Most vertebrates yawn in situations ranging
from relaxation to tension, but only humans and
other primate species that show mental state
attribution skills have been convincingly shown
to display yawn contagion. Whether complex forms
of empathy are necessary for yawn contagion to
occur is still unclear. As empathy is a
phylogenetically continuous trait, simple forms
of empathy, such as emotional contagion, might
be sufficient for non-primate species to show
contagious yawning. In this study, we exposed
pairs of male rats, which were selected for high
yawning, with each other through a perforated
wall and found that olfactory cues stimulated
yawning, whereas visual cues inhibited it.
Unexpectedly, cage-mate rats failed to show yawn
contagion, although they did show correlated
emotional reactivity. In contrast, stranger rats
showed auditory contagious yawning and greater
rates of smell-facilitated auditory contagious
yawning, although they did not show correlated
emotional reactivity. Strikingly, they did not
show contagious yawning to rats from a
low-yawning strain. These findings indicate that
contagious yawning may be a widespread trait
amongst vertebrates and that mechanisms other
than empathy may be involved. We suggest that a
communicatory function of yawning may be the
mechanism responsible for yawn contagion in
rats, as contagiousness was strain-specific and
increased with olfactory cues, which are
involved in mutual recognition.
-Moyaho A et
al Induced grooming transitions and open
field behaviour differ in high- and low-yawning
sublines of Sprague-Dawley rats Animal Behavior
1995;50:61-72
-Moyaho
A, Valencia J Grooming and yawning trace
adjustment to unfamiliar environments in
laboratory Sprague-Dawley rats J Comparative
Psychology 2002;116;(3):263-269
-Moyaho A
et al. Genetic and littermate influences on
yawning in two selectively bred strains of rats
Dev Psychobiol. 2009;51(3):243-248
General introduction
Yawning is a readily observable behaviour
that has been described in many vertebrate
species, including fish (Baenninger 1987), rats
(Anias et al. 1984), birds (Delius 1967) and
primates (Baenninger 1987). Many of these
species, especially mammals, may yawn
concurrently with stretching and spontaneous
penile erections (Holmgren et al. 1985), and
some of them show yawn contagion (Provine 1986;
Anderson et al. 2004). Despite having this
taxonomic information about yawning and the
increasing number of studies addressing its
immediate mechanisms (Collins and Eguibar 2010),
the literature still ignores other important
aspects, such as its biological meaning
(Guggisberg et al. 2010) and whether a wider
range of species are capable of showing yawn
contagion. Animals typically yawn in situations
that vary from calm (Provine et al. 1987) to
mild stress (Voigt and von Helversen 1999;
Voigt-Heucke et al. 2010) to conflict or tension
(Deputte 1994). Such a broad range of situations
suggests, according to the dimensional view of
emotion (Mendl et al. 2010), that yawning
manifests itself in the punishment avoidance
system, from low-arousal positive pleasant
states to high-arousal negative unpleasant
states. This association of yawning with
apparently contrasting emotional states might
explain the diversity of hypotheses about its
function (Smith 1999; Gallup and Gallup 2008;
Guggisberg et al. 2010). Overall, one can
distinguish two classes of hypotheses. One class
considers yawning as a physiological response
associated with the recovery of homeostasis (for
a review see Guggisberg et al. 2010). For
example, yawning may enhance physiological
arousal (Greco et al. 1993) or assist
thermoregulation (Gallup and Gallup 2008). The
second class, which has arisen from studies with
primates, suggests that yawning is a
communicative act (Deputte 1994). Accordingly,
yawning can be a displacement behaviour (Delius
1967), a threat display (Redican 1982) or a way
of synchronising a group's behaviour (Sauer and
Sauer 1967). Although there are some correlative
studies that have provided evidence for some of
these hypotheses, none of them have been
conclusively supported.
