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 



haut de page













mise à jour du
24 février 2021
Dev Psychobiol.
2021 Jan 27.
doi: 10.1002/dev.22094
 Earlier than previously thought:
Yawn contagion in preschool children
Cordoni G, Favilli E, Palagi E.


 Tous les articles sur la contagion du bâillement
All articles about contagious yawning

Yawning is a primitive and stereotyped motor action involving orofacial, laryngeal, pharyngeal, thoracic and abdominal muscles. Contagious yawning, an involuntarily action induced by viewing or listening to others' yawns, has been demonstrated in human and several non-human species. Previous studies with humans showed that infants and preschool children, socially separated during video experiments, were not infected by others' yawns. Here, the authors tested the occurrence of yawn contagion in 129 preschool children (ranging from 2.5 to 5.5 years) belonging to five different classes by video recording them in their classrooms during the ordinary school activi- ties. As it occurs in adult humans, children of all ages were infected by others' yawns within the 2 min after the perception of the stimulus. The yawn contagion occurred earlier than previously thought. For children, it appears that the natural social set- ting is more conducive to yawn contagion than the inherently artificial experimental approach. Moreover, children's gender did not affect the level of contagious yawn- ing. The neural, emotional and behavioural traits of preschool children are probably not sufficiently mature to express variability between boys and girls; nevertheless, children appeared to be already well equipped with the 'neural toolkit' necessary for expressing yawn contagion.

Le bâillement est une action motrice primitive et stéréotypée impliquant les muscles orofaciaux, laryngés, pharyngés, thoraciques et abdominaux. Le bâillement contagieux, une action involontaire induite par l'observation ou l'écoute des bâillements d'autrui, a été démontré chez l'homme et plusieurs espèces non humaines. Des études antérieures chez les humains ont montré que les nourrissons et les enfants d'âge préscolaire, séparés socialement lors d'expériences vidéo, n'étaient pas réceptifs aux bâillements des autres. Ici, les auteurs ont testé la survenue de contagion du bâillement chez 129 enfants d'âge préscolaire (de 2,5 à 5,5 ans) appartenant à cinq classes différentes en les enregistrant dans leurs salles de classe pendant les activités scolaires ordinaires. Comme cela se produit chez les adultes, les enfants de tous âges ont été réceptifs aux bâillements des autres dans les 2 minutes suivant la perception du stimulus. La contagion du bâillement s'est produite plus précocément que prévu. Pour les enfants, il apparaît que le cadre social naturel est plus propice à la contagion du bâillement que l'approche expérimentale intrinsèquement artificielle. De plus, le sexe des enfants n'affectait pas le niveau de contagion. Les traits neuraux, émotionnels et comportementaux des enfants d'âge préscolaire ne sont probablement pas suffisamment mûrs pour exprimer la variabilité entre garçons et filles; néanmoins, les enfants semblaient déjà bien équipés de la «boîte à outils neuronale» nécessaire pour exprimer la contagion du bâillement.

"Imagine a yawn. You stretch your jaws open in a wide gape, take a deep inward breath, followed by a shorter exhalation, and end by closing your jaws...You have just joined vertebrates everywhere in one of the animal kingdom's most ancient rites" (Provine, 2005 p. 532).
Yawning is a primitive and complex stereotyped motor response that requires the involvement of facial, oral, laryngeal, pharyngeal, thoracic and abdominal muscles (Provine, 2005). Among vertebrates, spontaneous yawning is a ubiquitous and evolutionarily conserved behaviour (Provine, 2005; Walusinski & Deputte, 2004). This activity
seems to serve important neurophysiological functions such as brain cooling and enhancement of blood flow to the skull, which, in turn, could stimulate cortical arousal and state change (for an extensive re- view see Massen & Gallup, 2017).
Spontaneous yawning is driven by physiological stimuli, it is widespread throughout the animal kingdom and it follows a daily fluctuation (Giganti & Zilli, 2011; Massen & Gallup, 2017). On the other hand, contagious yawning, an involuntarily action induced by viewing or listening to others' yawns (Provine, 2005), is socially driven and must follow a different 'pathway' compared to spontaneous yawning.
