., Piel, M. & irschbaum, C. Impact of oral contraceptive use on glucocorticoid sensitivity of pro-
inammatory cytoine production aer psychosocial stress. Psychoneuroendocrinology 28, 261–273 (2003).
40. Donges, U. S., ersting, A. & Suslow, T. Women’s greater ability to perceive happy facial emotion automatically: gender dierences
in aective priming. PLoS One 7, e41745 (2012).
41. Goldberger, A., Goldberger, Z. & Shvilin, A. Goldberger’s clinical electrocardiography: a simplied approach. (Elsevier/Saunders,
2013).
42. Littlewort, G. et al. e computer expression recognition toolbox (CET) (2010).
43. Pruessner, J. C., irschbaum, C., Meinlschmid, G. & Hellhammer, D. H. Two formulas for computation of the area under the curve
represent measures of total hormone concentration versus time-dependent change. Psychoneuroendocrinology 28, 916–931 (2003).
44. Wilhelm, F. H. & Pey, P. ANSLAB: Autonomic nervous system laboratory (Version 4.0). Available at the SP Soware epository:
http://www.sprweb.org. (2005).
Acknowledgements
e authors would like to thank Craig Berridge, Allison Blumenfeld, Stephanie Carpenter, Crystal Hanson,
Judy Harackiewicz, Roxanne Hoks, Cathy Marler, and Wendy Berry Mendes for their assistance and helpful
comments. Magdalena Rychlowska and Adrienne Wood provided critical feedback during the revision process.
Stella Mayerho, Josh Lucas, Sarah Hervdejs, and Mathias Hibbard were instrumental in collecting data. We
also acknowledge the tireless contribution of our two confederates A.J. Laird and Zeus Markos. This work
was supported by the National Institutes of Mental Health [grant number T32MH018931-26 to J.D.M]; the
U.S. – Israeli Binational Science Foundation [grant number 2013205 to P.M.N. and E.G.S.]; and the National
Science Foundation [grant number 1355397 to P.M.N.]. Further support for this research was provided by the
Graduate School and the Oce of the Vice Chancellor for Research and Graduate Education at the University of
Wisconsin-Madison with funding from the Wisconsin Alumni Research Foundation. Sources of nancial support
had no inuence over the design, analysis, interpretation, or choice of submission outlet for this research. Data are
available by contacting the corresponding author.
Author Contributions
J.D.M., P.M.N., and H.C.A. designed the study. J.D.M. collected and analyzed data. J.D.M. wrote the rst dra of
this manuscript, and all authors contributed revisions.
Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-21536-1.
Competing Interests: e authors declare no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2018
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
Available via license: CC BY
Content may be subject to copyright.