Contagious yawning&emdash;an increase in the
probability of yawning by a subject perceiving
another subject yawning&emdash; is a rare trait
shown by the few primate species that presumably
exhibit cognitive empathy (Provine 1986;
Anderson et al. 2004). However, according to the
perception&endash; action model (Preston and de
Waal 2002), empathy, which refers to an
individual's sensitivity to the emotional state
of others (Preston and de Waal 2002), is a
phylogenetically continuous phenomenon, and
therefore, even small mammals should show some
form of it. In fact, cagemate mice (Mus
musculus) showed emotional contagion&emdash; the
tendency to automatically mimic and synchronise
expressions, vocalizations, postures and
movements with those of another subject's and,
consequently, to converge emotionally&emdash;by
intensifying their response to pain when
perceiving other mice in pain (Langford et al.
2006). Additionally, rats (Rattus norvegicus)
showed empathic concern by liberating a cage
mate trapped in a restrainer (Ben-Ami Bartal et
al. 2011), although the interpretation of this
finding has recently been challenged (Silberberg
et al. 2014). Therefore, it is not surprising
that other species such as dogs (Canis
familiaris) and budgerigars (Melopsittacus
undulatus) may also show yawn contagion
(Joly-Mascheroni et al. 2008; Miller et al.
2012), although the evidence is still
controversial (Harr et al. 2009; O'Hara and
Reeve 2011).
Yawn contagion has generally been elicited
through the use of visual cues. For example,
human subjects (Homo sapiens) yawn when they
watch other subjects yawning on a TV screen
(Provine 1986) and chimpanzees (Pan troglodytes)
yawn when they watch animated chimpanzee yawns
(Campbell et al. 2009). Nonetheless, other
sensory cues, such as sounds, may make human
individuals feel an urge to yawn (Arnott et al.
2009), and because animal species vary in their
ecology, yawn contagion may use species-specific
senses. The present paper will examine three
aspects of yawn contagion using male rats that
were obtained from two populations of rats that
were selected for high yawning (HY) and low
yawning (LY) (Urba´-Holmgren et al. 1990).
First, we expose dyads of HY rats to different
sensory cues to determine whether they have a
differential effect on yawn contagion
susceptibility. Second, we further analyse which
sensory cues are involved in yawn contagion.
Finally, we assess whether the high-yawning rats
contagiously yawn to the low-yawning strain,
taking advantage of their genetic difference in
yawning rates.
General discussion
This study provides evidence for yawn
contagion in rats and suggests that it most
likely arises from a mechanism other than
empathy. Yawn contagion in HY rats may not be
the result of synchronised activity, as has been
claimed to happen, for instance, in lions
(McKenzie 1994). Circadian rhythms, such as
feeding and the sleep/wake cycle that are
associated with yawning (Holmgren et al. 1991),
remained unaltered throughout the test
situations. Stress-related yawning did not
account for yawn contagion because the results
persisted after we statistically controlled for
variation in defecation rates in the rats.
Therefore, we can confidently conclude that rats
did indeed demonstrate yawn contagion.
The stimulating role of olfaction on
auditory contagious yawning reflects the
importance of olfactory cues in social
interactions in most mammals (Johnston 2003).
The variety of chemicals that are released with
bodily secretions and urine during such
interactions provides information about
individual identity, sex and strain (Brennan and
Zufall 2006), and rats use this information to
distinguish between group members (Bradbury and
Vehrencamp 1998). In contrast, visual contact
seems not to convey such information, perhaps
owing to the impairment of albino rats from
seeing as pigmented rats normally do (Prusky et
al. 2002). This impairment may explain the
slight effect of visual cues on the yawning of
the rats in this study, which contrasts with the
function in primates, in which they undoubtedly
play a stimulating role (Provine 1986; Platek et
al. 2003; Anderson et al. 2004). Alternatively,
an evolutionary divergence in the type of
sensory modalities used in social interactions
may explain the opposite effect of visual cues
on yawning between primates and rodents
(Bradbury and Vehrencamp 2011). Both groups of
species share, however, the use of auditory cues
in contagious yawning, which suggests that this
sensory modality may have been involved in
contagious yawning earlier than vision.