Indeed, contagious yawning shows some differences in daily fluctuation (Giganti & Zilli, 2011) and it is not widespread across different animal species; even, this phenomenon occurs in some but not all species belonging to the same family. For example, within the hominid family, yawn contagion has been shown in chimpanzees (Campbell & Cox, 2019; Campbell & de Waal, 2011), bonobos (Palagi et al., 2014) and humans (Provine, 2005) but not in gorillas (Palagi et al., 2019), thus suggesting that phylogenetic closeness cannot necessarily predict, per se, yawn contagion (Palagi et al., 2020). Outside the primate order, evidence of contagious yawning are reported for dogs and wolves (Neilands et al., 2020; Romero et al., 2013, 2014; Silva et al., 2012), Asian and African elephants (Rossman et al., 2017; Rossman et al., 2020), sheep (Yonezawa et al., 2017), elephant seals (Wojczulanis-Jakubas et al., 2019) and parrots (Gallup et al., 2015; Miller et al., 2012). The occurrence of yawn contagion may be linked to the level of attention and responsiveness to conspecifics' stimuli. Studies car- ried out on humans showed a positive association between the prolonged duration in attentively fixing the eye region of fellows and the increase in the detection sensitivity to others' expressions including yawning (Bal et al., 2010; Chan & Tseng, 2017). In this view, in highly social species (e.g., wolves, chimpanzees, bonobos), the higher levels of affinitive interactions can promote the access to others' faces and facilitate the detection of others' expressions (Cordell & McGahan, 2004; Croes et al., 2020; Farroni et al., 2002). This condition may also favour the occurrence of yawn contagion between subjects. Indeed, in species with high exchange of affinitive and cooperative interactions, contagious yawning occurs at elevate rates (e.g., chimpanzees, Campbell & Cox, 2019; Campbell & de Waal, 2011; bonobos, Palagi et al., 2014; wolves and do- mestic dogs, Neilands et al., 2020; Romero et al., 2013; Romero et al., 2014; Silva et al., 2012; elephant seals, Wojczulanis-Jakubas et al., 2019; budgerigars, Gallup et al., 2015). In these species, the emergence of yawn contagion might be favoured by natural se- lection in order to promote synchronization between individuals in spatial ranging, coordinated foraging, predator defending and sleep/wake rhythm (Palagi et al., 2020).
The proximate factors underpinning contagious yawning are still under debate. For some researchers, yawn contagion can be the expression of a motor response whose individual variation may be related to the different levels of social attention by the observers (Attentional Bias Hypothesis, Massen & Gallup, 2017). In this view, those individuals that are considered more 'valuable' to the observer catch his/her attention more than the less 'valuable' companions do (Massen & Gallup, 2017; Massen et al., 2012; Yoon & Tennie, 2010).
On the other hand, the individual variability in contagious yawning according to the sex, age and level of familiarity between sub- jects (see Palagi et al., 2020 for an extensive review) could be linked to the individual variability in emotional contagion (Emotional Bias Hypothesis; Preston & de Waal, 2002). Following the Perception Action Model (Preston & de Waal, 2002), emotional contagion can be considered as a basic form of empathy defined as '[...] Any process that emerges from the fact that observers understand others' states by activating personal, neural and mental representations of that state, including the capacity to be affected by and share the emotional state of another; assess the reasons for the other's state; and identify with the other, adopting his or her perspective' (de Waal & Preston, 2017 p 498). Many neurological studies have demonstrated the involvement of the mirror neuron system in the emotional contagion phenome- non (Decety & Lamm, 2006; Gallese, 2007; Preston & de Waal, 2002; Rizzolatti et al., 2001; Uddin et al., 2007). The mirror neuron system is a network of visuomotor neurons firstly discovered in the area F5 of the macaque pre-motor cortex (Rizzolatti et al., 1996). In hu- mans, the mirror neuron system has been discovered in the premotor, primary somatosensory and inferior parietal parts of the cortex (Mukamel et al., 2010; Rizzolatti & Craighero, 2004). This set of neurons discharges both when a subject executes a motor action and when the same subject observes another individual perform- ing the same motor action (Ferrari et al., 2003). The firing of mir- ror neurons causes in the observer a similar physiological reaction occurring in the demonstrator thus stimulating a shared process between subjects. Hence, several studies have suggested that the mirror neuron system is involved in emotional sharing and other empathy-linked phenomena (Decety, 2010; Gazzola et al., 2006; Hess & Fischer, 2014; Palagi et al., 2020; Preston & de Waal, 2002; Prochazkova & Kret, 2017).
