Le bâillement, du réflexe à la pathologie
Le bâillement : de l'éthologie à la médecine clinique
Le bâillement : phylogenèse, éthologie, nosogénie
 Le bâillement : un comportement universel
La parakinésie brachiale oscitante
Yawning: its cycle, its role
Warum gähnen wir ?
Fetal yawning assessed by 3D and 4D sonography
Le bâillement foetal
Le bâillement, du réflexe à la pathologie
Le bâillement : de l'éthologie à la médecine clinique
Le bâillement : phylogenèse, éthologie, nosogénie
 Le bâillement : un comportement universel
La parakinésie brachiale oscitante
Yawning: its cycle, its role
Warum gähnen wir ?
Fetal yawning assessed by 3D and 4D sonography
Le bâillement foetal

mystery of yawning 





mise à jour du
30 novembre 2019
Sci Rep.
Spontaneous Yawning and its Potential Functions
in South American Sea Lions
(Otaria flavescens)
Palagi E, Guillén-Salazar F, Llamazares-Martín C.


-Palagi E, Leone A, Mancini G, Ferrari PF. Contagious yawning in gelada baboons as a possible expression of empathy. Proc Natl Acad Sci USA. 2009;106(46):19262-19267
-Palagi E, Norscia I, Demuru E. Yawn contagion in humans and bonobos: emotional affinity matters more than species Peer J 2:e519
-Palagi E, Norscia I, Cordoni G. Lowland gorillas (Gorilla gorilla gorilla) failed to respond to others' yawn: Experimental and naturalistic evidence. J Comp Psychol. 2019
-Palagi E, Guillén-Salazar F, Llamazares-Martín C. Spontaneous Yawning and its Potential Functions in South American Sea Lions (Otaria flavescens). Sci Rep. 2019;9(1):17226.
Spontaneous yawning is a widespread behaviour in vertebrates. However, data on marine mammals are scarce. In this study, we tested some hypotheses on the functions of yawning in a captive group of South American sea lions (Otaria flavescens). According to the Dimorphism Hypothesis, species showing low levels of sexual dimorphism in canine size do not show sex differences in yawning distribution; this was supported by our findings, since yawning did not differ between the sexes.
Yawning was more frequently performed during resting/sleeping contexts, thus supporting the Drowsiness Hypothesis. Yawning and self-scratching are considered reliable indicators of short-term anxiety in sea lions, since they immediately increased after conflicts both in aggressors and victims (Social Distress Hypothesis supported). In the long-term, yawning was not correlated with individuals' dominance status, thus showing that anxiety is similarly experienced by dominants and subordinates.
The last two findings can be explained by the social competition of this species, that involves individuals independently from their sex, age or ranking status. Therefore, the exposure to frequent stressful events can induce similar levels of anxiety in all the subjects (Resource Inequity Hypothesis supported). In conclusion, spontaneous yawning in sea lions seems to share similar functions with other social mammals, suggesting that this behaviour is a possible plesiomorphic trait.
Le bâillement spontané est un comportement répandu chez les vertébrés. Cependant, les données sur les mammifères marins sont rares. Dans cette étude, les auteurs ont testé certaines hypothèses sur les fonctions du bâillement dans un groupe captif d'otaries du Sud (Otaria flavescens).
Selon l'hypothèse du dimorphisme, les espèces présentant de faibles niveaux de dimorphisme sexuel, en taille de leurs canines, ne montrent pas de différence liée au sexe dans leurs différents bâillements. Leurs conclusions corroborent ces données, puisque les bâillements ne diffèrent pas entre les sexes.
Les bâillements sont plus fréquents lors des transitions repos / sommeil, corroborant ainsi l'hypothèse d'un lien avec la somnolence.
Le bâillement et l'auto-grattage sont considérés comme des indicateurs fiables d'une anxiété récente chez les lions de mer, puisqu'ils augmentent aussitôt après les conflits, tant chez les agresseurs que chez les victimes (hypothèse de détresse sociale soutenue).
À long terme, le bâillement n'est pas corrélé au statut de dominance des individus, ce qui montre que l'anxiété est vécue de la même manière par les dominants et leurs subordonnés. Les deux dernières conclusions peuvent être expliquées par la compétition sociale de cette espèce, qui implique des individus indépendamment de leur sexe, de leur âge ou de leur statut hiérarchique. Par conséquent, l'exposition à des événements stressants fréquents peut induire des niveaux d'anxiété similaires chez tous les sujets (hypothèse d'injustice relative aux ressources appuyée).
