Swallow is a fundamental behavior to maintain life in animals. Although it is partly mediated by reflexive neuronal activities at the brainstem level, multiple areas in the cerebral cortices, such as primary motor and somatosensory cortices and supplemental and premotor cortices, are involved in swallow (Jean, 2001; Ertekin and Aydogdu, 2003; Michou and Hamdy, 2009). Recently, event-related desynchronizations (ERDs) involved in the process of swallow have been reported in bilateral sensorimotor areas using magnetoencephalography (MEG) (Dziewas et al., 2003; Suntrup et al., 2013, 2014, 2015). Generally, ERDs are a decrease of the power in the alpha and beta frequency bands during voluntary movements when compared with that during the rest (nonmovement) (Pfurtscheller and Lopes da Silva, 1999; Li et al., 2018; Chen et al., 2020; Spadone et al., 2020; Xie et al., 2021). They are supposed to reflect cortical activities during voluntary movements (Pfurtscheller and Lopes da Silva, 1999). In the previous works using MEG by the same research group, ERDs were observed in the alpha and beta frequency bands during volitional swallows in the bilateral sensorimotor, premotor, and prefrontal areas (Dziewas et al., 2003; Suntrup et al., 2013, 2014, 2015; Suntrup-Krueger et al., 2018). The emergence of the ERDs detected by their methods was changed in patients with Parkinson’s disease without dysphagia (Suntrup et al., 2013), in patients with functional dysphagia without organic abnormality (Suntrup et al., 2014), in patients with poststroke dysphagia after transcranial direct current stimulation therapy (Suntrup-Krueger et al., 2018), mildly and severely in patients with dysphagic amyotrophic lateral sclerosis, and in healthy subjects after electrical stimulation on the pharynx (Suntrup et al., 2015). There have been few analogous works in electroencephalography (EEG) (Cuellar et al., 2016), whereas EEG is an easier and more economical method for evaluating cortical activities, compared with MEG. In addition to ERDs during voluntary movements, corticomuscular coherence (CMC) has been reported in EEG works using motor tasks of the upper and lower extremities (Mima and Hallett, 1999). CMC is supposed to indicate both corticomotoneuronal activities that projecting the anterior horn cell in the spinal cord (Mima and Hallett, 1999) and ascending sensory feedback from muscles to motor cortex (Witham et al., 2011; Liu et al., 2019). Transcranial magnetic stimulation (TMS) works showed changes in corticomotoneuronal activities by the evaluation of motor-evoked potentials in the pharyngeal muscles in the resting state after repetitive TMS, which increased the cortical excitability of the stimulated area (Michou et al., 2014). In the previous works using MEG, significant CMCs in theta, alpha, and beta frequency bands have been found during tongue protrusion task, which suggests that cortical motor commands can be evaluated in oral apparatus by CMC (Maezawa et al., 2014, 2016; Maezawa, 2017). Although it is important to evaluate the corticomotoneuronal activities during swallow movements with activated submental group muscles and pharyngeal muscles (Ding et al., 2002), there has been no analogous work on them. If ERD and CMC can be detected by EEG recordings using a similar way with the previous MEG studies, it would be easily applicable in clinical practice for the evaluation of swallow dysfunction in neurological patients. Therefore, we investigated them based on the hypothesis that ERDs and CMCs could be measured in the bilateral sensorimotor, premotor, and prefrontal areas as reported by the previous MEG works (Dziewas et al., 2003; Suntrup et al., 2013, 2014, 2015; Maezawa et al., 2014, 2016; Maezawa, 2017). The submental group muscle activity was used to differentiate the swallow activation from the baseline according to the previous MEG works (Dziewas et al., 2003; Suntrup et al., 2013, 2014, 2015) because the onset of the activity was close to the start of the pharyngeal phase (Ding et al., 2002). This work is the observational and descriptive one to investigate the EEG change such as ERD and CMC associated with swallowing.

