While comparative research on contagious
yawning has grown substantially in the past few
years, both the interpersonal factors
influencing this response and the sensory
modalities involved in its activation in humans
remain relatively unknown. Extending upon
previous studies showing various in-group and
status effects in non-human great apes, we
performed an initial study to investigate how
the political affiliation (Democrat vs.
Republican) and status (high vs. low) of target
stimuli influences auditory contagious yawning,
as well as the urge to yawn, in humans.
Self-report responses and a subset of video
recordings were analyzed from 118 undergraduate
students in the US following exposure to either
breathing (control) or yawning (experimental)
vocalizations paired with images of former US
Presidents (high status) and their respective
Cabinet Secretaries of Commerce (low status).
The overall results validate the use of auditory
stimuli to prompt yawn contagion, with greater
response in the experimental than the control
condition. There was also a negative effect of
political status on self-reported yawning and
the self- reported urge to yawn irrespective of
the condition. In contrast, we found no evidence
for a political affiliation bias in this
response. These preliminary findings are
discussed in terms of the existing comparative
evidence, though we highlight limitations in the
current investigation and we provide suggestions
for future research in this area.
INTRODUCTION
Accumulating research over the past decade
has vastly improved our understanding of
contagious yawning. Unlike spontaneous yawning,
which is evolutionarily older and believed to be
relatively widespread among vertebrates
(Baenninger, 1987), contagious yawning is a more
recently derived behavior present in relatively
few highly social species (Gallup, 2011). In
particular, contagious yawning has been reported
in humans (e.g., Provine, 1986; Platek et al.,
2003), some non-human primates including
chimpanzees (e.g., Anderson et al., 2004;
Campbell et al., 2009), bonobos (Demuru and
Palagi, 2012; Palagi et al., 2014), and gelada
baboons (Palagi et al., 2009); in wolves (Romero
et al., 2014), in domesticated dogs in response
to human yawns (Joly-Mascheroni et al., 2008;
Madsen and Persson, 2012; Silva et al., 2012;
Romero et al., 2013; but see Harr et al., 2009;
O'Hara and Reeve, 2011; Buttner and Strasser,
2014), in a sub-line of high frequency yawning
rats (Moyaho et al., 2015), and in budgerigars
(Miller et al., 2012a; Gallup et al.,
2015).
While spontaneous yawning appears to be
triggered by deviations in brain thermal
homeostasis (Gallup and Gallup, 2007, 2008;
Shoup-Knox et al., 2010; Gallup and Eldakar,
2013; Massen et al., 2014; Eldakar et al.,
2015), which could serve as a potential
mechanism to promote cortical arousal
(Baenninger, 1997) or state change (Provine,
1986; Liang et al., 2015), contagious yawning is
elicited simply by sensing or thinking about the
action (Provine, 2005). Consistent with this
mode of response activation, it has been
hypothesized that contagious yawning is rooted
within a perception-action mechanism tied to
basic forms of empathic processing (Preston and
de Waal, 2002).
A growing literature shows an indirect
association between contagious yawning and
empathy, both behaviorally (Platek et al., 2003;
Palagi et al., 2009; Campbell and de Waal, 2011,
2014; Demuru and Palagi, 2012; Norscia and
Palagi, 2011; de Waal, 2012; Romero et al.,
2013, 2014; Silva et al., 2012; Rundle et al.,
2015; but see Bartholomew and Cirulli, 2014) and
neurologically (Platek et al., 2005; Arnott et
al., 2009; Nahab et al., 2009; Cooper et al.,
2012; Haker et al., 2013; but see Schurmann et
al., 2005; Gallup and Church, 2015). Studies
investigating the developmental onset of
contagious yawning in children also generally
support this view (Anderson and Meno, 2003;
Millen and Anderson, 2010; Hoogenhout et al.,
2013), since contagious yawning develops in
parallel with empathy related capacities (e.g.,
Perner and Lang, 1999).
Initial reports on the absence of contagious
yawning in children with autism spectrum
disorder also supported this connection to
empathy (Senju et al., 2007; Giganti and Ziello,
2009; Helt et al., 2010), but subsequent
research shows that this result may be a
consequence of the reduced tendency for these
individuals to spontaneously attend to others'
faces (Senju et al., 2009; Usui et al., 2013).