The finding that stranger NVOC and OC rats,
rather than cage-mate rats, showed yawn
contagion contrasts with the prediction that
empathic processes are more likely to occur in
familiar animals (Preston and de Waal 2002;
Langford et al. 2006). Therefore, mechanisms
other than empathy are most likely involved in
contagious yawning in rats. Imitation cannot be
suggested as a valid process of yawn spreading
because yawning cannot be considered as a novel
or improbable response in the repertoire of the
rats, which is a necessary condition for
imitative behaviour (Byrne 1994; Zentall 2011).
Our results suggest the existence of a social
interaction effect that is more specific than an
indiscriminate influence on yawning. A social
facilitation mechanism, such as contagious
behaviour (Byrne 1994), may suffice for
contagiousness to occur. In such a case, yawning
would be released if the releasing stimulus
(e.g. the sound of a yawn) happened to be
another rat behaving in the same manner (Yoon
and Tennie 2010). However, this mechanism ought
to produce yawn contagion in both cage-mate and
stranger HY rats. Because stranger NVOC- and
OC-HY rats yawned at each other, contagion seems
to have simply occurred as a consequence of
using yawning as a form of communication. This
mechanism does not rely on empathy based on the
fact that only stranger NVOC- and OC-HY rats
showed contagious yawning.
Why did cage-mate and stranger HY rats
differ in their ability to show yawn
communication? The difference may be because of
the way in which they coped with the test
situations. Individual recognition based on
familiarity made cage-mate OC-HY rats, although
not NVOC-HY rats, empathise emotionally
(Hatfield et al. 1993) because their defecation
rates positively correlated. Nonetheless, the
yawn they showed was unlikely to influence each
other's behaviour; it possibly served for
monitoring their internal state. In contrast,
individual recognition based on familiarity
failed in stranger NVOC- and OC-HY rats. The
encounter between them could lead to a conflict
situation because they were placed next to each
other in a limited area in which they very
likely simultaneously attempted to gain control
(Hurst and Beynon 2004; Bradbury and Vehrencamp
2011). The conflict most likely led the rats to
gather information about the opponents' fighting
ability and motivation using cues and signals.
The rats could benefit if the conflict could be
settled without a fight, and therefore, they
would somehow signal their fighting ability to
convince each other to retreat. Yawning of HY
rats might be correlated with dominance status
through its ability to encode physiological
condition. The positive association between
spontaneous penile erections and yawning
underpins this suggestion because there is a
physiological link between the two traits that
depends on steroid hormones, such as
testosterone (Huston 1971; Holmgren et al. 1980;
Phoenix and Chambers 1982; Melis et al. 1994).
Androgen hormones, like testosterone, facilitate
aggressive behaviour, which is expected to
characterise dominant rats. Therefore, yawn rate
would be a reliable signal for indicating
physiological capacity&emdash;an animal's
overall quality in terms of acquired endurance,
due to circulating levels of testosterone,
necessary for prolonged displays and agonistic
interactions (Sinervo et al. 2000)&emdash; which
only dominant rats could attain.
Male&endash;male conflicts might explain the
finding that the yawning of littermates in HY
male rats increased with male-biased sex ratios
(Moyaho et al. 2009) and support a connection
between yawning and dominance status. This
evidence agrees with the observation that
dominant males out-yawn subordinate males in
stump-tailed macaques (Macaca arctoides) (Adams
and Schoel 1982) and is consistent with the
increase in yawning rate in male&endash;male
encounters in Siamese fighting fish (Betta
splendens) (Baenninger 1987). The finding that
HY rats possibly inhibited LY's yawning might
also be because of a male&endash;male conflict
in which a HY dominant male out-yawns a
subordinate LY one. However, further experiments
will be necessary for supporting this
hypothesis.
In summary, the findings of this study
suggest that yawn contagion is more common than
was previously thought and that it may result
from a communicatory function of yawning. We
believe that this function may advertise greater
physiological capacity during male&endash;male
conflicts.
