- Tous
les articles sur la contagion du
bâillement
- All
articles about contagious
yawning
-
- Abstract
- Yawning is highly contagious, yet both its
proximate mechanism(s) and its ultimate
causation remain poorly understood. Scholars
have suggested a link between contagious yawning
(CY) and sociality due to its appearance in
mostly social species. Nevertheless, as findings
are inconsistent, CY's function and evolution
remains heavily debated. One way to understand
the evolution of CY is by studying it in
hominids. Although CY has been found in
chimpanzees and bonobos, but is absent in
gorillas, data on orangutans are missing despite
them being the least social hominid.
-
- Orangutans are thus interesting for
understanding CY's phylogeny. Here, the authors
experimentally tested whether orangutans yawn
contagiously in response to videos of
conspecifics yawning. Furthermore, they
investigated whether CY was affected by
familiarity with the yawning individual (i.e. a
familiar or unfamiliar conspecific and a 3D
orangutan avatar). In 700 trials across 8
individuals, they found that orangutans are more
likely to yawn in response to yawn videos
compared to control videos of conspecifics, but
not to yawn videos of the avatar. Interestingly,
CY occurred regardless of whether a conspecific
was familiar or unfamiliar.
-
- They conclude that CY was likely already
present in the last common ancestor of humans
and great apes, though more converging evidence
is needed.
-
- Résumé
- Le bâillement est contagieux, mais son
(ses) mécanisme (s) immédiat (s)
et sa cause ultime restent mal compris. Les
chercheurs ont suggéré un lien
entre le bâillement contagieux (CY) et la
socialité en raison de son apparition
dans des espèces à vie sociale.
Néanmoins, comme les résultats
sont discordants, la fonction et
l'évolution de CY restent largement
débattues. Une façon de comprendre
l'évolution du CY est de l'étudier
chez les hominidés. Bien que CY ait
été trouvé chez les
chimpanzés et les bonobos, mais soit
absent chez les gorilles, les données sur
les orangs-outans sont manquantes bien qu'ils
soient les hominidés les moins
sociaux.
-
- Les orangs-outans sont donc
intéressants à étudier pour
comprendre la phylogénie de CY. Ici, les
auteurs ont testé
expérimentalement si les orangs-outans
bâillent de manière contagieuse en
réponse à des vidéos de
conspécifiques bâillant. En outre,
ils ont examiné si CY était
affecté par la familiarité avec
l'individu bâillant (c'est-à-dire
un conspécifique familier ou inconnu et
un avatar d'orang-outan 3D). Dans 700 essais
portant sur 8 individus, ils ont observé
que les orangs-outans sont plus susceptibles de
bâiller en réponse aux
vidéos de bâillements par rapport
aux vidéos de contrôle de
congénères, mais pas aux
vidéos de l'avatar bâillant. Fait
intéressant, CY s'est produit
indépendamment du fait qu'un
conspécifique soit familier ou
inconnu.
-
- Ils concluent que CY était
probablement déjà présent
dans le dernier ancêtre commun des humains
et des grands singes, bien que des preuves plus
convergentes soient encore
nécessaires.
-
-
-
- Yawning is an evolutionarily old phenomenon
as its associated motor features can be
recognized in different groups of animals1. It
follows a stereotyped pattern that, once
started, is unstoppable2. Apart from its
spontaneous form, it is also notoriously
contagious, at least for some species; i.e.
individuals yawn as an unconscious and automatic
response to seeing or hearing other individuals
yawn3. While a yawning-like pattern is observed
in a wide range of vertebrates1, contagious
yawning (CY) is less wide-spread. To date, CY
appears to be present in only a few, generally
social species, including tonkean macaques4 (and
possibly stumptail macaques5), gelada baboons6,
chimpanzees7,8,9,10,11,12,13,14, bonobos15,16,
dogs and wolves17,18,19, sheep20, elephant
seals21, budgerigars22, and rats23. In contrast,
studies failed to show CY in grey-cheeked
mangabeys and long-tailed macaques24,
mandrills1, common marmosets25, lemurs26,
horses27, lions1, tortoises28, and fish1, even
though some of these species are also very
social. Despite growing interest in CY, both its
proximate mechanisms (how it functions and
develops) and ultimate causes (why and how it
evolved) currently remain unclear.
-
- Several hypotheses have been put forward,
following a Tinbergian approach29. One view on
the proximate mechanism underlying CY is that it
is an automatic form of physiological or
emotional state-matching between individuals.