Consequently, the link between contagious
yawning and empathy remains debated (see Yoon
and Tennie, 2010). Nonetheless, while contagious
yawning may serve as a useful marker for
social-psychological functioning, future
research into this area is warranted.
While the adaptive value of contagious
yawning remains largely unclear, some recent
experimental research suggests a role in
promoting group coordination and/or vigilance
(Miller et al., 2012b). Indirect support for
this view also comes from a growing number of
behavioral reports investigating contagious
yawning in relation to group affiliation or
social closeness/bonding. In particular, an
influential study on chimpanzees demonstrated
evidence for an in-group bias for contagious
yawning (Campbell and de Waal, 2011). In
particular, captive chimpanzees shown video
stimuli of other chimpanzees yawning showed
contagion to in-group members but not unfamiliar
conspecifics. However, a similar familiarity
bias has not been demonstrated for chimpanzees
viewing human yawns (Madsen et al., 2013;
Campbell and de Waal, 2014), and at least one
study provided evidence that relationship
quality among chimpanzees within a single group
did not predict yawn contagion (Massen et al.,
2012).
Nonetheless, similar in-group findings
involving a measure of social closeness have
also been demonstrated in humans (Norscia and
Palagi, 2011), gelada baboons (Palagi et al.,
2009), bonobos (Demuru and Palagi, 2012; Palagi
et al., 2014), and wolves (Romero et al., 2014),
with mixed support for domesticated dogs to
catch familiar human yawns (O'Hara and Reeve,
2011; Silva et al., 2012; Madsen and Persson,
2012; Romero et al., 2013). For example,
naturalistic observations on humans have
revealed contagious yawning to be of
significantly higher frequency when witnessing
kin and friends yawn, in comparison to
acquaintances
and strangers (Norscia and Palagi, 2011). In
great apes, the status or dominance of the
target individual (i.e., the yawner) also seems
to influence the susceptibility for others to
catch this behavior (Demuru and Palagi, 2012;
Massen et al., 2012). For example, chimpanzees
and bonobos are more likely to yawn in response
to witnessing yawns from conspecifics of the
dominant sex (i.e., males in chimpanzees,
females in bonobos). It remains unclear,
however, whether the yawns of dominant
chimpanzees and bonobos are more contagious or
that individuals just pay more attention to
dominants, which is consistent with the view
that monitoring high status or dominant
individuals will provide important information
regarding changes in reproductive opportunities,
group state or vigilance (Chance, 1967; Keverne
et al., 1978; McNelis and Boatright-Horowitz,
1998; Shepherd et al., 2006; Overduin-de Vries
et al., 2012). Similarly, one could argue that
individuals pay more attention to their friends
and/or in-group members, explaining the related
differences in contagiousness of yawns.
Unfortunately, however, aside from that
mentioned above, little is known about such
in-group and status effects on contagious
yawning in humans.
The current research was designed to build
from this existing comparative literature by
investigating how some of these interpersonal
variables influence contagious yawning in
humans. Specifically, based on a study showing
an effect of in-group voter biases of reflexive
gaze-following to politicians (Liuzza et al.,
2011), we tested whether a similar effect would
be present when activating contagious yawning.
Since humans have developed large-scale
political behavior and evolved in relatively
large societies, political identity or
affiliation may provide salient in-
group/out-group cues that modulate other
unconscious social responses. Previous research
has shown that neural activity differs markedly
when viewing images of US Presidential
candidates of the same vs. opposing political
party (Kaplan et al., 2007). For example,
viewing opposing-party candidates activated
areas of negative emotion (e.g., insula,
anterior temporal cortex, anterior cingulate
cortex, and dorsolateral prefrontal cortex)
while, at least for Democrats, viewing
same-party candidates enhanced activation in
areas associated with positive emotion and
empathy (i.e., the medial prefrontal cortex).
Furthermore, distinct patterns of neural
activity have been shown to occur within brain
regions associated with theory of mind when
participants were asked to take the perspective
of ingroup vs. outgroup political candidates
(Falk et al., 2012). In particular, the
posterior cingulate cortex becomes more active
during perspective taking of an ingroup
political candidate, while the bilateral
temporoparietal junction was more active when
considering an outgroup candidate's political
views.
