Autisme spectrum disorder (ASD) is
characterized by core social impairments (c.g.,
reduced nonverbal social communication
behaviors, lack of perspective taking, and
deficits in social-emotional reciprocity.
Individuals with ASD may also exhibit a reduced
capacity for empathy, as defined and reviewed by
Bons et al. [2013], including
impairments in facial mimicry and emotion
recognition. The contagious yawn response (i.e.,
yawning in response to the perception of another
individual's yawn) has been found to be
positively associated with performance on
self-face recognition and theory of mind stories
[Platek, Critton, Myers, & Gallup Jr.,
2003]. Additionally, viewing yawns was found
to increase activity in the mirror neuron system
[Haker, Kawohl, Herwig, & Rosslcr,
2013]. As a result, contagious yawning has
been used in research studies as a proxy measure
of empathy [Senju et al., 2007].
Some studies have found contagious yawning
to be reduced in children with ASD compared to
typically developing (TD) children [Giganti
& Ksposito Ziello, 2009; Senju et al.,
2007]. However, not all studies have
replicated the impaired contagious yawn response
finding in children with ASD. Several such
studies have attributed the existence of this
phenomenon instead to group differences in
attention paid to yawn stimuli during task
performance [Senju et al., 2009; Usui et
al., 20131. Another possibility, however, is
that the impaired contagious yawn response
exists, but only in a subset of children with
ASD. This might be particularly true if
individual differences in the biology that
regulates social functioning drive contagious
yawning behavior in children with ASD, thereby
contributing to phenotypi-cally hetcrogenous
study cohorts and inconsistent study
outcomes.
Although the neurochemical underpinnings of
contagious yawning are unknown, a promising
biological candidate is the neuropeptide
oxytocin (OXT). In animal models, both OXT
administration and activation of oxytocinergic
neurons have been found to induce yawning
[Eguibar, Codes, Isidro, & Ugarte, 2015;
Kita, Yoshida, & Nishino, 2006].
Moreover, OXT promotes prosocial functioning in
mammals lAnacker & Beery, 2013] and
administration of OXT enhances empathy in humans
[Hurlemann et al., 2010]. Blood OXT
concentration bas also been found to predict
cerebrospinal fluid OXT concentration in the
same individuals [Carson et al., 2015],
suggesting that blood OXT concentration may be a
useful surrogate measurc of brain OXT activity.
For example, a previous study found low
circulating blood concentrations of OXT to be
associated with diminished theory of mind
ability, a cognitive form of empathy, in human
children |Parker et al., 2014]. The goal of
the present study therefore was to test whether
children with and without ASD differ in
contagious yawning behavior, while concomitantly
testing whether blood OXT concentration and
attention paid to yawn stimuli during task
performance impact this behavioral
phenomenon.
Discussion
Previous research on contagious yawning
behavior in children with ASD has produced
equivocal results. Some studies have reported an
impaired contagious yawn response in children
with ASD versus TD children [Giganti &
Esposito Ziello, 2009; Senju et al., 2007],
whereas other studies have failed to replicate
this finding [Senju et al., 2009; Usui et
al., 2013], instead attributing its
existence to group differences in attention paid
to yawn stimuli. The present study sought to
address this discrepancy by testing whether a
subset of children with ASD had this impairment
and if it could he biologically identified. This
was indeed the case, as children with ASD at
lower blood OXT concentrations exhibited an
impaired contagious yawn response but those at
higher blood OXT concentrations were
behaviorally similar to TD children.
Importantly, this finding was observed whether
attention paid to test stimuli was included in
the statistical model or not, and attention
never predicted yawn rate. The conclusion that
attention is not a significant driver of yawn
rates was further supported by our exploratory
factor analysis showing that attention rate and
IQ score significantly loaded onto one factor,
whereas blood OXT concentration and contagious
yawning rate significantly loaded onto
another.
We found no evidence that children with ASD
showed delayed perception of yawn stimuli, as
they did not differ from TD children in yawning
rates during neutral stimulus trials. We
likewise found no relationship between SRS total
score and contagious yawning in children with
ASD, and the relationship between blood OXT
concentration and contagious yawning held
whether or not SRS total score was included in
the model. These findings suggest that an
impaired contagious yawn response is not due to
social perceptual delays or greater clinical
symptom severity, but rather, that individuals
with low OXT concentrations may constitute an
ASD "subgroup" characterized by empathy-related
impairments in facial mimicry and reflexive
behavior copying [Massen 8c Gallup,
2017]. Additional research including
pharmacological manipulation of the OXT System
is now required to more fully test this
hypothesis.
Interestingly, the relationship between
blood OXT con-centration and contagious yawning
behavior was specific to children with ASD.
Exactly why TO children did not show this
relationship is unknown, but it is not due to
"ceiling" effects in their contagious yawning
behavior. One possible explanation for this
finding is that in individuals whose social
functioning is unaffected, higher OXT
concentrations may enhance social awareness of
contagious yawning, and thus lead to its
suppression [Gallup & Church, 2015].
Our participants completed the contagious
yawning task in a room containing both a webcam
and a researcher; the presence of each of these
individually bas been shown to decrease
contagious yawning behavior in TD individuals
[Gallup, Church, Miller, Risko, 8c
Kingstone, 2016].
This study had several limitations.
First, our sample was mad biased, in keeping
with ASD's population prevalence
[Christensen, 2016], but nevertheless
not powered to discern potential sex differences
in the relationship between blood OXT
concentrations and contagious yawning
behavior.
Second, this study related blood OXT
concentrations to a brain-mediated social
behavior. Whether or not peripheral OXT
concentrations are a robust indicator of
brain-related OXT activity remains to be
determined, but at least some research suggests
that this may be the case [Carson et al.,
201 S].
Third, although all of our study
participants underwent identical experimental
procedures, it should be noted that participants
nevertheless were drawn from several distinct
study populations.
Fourth, our ASD participants were not
medication free. Although their medications were
stable (i.e., for at least 4 weeks) before blood
collection, it is possible that blood OXT
concentrations and/or the contagious yawn
response were influenced by their medication
status, finally, this study manually coded
attention as a binary measure per stimulus
trial. Follow-up research could benefit from
measuring attention with a more precise
technology (e.g., eye tracking).
Despite potential limitations, the
present study provides the first evidence that
blood OXT concentration is positively associated
with contagious yawning behavior in children
with ASD. This study also identified a subset of
children with ASD with lower blood OXT
concentrations who exhibit an impaired
contagious yawn response compared to TD
children. These collective findings suggest that
a biologically defined subset of children with
ASL) may exhibit reduced empathy, as measured by
an impaired contagious yawn response, and that
discrepant reports of this behavioral phenomenon
in the literature may be attributable, at least
in part, to variable mean OXT concentrations
across ASD study cohorts.