In their research, Haker and colleagues (Haker et al., 2013) employed functional magnetic resonance imaging to assess human brain activity in response to others' yawns. When subjects observed videos of yawning faces as opposed to faces with a neutral expression, there was activation in right Brodmann's area 9, a portion of the right inferior frontal gyrus which is a region of the mirror neuron sys- tem. Since Brodmann's area 9 is involved in higher social cognitive processes (e.g., mentalizing, Ohnishi et al., 2004), the authors pro- posed the existence of a linkage between the mirror neuron system and more complex cognitive functions such as cognitive empathy. Cognitive empathy is a top-down process that consciously evaluates information not directly observable (e.g., taking another perspective; de Waal & Preston, 2017). The involvement of higher cognitive functions, which are not fully developed at birth, could explain why contagious yawning occurs later throughout the human development (Haker et al., 2013). Some experimental studies on yawn contagion in humans showed that both infants and preschool children were not infected by yawns of both familiar and unfamiliar subjects. Nevertheless, it is worth noting that children were not exposed to live stimuli (e.g., colour videotapes of adult individuals while yawning or smiling; Anderson & Meno, 2003) and were separated from their peers during the experiment (e.g., each child individually tested in a quiet room; Anderson & Meno, 2003).
Helt and co-workers (Helt et al., 2010), in order to assess the level of susceptibility to yawn contagion of both autistic and typically de- veloping children, employed a different experimental procedure by exposing subjects to a live, adult model. Yawn contagion in typically developing children was almost absent between 1 and 3 years of age and not strongly evident in children of 4&endash;5 years old (11 out of 40 subjects). In a second set of subjects, the authors showed that children with autistic disorders were less susceptible to yawn contagion compared to typically developing children.
Hoogenhout and co-authors (Hoogenhout et al., 2013) investigated whether cuing eye contact could influence yawn contagion in children. The authors presented both yawning and control video clips to the child while an examiner sat next to the subject by re- minding him/her to maintain eye contact with the screen. They re- corded yawning response only in one of six 3 years old children and seven of 17 older subjects (4&endash;5 years of age).
We carried out an ethological study on 129 preschool children aged from 2.5 to 5.5 years belonging to five different classes of an Italian public kindergarten. We gathered data on yawning events by following the children in their classrooms during their ordinary activities in order to test some hypotheses on spontaneous and conta- gious yawning in humans during the preschool phase.
Hypothesis 1 The effect of age on spontaneous yawning.
The frequency of spontaneous yawning during the 24 h changes throughout human life (Giganti & Zilli, 2011). For example, in preterm and near-term infants, the mean rate of yawning over a 24 h pe- riod is 1.10 yawn/h (Giganti et al., 2007), whereas in young adults (17&endash;35 years old) this activity is reduced to a mean of 7.1 yawn/day, that is about 0.3 yawn/h (Baenninger et al., 1996). If, at a finer scale, the age affects the frequency of spontaneous yawning also during the preschool phase (Hypothesis 1a), we predict that spontaneous yawning is more frequent in younger (2.5 years of age) than in older children (5.5 years of age) (Prediction 1a).
Recently, Gallup and colleagues (Gallup et al., 2016) demon- strated that both the frequency and the duration of spontaneous yawning are variable. The authors showed that, across mammal spe- cies, the duration of yawning is strictly correlated with the average brain weight and cortical neuron number. The larger-brained mam- mals, such as primates, in general and humans, in particular, tend to perform longer yawns than other mammals do. Furthermore, the duration of spontaneous yawning seems to be age related, at least in humans, with adults performing longer yawns than children (Gallup et al., 2016). During the preschool years children's brains are growing in multiple dimensions, including volume, cortical thick- ness and neural network (i.e., brain complexity and weight; Brown & Jernigan, 2012; Hagmann et al., 2010; Krogsrud et al., 2016; Prochazkova & Kret, 2017). If in the preschool phase the duration of spontaneous yawning is sensitive to the neural changes linked to the different ages of children (Hypothesis 1b), we predict that the duration of spontaneous yawning increases along with the age in the preschool period (Prediction 1b).
Hypothesis 2 The effect of age on the occurrence of yawn contagion.