This synchrony of states between individuals may
work via a perception&endash;action mechanism
(PAM), an adaptive mechanism that serves to
create and maintain relationships in highly
social species and that can give rise to
higher-order cognitive phenomena such as
empathy30. Some scholars argue that CY taps into
the same PAM as emotion contagion
(e.g.6,7,31,32), which is the tendency to
automatically synchronize emotional states with
another individual33. Following this line of
thought, CY can thus potentially be a proxy for
empathy (i.e. the CY-empathy
hypothesis)6,9,12,18,31,34,35. Indeed,
neuroimaging studies have shown increased brain
activity during CY in areas involved in theory
of mind and social cognition 36,37,38,
corroborating the idea that CY is linked with
emotional state-matching and perhaps even
empathy. Furthermore, individuals who score low
on empathy scales (e.g. individuals on the
autism spectrum) are less likely to engage in
CY39, and females yawn more frequently in
response to seeing others yawn than males do,
reflecting the idea that females show higher
levels of empathy than males because of their
investment in offspring care40. Nevertheless,
there are some studies that do not find such a
clear link between CY and empathy. For instance,
when people with autism spectrum disorder (ASD)
are instructed to pay attention to the eyes
(avoidance of the eyes is one of the
characteristics of ASD), they are just as likely
to yawn contagiously as neurotypical
individuals41. Furthermore, the gender bias is
not consistently found (e.g. 10,42) and heavily
debated43,44. For instance, in chimpanzees, it
appears that males yawn more frequently than
females in response to seeing other males
yawn10. Finally, while dogs do engage in CY, its
presence is not affected by whether the yawner
is prosocial versus the yawner being
antisocial45. The mixed findings in the studies
investigating the relationship between CY and a
complex construct such as empathy show that the
topic deserves more attention, and that it is
still debated (see Massen & Gallup (2017)
for a critical review).
-
- The emotional bias hypothesis is a more
detailed specification of how CY can be socially
modulated through a shared PAM, namely via
social closeness and familiarity. The hypothesis
predicts that individuals who are socially, and
thus emotionally close are also more likely to
yawn contagiously in response to each
other15,16,18,19,34,46,47. Additionally,
individuals from a group (i.e. familiar others)
are more likely to yawn in response to each
other than to unfamiliar others9,18. A potential
issue that has been raised is that these studies
often fail to rule out simple alternative
explanations for CY that do not require
higher-order cognition48. For instance, effects
of familiarity on CY may be explained by a
general tendency to bias attention to familiar
and socially close others48. Nevertheless, in a
recent study investigating auditory yawn
contagion in humans, yawns were most contagious
between family and friends while controlling for
the potential effects of increased attention to
socially close others using a non-visual
stimuli34. Still, in quite some social species,
the linkage between CY and social closeness or
familiarity is not found10,45,49,50,51. For
example, a recent study analyzing a large
dataset on CY in dogs shows CY is present in
dogs, but is not affected by familiarity or
other potential mediators such as sex or
prosociality45. It therefore remains possible
that mechanisms other than the same PAM that
underlies emotion contagion or empathy are
mediating CY. For instance, CY may result from
stress induced by a common stressor in the
environment5,52. Thus, rather than being
mediated by seeing others yawn, yawning occurs
as a response to the stressor. Individuals that
are stressed are known to show higher rates of
self-directed behaviors, of which yawning and
scratching are examples53, and indeed, in one
study involving stumptail macaques, monkeys
yawned more frequently in response to a video
clip of yawns as compared to a control, but also
scratched more5. The authors concluded that
tension was most likely mediating the occurrence
of yawning in the yawn condition. In short,
while it is likely that CY is a social
phenomenon, its exact mechanisms remain an
active field of investigation.
-
- Notwithstanding the debate on proximate
mechanisms, little attention has been given to
more ultimate explanations for CY. One of the
few hypotheses out there is that CY is an
adaptive mechanism that helps with social
coordination54. Accumulating evidence suggests
that yawning itself serves to cool the brain as
to maintain homeostasis55,56,57,58,59,60 and
consequently may increase alertness and aid in
vigilance. Within this social coordination
hypothesis, CY, in turn, may help to spread
vigilance within the group, for instance to
remain alert for potential predators54,57.
Specifically, it may be adaptive to match the
state of a vigilant conspecific as it may have
sensed a predator, which the individual itself
did not yet sense. To date, however, the social
coordination hypothesis remains untested, and
the thermoregulatory function of yawning is
still debated (e.g.61,62, but see58 for a
response to the critique).