While it has been confirmed that both seeing
and thinking about yawns can trigger contagious
yawning in humans (i.e., Provine, 1986), it has
been reported that hearing yawns can produce the
same effect (Provine, 2005). One study has used
audio recordings of yawns to assess the
self-reported urge to yawn in humans during fMRI
scans (Arnott et al., 2009), but to date there
is no evidence showing that auditory cues
presented in isolation can elicit yawn contagion
in humans. Similarly, comparative studies have
primarily used visual stimuli for tests of
contagious yawning in non-human animals.
Currently only two studies have addressed
whether auditory cues alone can trigger this
response. Experimental research has provided
evidence for cross-species (human to canine)
auditory contagious yawning in domesticated dogs
(Silva et al., 2012), and an observational study
has shown an increase in yawning frequency among
gelada baboons when in the presence of auditory
cues from nearby conspecifics (Palagi et al.,
2009). In this initial investigation we
manipulated the political affiliation (Democrat
vs. Republican) and status (former US Presidents
vs. Cabinet members) of our target stimuli,
which were paired with either breathing or
yawning sounds. This is the first study to test
the effectiveness of this priming stimulus in
actually eliciting yawns in humans. Consistent
with previous comparative research on empathy
and perspective taking, social affiliation and
status, and auditory contagion, we hypothesized
that participants would yawn significantly more
when hearing yawning sounds paired with images
of same-party, high status political
figures.
DISCUSSION
These findings add to our understanding of
the sensory, perceptual and interpersonal
factors contributing to (contagious) yawning in
humans. The current study is the only one to our
knowledge to show auditory contagious yawning in
humans. This finding is consistent with an
earlier study reporting that humans' urge to
yawn increases after having heard someone yawn
(Arnott et al., 2009). Moreover, this finding is
in line with a previous report on domesticated
dogs demonstrating that hearing a yawn, albeit
heterospecific (i.e., human), elicits contagious
yawning in this species (Silva et al., 2012). In
addition, there are observational indications
that gelada baboons increase their yawning in
the presence of yawning vocalizations of nearby
conspecifics (Palagi et al., 2009). The current
findings suggest auditory contagious yawning may
also be a preserved mechanism present in
mammalian species that show yawn contagion and
not just restricted to those that are
particularly well adapted with regard to
hearing; e.g., dogs.
Our results also show that an auditory yawn
stimulus is equally effective as visual stimuli
for priming contagious yawning in humans, since
49.2% of the participants that heard a yawn
reported a yawn in comparison to roughly 40-55%
who yawn after seeing pictures or videos of
people yawning (e.g., Provine, 1986; Platek et
al., 2003; Gallup and Gallup, 2007; Massen et
al., 2014). In addition, although it is possible
that the participants suspected that the audio
clips were not from the politicians themselves,
this stimulus proved to be a good tool for
priming contagious yawning with regard to
particular visual stimuli, as we found several
patterns in the reported contagiousness of
yawning in response to hearing someone yawn that
could only be attributed to the corresponding
visual stimuli. Therefore, this methodology
could be preferred when, as in the current
study, visual yawning stimuli are highly
variable and/or difficult to acquire.
Aside from auditory signal, our results
demonstrate that the self-reported response to
yawn was influenced by the status of the
politicians in the pictures (see below for
extended discussion of methodological
limitations). We found that regardless of the
auditory priming stimulus, be it yawning or
breathing, participants reported yawning more
when seeing a low status political figure than a
high status political figure. This contrasts
with findings on bonobos and chimpanzees, where
yawns of high status/ranking individuals
(females among bonobos, males among chimpanzees)
were more contagious than those of low ranking
individuals (Demuru and Palagi, 2012; Massen et
al., 2012). For apes this effect could be a
result of differences in attention since the
monitoring of dominant animals might provide
several advantages regarding changes in
reproductive opportunities, group state or
vigilance (Chance, 1967; Keverne et al., 1978;
McNelis and Boatright-Horowitz, 1998; Shepherd
et al., 2006; Overduin-de Vries et al.,
2012).
