128 - 133: Exploring Yawning with Neuroimaging

The genesis and function of the yawn has baffled humans since ancient times. Yawning requires no conscious motor planning and manifests in a stereotypic manner regardless of the animal involved, human or otherwise. This primitive motor program requires the concerted activity of facial, oral, laryngeal, pharyngeal, thoracic and abdominal musculature. Despite the ubiquity of spontaneous yawning across species, only primates [1] and possibly dogs [2, 3] yawn to a contagion, or in response to viewing, hearing or simply thinking about a yawn [4].

Despite numerous theories regarding the mechanism or function of yawning, little is known about the underlying physiological mechanisms of this primitive behavior. Attempts to isolate the neural correlates of yawning have yielded a wide range of results. Involvement of the brainstem cranial nuclei is clear, since these neural structures mediate the outflow to the muscles required to carry out a yawn. Seizures involving the brainstem have been associated with abnormal behaviors including head bobbing, sniffing and yawning [5]. The upstream mechanisms that trigger spontaneous yawning or the brain regions involved in the processing of contagious yawning remain poorly understood.

Various studies have, however, shed light on the neural structures involved in yawning. A report by Jurko and Andy [6] observed yawning during hyperventilation that occurred in certain patients who underwent thalamotomy. All of the lesions that elicited yawning (during the routine recording of electroencephalograms) were localized to the medial portion of the centro-median nucleus of the thalamus. In rats, electrical and chemical stimulation of the paraventricular nucleus (PVN) of the hypothalamus also stimulated yawning [7, 8]. Microinjection of histamine into the medial parvocellular subdivision of the PVN also elicited yawns and an arousal shift was observed using electrocorticography, as evidenced by lower voltage and faster rhythms. Microinjection of HTMT (histamine trifluoromethyl toluidide) dimaleate, a histamine H₁ receptor agonist, into the PVN also caused the yawning/arousal response, whereas pretreatment with pyrilamine, an H₁ receptor antagonist, inhibited the histamine-induced yawning behavior [9].

The advent of high-resolution structural and functional neuroimaging modalities such as MRI and PET provide new and more sensitive methods to explore the neural underpinnings of yawning. Lesion-based inferences utilizing structural imaging have identified additional components of the larger yawn neural network. Walusinski et al. [10] reported that in some cases of hemiplegia following a stroke in the internal capsule, yawning was associated with the involuntary raising of the subject‘s paralyzed arm. Two additional cases of brainstem strokes involving the upper pons and the ponto-mesencephalic junction also presented with transient excessive pathological yawning [11].

Despite the large number of brain regions implicated in spontaneous yawning, little has been ascertained about how these regions interact as a cohesive network to generate a yawn. Newer neuroimaging methods such as blood-oxygen-level-dependent functional MRI (BOLD-fMRI) provide this ability to characterize the neural network. fMRI may also utilize behavioral paradigms to probe the mechanisms underlying contagious yawning. Schurmann et al. [12] utilized BOLD-fMRI to explore the neural correlates of the yawn contagion, and hypothesized that yawning would not significantly activate mirror neuron networks since yawns are produced automatically without the need for imitation. The investigators conducted a block-design experiment whereby participants viewed videos of unknown actors performing yawns or control movements involving the tongue for approximately 25 s, interspersed by periods of rest. Their findings showed differences between yawn and control stimuli in the right posterior superior temporal sulcus (STS) and bilaterally in the anterior STS. In addition, they identified no yawn-specific activations in Broca's area that supported their hypothesis, suggesting the mirror neuron system (MNS) was not involved in the genesis of yawning to a contagion. Finally, they noted that their subjects’ yawn susceptibility negatively covaried with amygdalar BOLD activity and suggested this was related to ‘ the effectiveness of yawn contagion and the face-processing-related emotional analysis during social interaction’.