-
- Another fruitful way to explore evolutionary
hypotheses is through phylogenetic comparisons.
Palagi et al. (2019) proposed the common trait
among hominids hypothesis which states that,
given the shared phylogeny between humans and
great apes, CY may find its roots in a shared
underlying socio-cognitive mechanism that was
already present in at least the last common
ancestor (LCA) of all hominids. Moreover, since
CY is also present in some Old-World monkeys and
non-primate species, its roots could be much
older, or CY is an example of convergent
evolution. To date, few data exist to perform
comparisons and most interestingly, the picture
among the great apes is not yet clear. There is
convincing evidence for CY in
chimpanzees7,8,10,12,14. In bonobos, two
observational studies15,16 and an experiment63
show clear evidence for CY, while one
experimental study did not12. However, the
latter study only tested four individuals, thus
making it very likely that CY is, indeed,
present in bonobos. Finally, the first
comprehensive study on gorillas combining an
experimental and naturalistic approach found no
evidence for CY64. Notoriously absent are data
on CY in orangutans, which, considering their
semi-solitary lifestyle65 may be of comparative
interest for a social phenomenon like CY. To
date, the only existing study involving
orangutans failed to find evidence for CY12, yet
the sample size was too small to be conclusive.
In general, orangutans in the wild roam mostly
solitarily: males travel alone, and mothers
travel with their offspring66. Due to
overlapping home ranges, occasional encounters
and affiliation are possible, but generally do
not occur frequently66,67. Consequently, finding
out whether CY is present in orangutans will
further help elucidate the hypotheses previously
discussed.
-
- The current study attempts to clear up the
picture of CY in hominids in two ways. First, we
aim to find a convincing answer to whether CY is
present in orangutans or not via an experimental
design involving the presentation of yawning and
neutral stimuli of orangutans to 8 orangutans.
Second, we also investigate whether this
potential yawn contagion is affected by a
familiarity bias, i.e. whether CY is stronger
between individuals that know each other versus
unfamiliar individuals. To this end, we exposed
orangutans to videos showing either yawn or
control clips of familiar (i.e. conspecifics
living in close proximity) and unfamiliar
orangutans, as well as a 3D avatar68 and
measured their response (yawns). Additionally,
we also measured the occurrence of scratching to
rule out potential effects of stress on the
occurrence of yawning53. So far, CY appears to
be exclusively present in highly social species,
and because orangutans do not show high
affiliative tendencies, we therefore expected
that orangutans do not show CY.
- Discussion
- Here we find that orangutans yawn
contagiously in response to conspecifics
yawning, independent of whether the conspecific
is a familiar or unfamiliar individual.
Furthermore, orangutans were not susceptible to
yawns of an avatar. Additionally, the videos
used in our experiment appeared to be similarly
arousing. That is, there was no difference in
scratching (an indicator of stress) between the
conditions. We here discuss the consequences of
our findings for the different proximate and
ultimate hypotheses that currently exist.
-
- CY has thus far been observed in highly
social species6,7,15,17,19,20,21,22 (but
see:1,24,25,26,27). Orangutans have meaningful
social interactions that occur more often than
is expected by chance alone76, but these
interactions occur at a much lower frequency
compared to bonobos and chimpanzees66,67.
Interestingly, our results show that orangutans
exhibit CY, suggesting that a high degree of
affiliation within a species is not necessary
for CY to occur. This also indicates that more
studies are needed that investigate the presence
or, importantly, absence of CY in a variety of
species that differ on their social organization
and affiliative tendencies. At the same time, it
has to be noted that our sample consists of
zoo-housed orangutans that were also born in
captivity. In captivity, frequencies of
affiliation can exceed those observed in the
wild77, thus potentially increasing the
likelihood of CY to occur. Nevertheless, our
results do show the presence of CY in orangutans
and the few generations of zoo-living
individuals cannot inform us about any selection
pressures that have resulted in this tendency in
orangutans. Our results must therefore be
discussed in light of the orangutans' natural
behavior and social environment.
-
- In our study, we did not find an effect of
familiarity on CY, suggesting that at least in
orangutans, social modulation of CY may not be
present. While presence of social modulation of
CY is often used as confirmation of CY and
emotion contagion sharing the same underlying
perception&endash;action mechanism9,15,16,18,35,
its absence in our data makes it more difficult
to interpret the emotional bias hypothesis.