Several behavioural studies highlighted the early develop- ment of emotional contagion in humans that is largely reflected in emotional mimicry (Decety, 2010; Leppa_nen & Nelson, 2008; Panksepp & Panksepp, 2013). For example, newborns and infants become vigorously distressed by another cry (Dondi et al., 1999; Trevarthen, 2005) and, from the second half of the first year of life, children become able to identify and mimic discrete facial expressions of emotion such as smiles (Leppa_nen & Nelson, 2008). The mechanism allowing a child to reflexively mimic a smile is thought to be the same allowing the reflexive mimicry of yawning (Dimberg et al., 2000). Additionally, children are able to distinguish yawning movements from other mouth movements from 5 months of age. The view of yawning motor patterns activates neurons in the tem- poral areas of 5/8-month-old infants thus suggesting that the neural mechanism underpinning yawning perception begins early in infancy (Tsurumi et al., 2019). If the neural mechanism allowing children to reflexively mimic others' yawns has already developed at the beginning of preschool period (Hypothesis 2a), we predict that, starting from 3 years of age, children can not only perceive but also respond to others' yawns in a congruent manner (Prediction 2a).
The ability of children to process others' facial expressions and, consequently, to correctly perceive others' emotions seems to follow a gradual and continuous development starting from 7 to 8 months of age (Batty & Taylor, 2006; Grossard et al., 2018; Xie et al., 2019). Some studies have investigated the length of the emo- tional processing in human subjects by recording event-related potentials (Boutet et al., 2020; Grossard et al., 2018; Kappenman & Luck, 2012; Xie et al., 2019). Batty and Taylor (2006) studied the developmental pathway of facial emotion perception in children (4&endash;15 years) and found that the latency in face processing decreased with age. The authors suggested that, even if the neural processes involved in the facial expression recognition and emotional perception are at work from early childhood, the processing of emotions develops and matures on a stepwise fashion until adulthood. If older children (4&endash;5 years old) have developed more 'neural experience' in facial expression processing and emotional perception (Hypothesis 2b), we predict that the latency in responding to the perceived yawn decreases along with the increase in age (Prediction 2b).
Hypothesis 3 The effect of gender on the distribution of yawn contagion.
The contagious aspect of yawning shows high individual variability in humans (Bartholomew & Cirulli, 2014; Takahiro & Akio, 2018). Yawn contagion can be affected by different variables, such as the time of the day (Giganti & Zilli, 2011), age of the sub- jects (Bartholomew & Cirulli, 2014) and level of familiarity between individuals (Norscia & Palagi, 2011). A 5-year long naturalistic observation on humans demonstrated that females were more susceptible to others' yawns than males (Norscia et al., 2016). The authors discussed their result in the light of the empathic nature of yawn contagion and the gender differences in performing empathy-based behaviours with females showing higher empathic abilities. If gen- der affects the sensitivity to others' yawns in the preschool period (Hypothesis 3), we predict that in children, girls show higher levels of contagious yawning compared to boys (Prediction 3).
Hypothesis 4 The effect of sleep/awake activity on the fluctuation of spontaneous and contagious yawning.
The frequency of spontaneous yawning varies across the day- time with an increase in the early morning, just after awakening and in the evening, before sleep (Giganti & Zilli, 2011; Zilli et al., 2008). If also during the preschool phase the spontaneous yawning is linked to the sleep/awake cycle (Hypothesis 4a), we predict that the levels of spontaneous yawns fluctuate over the 8-h school day by showing a peak around the sleeping time after lunch (Prediction 4a). If con- tagious yawning follows different developmental trajectories com- pared to spontaneous one (Anderson & Meno, 2003; Gallup, 2011; Helt et al., 2010; Hoogenhout et al., 2013; Krestel et al., 2018) (Hypothesis 4b), we predict that the fluctuation of contagious yawn- ing is not necessarily linked to the resting/activity phase occurring during the 8-hr school day (Prediction 4b).
In the current study we employed an ethological approach for investigating the dynamics of spontaneous and contagious yawning in preschool children during their ordinary activities in their ordinary social context (i.e., classroom).