To expand on prior work demonstrating that susceptibility to contagious yawning was negatively correlated with schizotypal personality traits, Platek et al. [13] conducted an fMRI study to investigate whether theory of mind or empathy networks were activated while subjects viewed yawn videos [14]. This study involved 10 subjects undergoing fMRI while viewing 7-second videos consisting of 1 of 3 conditions: neutral, laughing and yawning. The findings provided evidence for the hypothesis that contagious yawning may be a primitive form of empathic modeling based on the activation of the posterior cingulate and precuneus. Additional activations in the bilateral thalami and parahippocampal gyri were attributed to visual and sensory processing of the stimuli. It should be noted that a more recent fMRI study exploring the neural correlates of hearing a yawn did demonstrate involvement of the MNS [15].

Based on these preliminary findings, our own group hoped to unify and expand on the previous findings regarding the yawn neural network by exploring the brain responses induced when subjects viewed yawns and similar non-contagious facial actions [16]. We used single video clips of 4 seconds’ duration showing an unfamiliar actor performing 1 of 4 facial expressions: yawning, gaping, coughing and a “blank” expressionless face with limited blinking. We used a single stimulus set to avoid confounds related to viewing novel faces or facial expressions. The blank face and cough stimuli served to control for facial perception and visual motion, whereas the gape action involved the actor opening the mouth wide then closing with similar timing to a yawn, but while maintaining the eyes wide open and having a more robotic appearance. Thus gaping mimicked all aspects of facial movement, complexity and timing associated with yawning, without conveying the behavioral or emotional components of yawning. To obtain a more objective measure of the various stimuli, we measured the relative motion characteristics over time (MoCharT) using an inhouse [Matsuhashi M, unpublished] Matlab script (Mathworks) that compared frame-by-frame pixel changes as an approximation of overall motion (fig. 1).

The shorter stimulus duration and slow event-related design we implemented allowed for the display of a complete yawn cycle, while minimizing potential confounders such as response habituation or potentiation. Eighteen healthy right-handed subjects participated, while remaining naïve to the experimental goals. At the completion of the MRI session, subjects were informed about the goals of the experiment and completed a binary questionnaire describing their susceptibility to contagious yawning, both in general and during the experiment.

Our fMRI results provided both expected and novel results. In addition to standard activations in visual (all stimuli) and visual motion areas (cough, gape, yawn stimuli), we found common activations in right inferior frontal cortex and right STS during the cough and gape stimuli. The gape condition showed a number of additional activations, including the bilateral ventral premotor areas, the parahippocampal gyrus near the ventral posterior cingulate (BA 23) and pre-supplementary motor area. For our contrast of interest (yawn-gape), we found a significant difference only in the region of the ventromedial prefrontal cortex (vmPFC; fig. 2).

The findings of activations in inferior frontal cortex, STS, ventral premotor and the posterior cingulate validated prior study results while demonstrating their association with our non-contagious control stimuli. Despite the activation of human mirror neuron areas, such as the inferior frontal cortex and STS, we found no significant MNS activations in our yawn-gape contrast since these regions were involved in both conditions and unlikely to be the primary regions mediating contagious motor programs such as yawning. The novel finding of vmPFC activation associated with viewing contagious yawning does however seem a reasonable candidate region. The vmPFC has been associated with emotional processing of internal and external stimuli and representation of emotional responses [17]. The identification of a cortical correlate of contagious yawning may seem unlikely, considering its primitive origins and its association with previously reported lower brain regions such as the hypothalamus and brainstem. A cortical ‘ releasing’ mechanism, however, helps to explain why yawning is ubiquitous while contagious yawning is primarily restricted to humans and non-human primates. Furthermore, the role of the vmPFC in contagious yawning may explain the findings of Anderson and Meno [18] showing that contagious yawning did not develop in children below the age of 5 years. We hypothesize that the processing of contagious motor programs only begin after the child has reached some developmental or biological milestone (e.g. axonal myelination, synaptic plasticity or empathic processing), and that this process may be heralded by the disappearance of other primitive reflexes in the toddler.

The neural structures involved in the generation of the yawn appear to span multiple levels of the neural axis, including the brainstem, subcortical regions and prefrontal cortex. The roles and interplay of these regions in the motor production of the yawn, as well as the processing of contagious yawning, remain poorly understood. Further work is needed to integrate our knowledge of contagious motor programs such as yawning into a model explaining the functional roles of each brain region and its contribution to the larger neural network.