Our participants were specifically requested
to pay equal attention to all stimulus images,
and thus these results possibly suggest a
discontinuity between apes and humans in the
evolution of yawning in response to, or simply
in the presence of, individuals of different
status. This discontinuity may be a cultural
effect, portraying that for humans yawning is
often viewed as a sign of boredom or disrespect
in the presence of others (Schiller, 2002),
which may inhibit yawning in general when
viewing individuals of extremely high status and
reputation like the former presidents in our
study. Such an inhibition hypothesis, however,
would predict that the urge to yawn would be
equal for both high and low status politicians,
though that it would only be inhibited in case
of the high status former presidents. This
interpretation is supported by recent research
demonstrating that administration of intranasal
oxytocin produces a significant discrepancy
between the self-reported urge to yawn and
actual yawning rates of human participants,
suggesting that individuals may consciously
inhibit this response (Gallup and Church, 2015).
Conversely, we found a similar status effect
with regard to the urge to yawn. That said, the
reduced self-reported yawning (and urge to yawn)
in response to past US presidents might not
generalize across other high status target
stimuli. In fact, there may be a similar high
status copying bias as seen in other great apes
when using target individuals of varying social
dominance/rank that are within the participants'
social network, such as among friends, coworkers
or different social organizations.
Unlike previous research showing an in-group
political bias for gaze-following (Liuzza et
al., 2011), we found no evidence for similar
results of self-reported contagious yawning.
Multiple factors could explain this null effect.
Most notably, our sample was rather homogeneous
in their political views, with 63.6% identifying
as Democrats and only 11.0% identifying as a
Republican. Thus, future research could aim to
draw from a larger and more representative
population to test these effects. However, we
found no evidence for an in-group bias among the
Democrats to yawn more in response to President
Clinton or former Secretary
Daley. It should be noted here that we ran
models with the actual political affiliation
(which party) of our subjects as well as with a
categorization of Democrats vs. all other
parties, though the results were the same and
neither variable was significant in the best
fitting models. Since none of the political
figures used as our target stimuli are still in
office, and were so only when the participant
sample was quite young, it also remains possible
that using contemporary politicians or older
participants would produce a greater
in-group/out-group affiliation.
Alternatively, despite differences in neural
activity associated with viewing images or
taking the perspective of politicians from
congruent or opposing parties (Kaplan et al.,
2007; Falk et al., 2012), political affiliation
and/or political in-group affiliation may simply
not be a strong enough cue to affect contagious
yawning.
Although our results provide effects of
unique interpersonal factors influencing yawning
in humans, there are additional limitations to
acknowledge within this preliminary study. In
particular, the current study only utilized a
single exemplar for each of the four stimulus
conditions (HS Dem, LS Dem, HS Rep, LS Rep), and
thus further investigation is needed in order to
replicate and clarify these results.
Furthermore, the self-report nature of our data
is suboptimal. In order to keep the goals of the
experiment implicit participants were asked to
self-report on their yawning behavior at the end
of the task, but this may have increased the
chances of measurement error. The inclusion of
the subset of video recorded participants
confirmed the validity of overall self-report
yawning behavior, and showed a similar
distribution of individuals that yawned in
respect to the auditory treatment.
However, validation of recall of the
specific political stimuli was only fair, and
the distribution of individuals that yawned
during exposure to high and low status political
figures tended to differ between video and
self-report data. Since the images of the
politicians were explicitly labeled for status,
it is possible that expectancy bias contributed
to participants assigning fewer yawns in the
presence of high status figures. Therefore, bias
in self-reporting may have contributed to the
status effect. Nonetheless, subjective responses
in the current study provide understanding of
human perception and may reveal important
factors contributing to the social stigma
associated with yawning in the presence of
others. Future research could be conducted to
improve upon methodological and sampling
limitations and further explore the interplay
between perceived and actual inhibition of
behavioral contagion as it relates to social
context.
In conclusion, the results of this
investigation validate the use of auditory
stimuli to prompt yawn contagion and provide
evidence for an effect of target status
influencing the self-reported nature of this
response in humans. Despite the limitations in
our sample and design, these initial findings
broaden the existing literature and provide a
clear framework for pursuing further research in
this area.