En conclusion, le bâillement spontané chez les lions de mer semble partager des fonctions similaires avec d'autres mammifères sociaux, ce qui suggère que ce comportement est un trait plésiomorphique possible.
Yawning is a ubiquitous behaviour present in all classes of vertebrates. It is a stereotyped behaviour that once triggered, is uncontainable and unstoppable. In air-breathing animal species, the first phase of a yawning event is characterized by a slow and wide opening of the animal's mouth and is accompanied by a deep inhalation. The second phase includes a quick closure of the mouth and a short exhalation. Depending on the species, yawning may be also accompanied by eye closing, vocalizations, body stretching, pandiculation and even tongue protrusion.
Many hypotheses regarding the potential functions of spontaneous yawning have been proposed. These hypotheses can be classified into two groups, the physiological and social hypotheses. In the former group, yawning is posited to act as a homeostatic restoring mechanism, and, with the intensification of the studies, some of these hypotheses such as brain cooling, anxiety and drowsiness hypothesis have found an increasing support.In the latter group, the performance of yawning is posited to express the emotional state of the yawner as a communicative tool (e.g., threat yawns), since other members of the group would be able to associate it with certain contexts or behaviours and so could act accordingly. Moreover, yawning has been proven to have an infectious nature (contagious yawning), suggesting that it may be a primitive form of emotional contagion.
Spontaneous yawning is mainly displayed during resting contexts characterized by the absence of changes in social and environmental stimuli. The effect of yawning can be linked to a state of drowsiness in which animals may change from an awake to a sleep phase and vice versa1. It has been proposed that the role of yawning in such situations is to increase the alertness state, thus making animals able to respond effectively to sudden and urgent situations.
Spontaneous yawning can also be affected by social stimuli and conditions. For example, in birds, rats, lemurs and monkeys, yawning increases after an anxiogenic event which can affect the homeostasis of the subject. In the wild, Goodall observed that chimpanzees yawned more frequently in the presence of human observers and, in captivity, the presence of humans in front of animal facilities produced an increase of yawning in monkeys (lion-tailed macaques, Macaca silenus). For this reason, along with self-directed behaviours such as self-grooming and self-scratching, yawning can be considered an indicator of anxiety. For example, in chimpanzees, Baker and Aureli showed that individuals tended to yawn and self-scratch more frequently during high social tension conditions that provoked an increase of anxiety and arousal in the subjects. Recent findings suggest a possible effect of yawning as a stress-releaser, which helps restoring physiological/emotional homeostasis.
In species showing an evident sexual dimorphism in their canine size, yawning is more frequent in males than in females. The development of canines in males is a secondary sexual characteristic that provides them advantages both in intra- and inter-sexual competition. The canine exposure during a yawning event may convey information about the transitional phase experienced by the yawner that can pass from a relaxed to a tense emotional state or vice versa. Hence, yawning can anticipate a forthcoming aggressive behaviour ("threat displays") or a reduction of the levels of anxiety in the yawner.
In-group conflicts are undoubtedly a source of social tension for both aggressors and victims, which often increase self-directed behaviours in the immediate period after an agonistic event. This is also true for yawning, which shows a peak in the minutes after the end of an agonistic contact (Nazca booby, Sula granti; ring-tailed lemur, Lemur catta and Verreaux's sifaka, Propithecus verreauxi; macaques, Macaca tonkeana).
This wide array of results strongly suggests the multifunctional nature of this behaviour, which has been widely studied in primates and rodents, but has often been neglected in other species such as marine mammals. Here, we test some hypotheses, not necessarily mutually exclusive, on spontaneous yawning in South American sea lions, a species living in harems and characterized by high levels of competition. Sea lions show a strong sexual dimorphism in body size and morphology (males possess a mane around their heads), but show a reduced sexual dimorphism in canine length. The Sexual Dimorphism Hypothesis predicts that yawning is more frequent in males when the species presents evident sexual differences in canine size. We thus predict that males and females of South American sea lions do not necessarily differ in the performance of yawning (Prediction 1).
In species showing an unequal distribution of social power, subordinates experience greater levels of anxiety in the long-term. Within social groups, hierarchy affects stress levels, which can be evident from the baseline frequency of self-directed behaviours (e.g., with the dominants scratching more than subordinates or vice versa). The self-scratching trend across hierarchy depends on social organization in different species or populations, with subordinates showing the highest long-term stress levels in hierarchically stable social groups of despotic species. According to the Resource Inequity Hypothesis, since yawning is a reliable indicator of stress, it will be more frequently displayed by subordinate than by dominant individuals, reflecting a baseline anxiety in these kinds of social groups. For example, in species such as Japanese macaques (Macaca fuscata) that are characterized by strong power asymmetry, yawning reflects a baseline anxiety state in subordinate individuals that show higher yawning frequencies than in dominant individuals, particularly during periods of social stability. Although South American sea lions establish dominance hierarchies between group members, a strong competition over resources is present at all ages and in both sexes (e.g., females, shade areas for nursing or water access for thermoregulation), thus reducing the monopolization of resources by high ranking individuals. For this reason, we expect that yawning does not strictly follow a dominance gradient in this species (Prediction 2).