Orangutans do have some preferences when it
comes to their interaction partners, thus one
could expect social modulation of CY under the
emotional bias hypothesis. For instance, related
female orangutans are known to associate more
often than unrelated females78, and prefer the
long-calls of dominant males79. Additionally, in
a recent study, orangutans were shown to scratch
contagiously in response to conspecifics
scratching, suggesting a potential case of
emotion contagion80. Interestingly, scratch
contagion was stronger between weakly bonded
individuals during tense situations, which shows
a social closeness bias in the opposite
direction. This suggests that a familiarity bias
may be more flexible depending on the situation
individuals are in (e.g. relaxed versus
stressful contexts) and the nature of the
behavior that is copied (e.g. scratching as an
expression of tension). At the same time, there
are other studies on highly social species that
do not show a familiarity bias (e.g.
chimpanzees10, dogs45, macaques24, and
marmosets25). As such, there may be (currently
unknown) species-specific traits that determine
whether a familiarity bias occurs or not. The
exact (social) function of CY remains unclear
and thus alternative explanations that do not
involve the PAM that is underlying empathy may
still be possible (e.g. spreading of vigilance).
As has been pointed out by others, solving this
issue requires a more systematic study of CY
that includes a bigger variety of animals,
including solitary animals such as reptiles and
amphibians48.
-
- From an evolutionary perspective, our
results pose an interesting conundrum: while we
found CY in orangutans, it is not present in
gorillas, even though the split between
orangutans and other hominids is evolutionarily
older than the split between gorillas and other
hominids81. It is possible that the number of
trials in the study by Palagi et al. (2019) were
not sufficient to detect CY, as in our study,
even with a large number of trials, we only
detected yawns in 11.9% of all cases.
Nevertheless, studies with chimpanzees that have
few trials were able to establish CY in the
past, albeit with a relatively large number of
subjects8,10,12, and there was also no evidence
for CY in naturalistic observations in
gorillas64. Interestingly, it has been argued
that in the past, orangutans may have been more
social, but that due to long periods of low food
availability, orangutan gregariousness was no
longer viable82. This may suggest that the
ancestor of all apes already possessed the
mechanism underlying CY. However, based on
observational and relatedness data, it has been
suggested that this hominid lived in a group
with gorilla-like structure in which one male
could monopolize multiple females82. In this
sense, it is difficult to explain why, given a
similar social structure, gorillas do not show
CY and orangutans do. It is possible that CY was
somehow lost in the gorilla lineage, or that CY
developed multiple times over the course of
evolution. The loss of CY is theoretically
possible, given that CY has been found in some,
but not all primates1,64. Here, there is a role
for the type of social system that characterizes
a species in the loss (or occurrence) of CY64.
There is, however, not yet enough variation in
data on CY in different species of primates to
draw clear conclusions. Furthermore, it is
possible that the measures to detect CY in
certain species are simply not sensitive enough.
All these explanations can be true, given that
the occurrence of CY is highly variable in
primates in general. It is clear that more
studies are needed in order to draw robust
conclusions about the evolution of CY.
-
- In our study orangutans did not
significantly respond to the avatar, which
contrasts with findings in chimpanzees8.
Potentially, orangutans experienced the uncanny
valley phenomenon in which the avatar looks very
realistic, yet fails to behave like a real
orangutan, therefore violating natural
expectations of orangutan behavior. Indeed,
previous research on monkeys showed that they
preferentially looked at real or completely
unrealistic 3D model monkeys compared to very
realistic 3D models83. Nevertheless, this would
likely have increased scratching when viewing
the avatar, which was not evident in our study.
Furthermore, a recent study investigating the
uncanny effect in macaques showed that looking
times did not differ between the Primatar (3D
monkey head) and real or unrealistic images,
indicating that the use of virtual stimuli can
still be a promising way to study social
cognition84. Future studies will have to verify
whether the lack of evidence for CY using an
avatar in our study is because the effect is
truly absent, for instance by looking
specifically at how similarity with another
individual (on a physical level) affects CY. In
humans, there is ample evidence that the more
similar that individuals are in terms of
physical characteristics, but also personal
convictions and views, the more likely they are
to automatically mimic behavior85.
-
- Future studies can improve on the current
study design in several ways. First, we only
used orangutan males as stimuli. In previous
studies with chimpanzees10 and bonobos15, the
sex of the triggering yawner affected the
occurrence of CY; i.e. in chimpanzees, male
yawns were more contagious whereas in bonobos,
female yawns were more contagious. In gelada
baboons, CY is more prevalent among females,
especially when they are closely bonded6. It is
possible that these results can be explained by
emotional closeness between individuals, as in
chimpanzees males typically form strong social
relationships86, and in bonobos and gelada
baboons it is mostly females that bond87,88.