As regards spontaneous yawning, our finding demonstrated that none of the tested variables affected the duration of the yawning (Prediction 1b not supported) while the variable gender significantly affected the yawning frequency distribution with boys showing higher levels of yawns compared to girls (Prediction 1a not con- firmed). Several experimental studies highlighted a strict linkage between sex hormones and spontaneous yawning frequency. For example, by treating castrated male rats with testosterone, the lev- els of their spontaneous yawning were raised until the normal levels of intact males (Homgren et al., 1980; Melis et al., 1994). Moreover, the testosterone provided to female rats induced an increase also in their yawning rates (Homgren et al., 1980). In guinea pigs, daily treatment with testosterone potentiated the effectiveness of ACTH in inducing stretch-yawning (Rodriguez-Sierra et al., 1981). Graves and Wallen (2006) demonstrated that the treatment with androgen hormones promoted yawning behaviour in female rhesus monkeys by increasing their yawning levels from 0.3 to 4.7 yawn/min. Also, in Japanese macaques, testosterone induced an increase in yawning behaviour of males. Intriguingly, in this study, the increase in yawning levels positively correlated with the increase in aggression levels (Bethea et al., 2013). It has been suggested that in humans, as in other animals, yawning can be considered as a sign of increasing arousal or stress in the subject (Corey et al., 2011; Matikainen & Elo, 2008; Troisi, 2002; Walusinski, 2006). In a study carried out by Cordoni et al. (2016) on the same preschoolers of the current research, the authors showed that boys of all ages played in a more competitive manner and engaged in more aggressive interactions than girls. Another study on preschool children, by measuring their salivary testosterone concentration, found a positive correlation between the levels of the hormone and serious aggression in boys (Sa_nchez-Marti_n et al., 2000). Hence, if spontaneous yawning ex- presses an arousal state which is linked to testosterone and aggressiveness, we can speculate that in our study the higher levels of spontaneous yawning performed by boys may be due to their higher levels of arousal. It is worth noting that while boys showed more spontaneous yawning than girls, no difference was found in the levels of contagious yawning according to gender. The same result was obtained by Campbell and Cox (2019) for captive chimpanzees. Certainly, further studies applying both hormonal and behavioural evaluation are needed for a deeper investigation of these issues.
Yawning occurs under both spontaneous and contagious conditions (Provine, 2005). Previous studies on humans reported that yawn contagion did not occur below the age of 6 years (Anderson & Meno, 2003; Helt et al., 2010; Hoogenhout et al., 2013; Millen & Anderson, 2011). In the current study, we demonstrated the presence of yawn contagion in 3-year-old children (Prediction 2a confirmed).
A yawning event can be a strong releasing stimulus for a specific behavioural fixed action pattern that is hard wired and on which is built the yawn contagion phenomenon (Provine, 1986; Tinbergen, 1951). In general, by copying companions' actions children increase the synchronization of their activities (e.g., sleep/awake rhythm) that is a widespread adaptive phenomenon in human and non-human ani- mals (Duranton et al., 2017; Duranton & Gaunet, 2016). Both unconscious and conscious replication of other's action can be a strategy to create and strengthen relations with companions and become part of a social group (Lakin & Chart rand, 2003; Lakin et al., 2008). Children synchronize their behaviour to both identify and affiliate with their group members (Over & Carpenter, 2012).
Yawn contagion can be also considered an expression of emo- tional contagion (see Palagi et al., 2020 for an extensive review). At a neural level, subcortical brain systems (including the amygdala) are functional at birth and have a role in orienting infants' attention towards others' faces and in enhancing response activity in specific cortical areas, even though this function probably does not emerge until the second half of the first year of life (see Leppa_nen & Nelson, 2008 for an extensive review). During early childhood the cortical connectivity in frontal, parietal and cingulate areas increases (Long et al., 2017) and shapes the architecture of the 'future' brain before full maturation and stabilization (Boutet et al., 2020). The key components of the emotion-processing neural network seem to emerge at this stage, and children develop the capacity to ex- perience, express and manage emotions concomitantly with the in- crease in motor control and cognitive abilities (Decety, 2010; Tarullo et al., 2009). Taking this evidence into account, we could argue that, at the beginning of the preschool phase (3 years of age), children have already developed the necessary 'neural toolkit' for perceiving others' facial expressions, copying the motor programme underpinning those expressions, and therefore adopting the same emotional state. These phases of neural maturation are probably at the basis of the motor replication and subsequent emotional contagion that allow children to synchronize their behaviour and develop affiliation and social bonding with others (Over & Carpenter, 2012).
