Yawn contagion is not restricted to humans
and has also been reported for several non-human
animal species, including chimpanzees (Pan
troglodytes). Contagious yawning may lead to
synchronisation of behaviour. However, the
function of contagious yawning is relatively
understudied. In this study, we investigated the
function of contagious yawning by focusing on
two types of signal providers: close social
associates and leaders. We provided a captive
chimpanzee colony with videos of all individuals
of their own group that were either yawning, or
at rest. Consistent with other studies, we
demonstrated that yawning is contagious for
chimpanzees, yet we did not find any effect of
relationship quality on yawn contagion. However,
we show that yawn contagion is significantly
higher when the video model is a yawning male
than when the video model was a yawning female,
and that this effect is most apparent among
males. As males are dominant in chimpanzee
societies, male signals may be more relevant to
the rest of the group than female signals.
Moreover, since chimpanzees form male-bonded
societies, male signals are especially relevant
for other males. Therefore, we suggest that the
sex-differences of yawning contagion among
chimpanzees reflect the function of yawning in
the synchronisation of behaviour.
Introduction
Recently, interest has grown in contagious
yawning due to its proposed link with empathy
[1]-[4]. Yawning contagion,
where yawning in one individual enhances yawning
frequency in observing individuals, or
contagiousness of behaviour in general, is
considered part of emotional contagion, the
first and most basal level of empathy
[1]; i.e. being affected by the
emotional or arousal state of another individual
[2]. Apart from theory, there is also
empirical support that in humans contagious
yawning is related to empathy, albeit indirect.
The susceptibility to contagious yawning is
positively correlated to questionnaire measures
of empathy [3], and in a large
multi-cultural data set the degree of yawn
contagion could only be linked to the strength
of the social bond [4]. Moreover, people
with impairments in empathy (i.e. schizophrenics
or people with autism spectrum disorder) appear
to show impaired yawn contagion [3],
[5]-[7]. From an ontogenetic
perspective, however, children only develop yawn
contagion after they have developed other
empathy-based behaviours, even when the yawning
model was their own mother [8].
Yawn contagion was first believed to occur
only in humans (Homo sapiens) [9], but
has now also been reported in chimpanzees (Pan
troglodytes) [10], [11] and
gelada baboons (Theropithecus gelada)
[12]. Similar findings were reported for
stump-tail macaques (Macaca arctoides)
[13], yet it remains unclear whether
these resulted from actual yawn contagion or
were driven by tension [13]. In
addition, two studies also reported contagious
yawning in domestic dogs (Canis familiaris)
[14], [15], yet two other
studies failed to replicate these results
[16], [17]. Furthermore, three
recent studies in non-human animals provided
indirect evidence for a potential link between
yawn contagion and empathy [12],
[15], [18]. Empathy is related
to the relationship with an individual, and is
expected to be more apparent when individuals
are socially closer, more familiar or more
similar [19]-[21]. In line with
this argument, a study on gelada baboons
demonstrated a positive correlation between the
contagiousness of yawning and grooming levels
between dyads; i.e. individuals that were
socially closer appeared to be more prone to
contagion of each other's yawns [12].
Similarly, a recent study on chimpanzees
demonstrated an ingroup-outgroup bias for yawn
contagion; i.e. individuals yawned more in
response to a yawning individual from their own
group than in response to yawning strangers
[18]. Finally, a study on keeper-dog
contagious yawning found that the yawning of a
familiar caretaker was more contagious to dogs
than the yawns of unfamiliar humans
[15]. Although indirect, these results
strengthen the idea that also in non-human
animals a link exists between contagious yawning
and empathy. Nevertheless, empathy may operate
as a (proximate) mechanism affecting the
susceptibility to yawning contagion, yet does
not provide an evolutionary (ultimate) function
of this behavior. Consequently, the function of
yawn contagion was not investigated in these
studies and remains relatively
understudied.
From a functional (ultimate) perspective, it
has been suggested that yawn contagion causes
synchronization of group behaviour, by
synchronization of rest-activity transitions
[22]. Emotional contagion, suggested to
be present in chimpanzees [23], may
facilitate this process, as it causes adoption
of another's emotional or arousal state
[22] and may in turn result in better
behavioural synchronisation with preferred
animals [cf. 12], group members [cf.