If yawning and self-scratching are behavioural responses linked to an anxiety state of South American sea lions in response of an immediate perturbing event (the Social Distress Hypothesis), we expect that after an agonistic event, these two behavioural patterns will increase by following a similar trend in both the aggressor and the victim (Prediction 3).
Finally, if yawning, as it occurs in other mammal species, is a mechanism linked to the alertness state of subjects (the Drowsiness Hypothesis), we expect it to be concentrated during periods of inactivity characterized by an alternation of resting and sleeping phases (Prediction 4).
The aim of this study was to test, for the first time, hypotheses on the potential functions of the spontaneous yawning in a marine mammal species, the South American sea lion (Otaria flavescens). We have to note that the hypotheses tested here are not necessarily mutually exclusive due to the multifunctional nature of yawning.
The LMMs revealed that the variable 'context' strongly affected the distribution of yawning frequency but did not have a strong effect on its mean duration. Yawning was mainly performed in the resting/sleeping context, when sea lions were lay on the platforms maintaining their eyes closed without engaging in any social behaviour with conspecifics (inactive state). They sometimes opened their eyes and, from time to time, moved their body parts. More rarely, their state of alertness was increased and they could interrupt the inactive phase by socially interacting with other sea lions nearby. Despite our difficulty in determining whether animals were actually sleeping, the sea lions often alternated long periods of total inactivity (eyes closed) with short periods of low activity (eyes opened). Our findings are consistent with the association of yawning with drowsiness already found in human and non-human primates. These studies also reported that yawn distribution changed according to the sleep-wake rhythms of the different subjects with peaks of yawning during the transitional phases between the two states.
Sex does not seem to have a strong effect on either the frequency or duration of yawning, thus supporting the Dimorphism Hypothesis. In agreement with our results, in ring-tailed lemurs and Verreaux's sifaka - two prosimian species showing monomorphism in canine and body size - yawning distribution did not significantly differ between sexes. Although our results support the Dimorphism Hypothesis of yawning, due to the small number of males present in our study group, this support must be taken with caution. Due to the social structure of the species (one-male group or harem), the exploration of the Dimorphism Hypothesis of yawning in sea lions would need several data collections in different captive and wild groups to control for the possible male individual variability in yawning propensity and canine size.
Neither the frequency nor duration of yawning was influenced by the NDS values of the subjects (a quantitative measure of hierarchical dominance) thus confirming that in our species, dominants and subordinates do not show a strong bias in the levels of baseline anxiety as a function of a different inequity power distribution (Resource Inequity Hypothesis supported).To explain this finding the social features of South American sea lions have to be discussed. All the subjects of a colony strongly compete for shaded areas for thermoregulation, access to the sea for foraging, mating partners and for enough safe places for lactating/nursing49. Therefore, aggressive interactions are not limited to adults, as female aggression towards pups and infanticide attempts by sub-adult males have also been reported. Therefore, the aggressive propensity characterizing the species can make each animal, independent of the role it takes in an agonistic encounter or its individual features (age, sex or rank), uncertain about the probability of being the recipient of aggression. This is in line with the findings on the distribution of self-scratching and yawning immediately after agonistic interactions. Self-scratching has been proven to be a reliable indicator of anxiety in human and non-human primates, since it has been described in a wide range of stressful situations (e.g., crowded conditions, unfamiliar encounters, proximity of high-dominant individuals or post-conflicts). As for self-scratching, yawning has also been associated with the presence of environmental and social stressful stimuli. In our study, we found that yawning and self-scratching significantly increased immediately after conflict for both the aggressor and the victim. This clearly indicates that both of these behavioural patterns are sensitive to social stressors and may have a similar function in restoring the homeostasis of the subjects independent of the role they play during the conflict.
In conclusion, our findings reveal that yawning is a multifunctional behaviour also seen in the South American sea lion. Spontaneous yawning seems to serve similar functions in different mammals living in well-structured social groups (e.g., primates, rodents and pinnipeds), independent of their phylogenetic distance and terrestrial/aquatic habits, thus suggesting that this behaviour may be a plesiomorphic trait.