The sharing of the natural social setting in which children ordinarily interact can make manifest the expression of some phenomena, such as yawn contagion, thus limiting the inhibition caused by the experimental apparatus or unfamiliar setting employed in previous studies (Anderson & Meno, 2003; Gallup, Church, Miller, et al., 2016; Helt et al., 2010; Hoogenhout et al., 2013). Nakahashi and Ohtsuki (2015) stated that emotional contagion represents an efficient social learning strategy when the trigger and the observer share the same environment or context. In general, individuals are more prone to automatically replicate the emotions of in-group than out-group members thus creating social bonds with companions and, therefore, improving the probability to become part of the group (Seyfarth & Cheney, 2013). Following the 'Like Me' frame- work (Meltzoff, 2007), 'whoever is acting like me, is probably like me' (Rabinowitch & Knafo-Noam, 2015, p. 8). Children attending kindergarten and sharing the same environmental and social context can be affected more by yawns of individuals 'similar to them', such as classmates, rather than by yawns of unfamiliar or dissimilar subjects (e.g., unfamiliar adults; Anderson & Meno, 2003). Those children who developed greater abilities in identifying and sharing others' expressions of emotions can be more prosocially responsive to their peers (Denham et al., 2003).
The motor coordination and synchronous interaction could rep- resent two of the factors that increase the positive social attitude between interacting children thus promoting the development of pro-social behaviours (Rabinowitch & Knafo-Noam, 2015; Shamay- Tsoory et al., 2019). A recent study suggested that motor synchrony, emotional contagion and conformity (i.e., the alignment with group actions, thoughts, perception), even though different phenom- ena, can be intertwined. They may concur in fostering the social alignment between group companions, such as classmates (Social Alignment Model; Shamay-Tsoory et al., 2019).
Intriguingly, different from mimicry of other expressions (e.g. play face, Palagi et al., 2019), yawn contagion usually shows an ex- tended latency between the emission of the yawn by the trigger and the motor response by the observer (chimpanzees, Campbell & Cox, 2019; humans, Provine, 2005). Outside the primate order, several studies have shown that yawn contagion in adults mainly occurs within the first 90 s after the perception of the stimulus. In budgerigars (Melopsittacus undulatus), a social flock-living parrot, individuals show contagious yawning with a latency of 40 s (Miller et al., 2012). Yonezawa et al. (2017) found that sheep yawned within 1 min after the trigger. Similarly, in canids (i.e., wolves and dogs) the latency of yawning response was between 9 and 90 s (Joly- Mascheroni et al., 2008; Romero et al., 2013). In adult human and non-human primates, contagious responses peak within the first minute after the perception of the stimulus (Norscia & Palagi, 2011; Palagi et al., 2009, 2014). In the present study, preschool children, independently from their age, concentrated the majority of their infected yawns in the first 2 min after perceiving the stimulus (Prediction 2b not confirmed). This finding shows that the latency in the yawning response is comparable with those recorded for adult human and non-human primates.
Neither gender nor age influenced the frequency and the duration of contagious yawns (Predictions 3 not supported). While spontaneous yawning can reflect the internal affective state of the subject (e.g., arousal), the contagious yawning seems to belong to a communicative domain. The neural, emotional and behavioural traits of preschool children are probably not yet mature to express variability between boys and girls as well as younger and older children, although they seem to be already well equipped with the 'neural toolkit' for ex- pressing yawn contagion (Decety, 2010; Leppa_nen & Nelson, 2008; Tarullo et al., 2009; Tsurumi et al., 2019). Probably, the age- and gen- der-related difference in yawn contagion shown in adults (Norscia et al., 2016) emerges later in development of adolescence.
It has been proposed that contagious and spontaneous yawn- ing can be driven by different mechanisms (Giganti & Zilli, 2011). In 2011, Giganti and Zilli demonstrated that in young people (mean years 22.9 ± 2.7 SD) the fluctuations of contagious and spontaneous yawns only partially overlapped, although their frequencies over the wakefulness period were both influenced by the hour of the day (i.e., early morning and late evening). Our five kindergarten classes followed the same activity plan and, thus, the wakefulness period was the same for all children. According to Giganti and Zilli (2011), in our study, contagious and spontaneous yawning followed a partial overlapped fluctuation (Prediction 4b supported). Indeed, contagious yawns showed a clear peak of frequency during the guided activities (T2), whereas spontaneous yawns showed a wider peak of frequency from T2 to T4 (Prediction 4a not supported). However, we have to un- derline that our analysis was limited to the 8-hr school day. Further studies are needed to obtain more complete data on the frequency of both types of yawn during the day time.
In conclusion, physiological, neural and behavioural approaches are needed to draw the possible ontogenetic trajectories and modulation of yawn contagion in humans. Yet, we encourage observations carried out under naturalistic conditions that could complement the more controlled approaches carried out under laboratory conditions. Observations in the real world may reveal patterns not found in inherently artificial experimental settings.