18], or caretakers [cf. 15].
Here, we investigated the function of yawn
contagion and its proposed link to empathy in a
captive group of chimpanzees. We hypothesised
that if contagious yawning has a communicative
function, yawn contagion will be higher when the
relevance of the signal is high; i.e. when
provided by a socially close individual and by
those individuals that decide upon group
movement. This results in two predictions.
First, we expect that yawn contagion is more
prevalent among individuals with a good than a
bad quality relationship. Second, as in
chimpanzees males are dominant [24] and
guide the movements of their groups
[25], [26], we predict that yawn
contagion will be higher if a male provides the
signal. We presented 15 adult chimpanzees yawn
and control videos of all group-members and
measured their response (i.e. yawning).
Thereafter, we correlated the contagiousness of
yawning with the relationship quality with the
target individual. Moreover, we compared the
contagiousness of yawns of the different
sexes.
Discussion
Our study confirms previous studies
[10], [11], [18], and
shows that yawning is contagious for
chimpanzees. This, however, was the first study
that used an experimental group set-up to
investigate yawn contagion. In contrast to
studies that voluntarily separate animals, this
group set-up allowed us to test all animals in
our study population, and also avoided any
negative effects of the stress of separation, a
possible confound in yawning studies
[13], [22]. A group set-up,
however, makes the analyses of yawning contagion
more challenging, since it is hard to
distinguish who causes the yawning contagion;
either the actual model or a conspecific that
previously responded to the model with a
yawn.
In contrast to other studies [4],
[12], our study did not show an effect
of relationship quality on contagious yawning.
The lack of an effect of relationship quality on
yawn contagion in our group of chimpanzees may
be a result of possible small variation in
relationship quality due to the relatively small
size of the group in comparison to natural
groups [24], or due to the long and
stable residence of the members of our study
group. Alternatively, differences in outcomes
may result from different methods; whereas the
studies on humans [4] and gelada baboons
[12] rely on observed spontaneous yawns,
our study employed an experimental approach and
stimuli may be somehow artificial. Nonetheless,
the supposed link between contagious yawning and
relationship quality follows from the prediction
that empathy may be the proximate mechanisms
driving the contagiousness of yawning
[1]. However, as our data do not support
a link between yawn contagion and factors
indicating strong social bond, the proposed
relationship between yawn contagion and empathy
[1-4, 12, 15, 18, but see 8] is not
supported by our findings. Altogether, the
relationship between empathy, relationship
quality and yawn contagion in animals needs to
be further investigated.
This study does show strong sex effects of
both the model and the subject on the
contagiousness of yawning. We show that male
yawns are far more contagious than those of
females, and that this effect is most apparent
when the subject is a male too. From an ultimate
perspective, these results follow the prediction
that yawning contagion functions to facilitate
the synchronisation of behaviour [22],
one would then predict that yawning is only
contagious if the model's behaviour is relevant
to others. In chimpanzees the higher contagion
of male than female yawns is consistent with
males being the dominant sex [24] and
initiating group movement [25],
[26]. In addition, individuals of the
bonded sex (males [29]) are also
expected to influence each other more, and this
is consistent with the strong effect males have
on other males. Similar patterns have been found
among female-bonded gelada baboons, where yawn
contagion was most apparent among females
[12]. Moreover, this line of thought is
bolstered by recent findings that the yawns of
ingroup members are far more contagious than
those of outgroup individuals [18],
since signals of outgroup individuals, although
outgroup members and their signals are very
interesting [18], have little relevance
with regard to the synchronization of group
behaviour. Also, these results are in line with
recent findings showing that for dogs the yawns
of familiar caretakers are more contagious than
those of unfamiliar humans [15], since
the signals of caretakers are for more salient
for dogs with regard to the synchronisation of
behaviour than those of unfamiliar humans. And,
although not supported by our findings, also the
correlations between measures of social bonds
and yawn contagion [4], [12] are
consistent with this hypothesis, since signals
of socially close individuals will be more
relevant than those of less close individuals.
Nonetheless, our data do not reveal whether yawn
contagion actually leads to the synchronisation
of behaviour, and consequently, studies that
truly explore the ultimate function of yawn
contagion are still needed.