Alternatively, results could be explained by the
differences in hierarchy with chimpanzees being
male dominant89 and bonobos female dominant87,
and by the strong matrilineal bonds between
gelada baboons90. Investigating whether there is
an interaction between sex of the stimulus and
of the responder in orangutans could help
elucidate the roots of the observed sex effects
in CY in some species. The restricted selection
of stimuli and the low sample size did
unfortunately not allow us to perform such
analyses. It is noteworthy, however, that the
males in our study yawned more frequently than
the females (i.e. the total yawning rate of
males was 74, whereas females yawned only 9
times. See Table S1a). Yawns occur more
frequently in males of species with canine
polymorphism, and also during aggressive
contexts91. Given that all our stimuli were
male, perhaps there is a role for dominance or
rivalry in the occurrence of CY in orangutans23.
Nevertheless, one could argue that this leads to
tense situations, thus leading to more
scratching when observing yawns of others, which
is not what we found.
-
- Additionally, all of our videos contained
flanged males. Flanged adult males are often
preferred over unflanged males by receptive
female orangutans92, and can be viewed as
threatening by unflanged males93. As such, in
addition to interactions between the different
sexes and CY, it may also be interesting to
study potential effects of the two different
morphs of orangutan males on CY.
-
- Furthermore, due to power issues, we could
not reliably test effects of age on CY. In
humans, while spontaneous yawns can occur
already before birth94, CY does not seem to
appear until the age of four to five95,96,
although when children of 3 years old are
specifically told to look at the eyes of the
stimulus they show CY as well97. Similar
developmental trajectories of CY have been
reported in other animals6,7,11,50. In our
study, there were only two individuals younger
than 5; one 15 months (Indah) and one three-year
old (Baju). We observed one yawn occurrence in
Indah (in the yawn condition), in Baju we
observed six events (four in the yawn and two in
the control condition). We decided to include
these individuals in our study because while it
is true that CY shows a relatively slow
developmental pattern in humans, orangutans are
born more precocial, and developmental rates in
nonhuman primates are much faster compared to
humans98. Therefore, CY may possibly also occur
earlier in development in orangutans, but with
only anecdotal evidence we cannot verify this in
our study.
-
- Third, while we tested effects of
familiarity in our study by including both
familiar and unfamiliar yawners, the fact that
we only had yawns from the two adult males to
use as stimuli restricted any potential
investigation of the potential link between
social closeness of the responders and the
familiar individuals on the stimuli. The
positive effect of social closeness on the
occurrence of CY is well established in
humans99, chimpanzees (but see10), and
bonobos15, but is strongly debated in other
species such as dogs45 and budgerigars51. For
dogs, it should be noted that CY is
interspecific, and that domestication might have
had influential effects on how CY is modulated.
Inverse effects have also been reported. For
instance, a large study in rats has shown a
familiarity bias in the opposite direction with
rats being more likely to yawn in response to
unfamiliar yawns23. Similarly, a recent study
investigating scratch contagion in orangutans
found that during tense situations, orangutans
are more likely to take over scratching from
individuals with whom they have a weak bond80,
indicating a (negative) correlation between
social closeness and the contagiousness of a
behavior or motor pattern. Thus, it remains
possible that social modulation of CY is present
in orangutans, at least in those living with
conspecifics in captivity, although its presence
was not shown in our sample. Yet, given our
small sample size, replications that test for
the presence and subsequent direction of social
modulation of CY in orangutans are needed.
-
- Finally, we could not quantify attention to
the screen, which is one of the common
methodological issues raised by Massen et al.
(2017). We tried to maximize attention to the
screen by using attention-grabbing videos of
caretakers at the start of every video sequence,
and by adding colored screens in-between
stimulus presentations. Furthermore, we made
sure that orangutans had a direct line of sight
towards the screen at the start of the
experiment, and only recorded yawns when they
directed their attention to the screen at least
once during stimulus presentation. Nevertheless,
quantification of attention to the stimuli
(either measured as a continuous variable or a
frequency of gazes) remains the most robust way
to control for potential effects of attentional
bias.
-
- To summarize, our findings contribute
to understanding the evolutionary basis of CY in
hominids by showing that orangutans, like
humans, chimpanzees and bonobos, yawn
contagiously.
-
-
-
-
|