The numbers of volitional swallows were 206 ± 14.4 times during EEG recordings. The participants’ EMG swallowing parameters were as follows: swallow duration, 1.95 ± 0.69 s and EMG peak-to-peak amplitude, 72.8 ± 36.6 μV. Head movement during EEG recordings was visually inspected. We found the ERDs by EEG recordings similar to the previous works using MEG recordings and CMC during swallow movements.

We found the ERDs during the volitional swallow by EEG recordings in the bilateral sensorimotor, premotor, and prefrontal areas in a consistent way as previously reported in MEG works. Furthermore, we found the significant CMC in those areas during the volitional swallow movement.

Event-related desynchronizations in the beta frequency band were most evident in C3 and C4, which corresponds to the middle precentral gyrus in this work. The somatotopy for pharyngeal muscles is arranged in a more medial part than that for tongue muscles (Hamdy et al., 1996; Dziewas et al., 2003). The previous MEG work reported the ERDs during volitional swallow in more medial parts, compared with the ERDs during tongue movement (Dziewas et al., 2003), whereas the previous TMS work showed that pharyngeal muscles were arranged more medial and oral muscles were more lateral in the cortical representations (Hamdy et al., 1996). Our findings indicate cortical activities related to pharyngeal muscles during swallow.

The other brain imaging works also showed swallow-related cortical areas (Kern et al., 2001; Martin et al., 2001; Kober et al., 2015; Kober and Wood, 2018). The bilateral primary sensorimotor cortices, insula, cingulate, and parietal regions were represented during volitional swallow in the fMRI works (Kern et al., 2001; Martin et al., 2001). Although the activity of deep brain structure is difficult to measure with the EEG recordings, it may have influenced the observed sensorimotor activities by the EEG recordings. The previous fNIRS works showed the increased blood flow of the bilateral inferior frontal gyri (Kober et al., 2015; Kober and Wood, 2018), which were included in the analyzed areas for the ERDs in this work.

Event-related desynchronizations in the sensorimotor area are associated with the neuronal activities in not only the production of appropriate motor outputs but also in the processing of somatosensory information from the moving muscle (Pfurtscheller and Lopes da Silva, 1999). Processing of oropharyngeal sensation and complex motor execution are essential during swallow to prevent aspiration (Dziewas et al., 2003; Ertekin and Aydogdu, 2003). Therefore, alteration in ERDs can occur according to the dysphagia symptoms and recovery process (Suntrup et al., 2014; Suntrup-Krueger et al., 2018).

In the process of swallow, the tongue movement is important. To transit a bolus of water posteriorly at the oral phase, the apex of the tongue elevates and contacts with the hard palate and the bolus is propelled posteriorly in the oral cavity. At the pharyngeal phase, the tongue is retracted posteriorly to keep the oropharynx closed at the same time with the elevation and closure of the larynx. After the bolus transits pharynx, the tongue is relaxed and returned to the resting position (Logemann, 1988). In the previous work using the ROIs that covering the lateral areas from the C3 and C4, the tongue thrust execution induced the alpha ERD in the left hemisphere whether the tongue thrust left or right, and beta band ERD in the left hemisphere when the tongue thrust right, suggesting the left-sided dominance in the tongue movement (Sakihara and Inagaki, 2015). In our work, beta band ERD during the swallow was induced in the bilateral sensorimotor cortices. As the tongue movement is not lateralized during swallow, the side-dominance may not have been found. In the other previous work, ERD, ERS, or nothing was induced during the press of the tongue on the roof of the mouth in the primary motor cortex (Morash et al., 2008). The ERD may depend on how the tongue moves. In the previous MEG work that using the tongue electrical sensory stimulation, 20-Hz ERD was shown around 400 ms after the stimulation in the bilateral temporoparietal areas. The sensory inputs to the tongue also induced the beta band ERD. In our work, the ERD during swallow was induced in more medial and rostral parts probably because it was mixed with motor efferents and sensory afferents to the tongue and pharynx. Swallow movement itself requires various muscle activities in the tongue and the pharynx. It is difficult to differentiate specific brain activities for a specific muscle because complex activities of various muscles are necessary for swallow movement. The ERDs found in this work are brain activities during swallow including tongue and pharyngeal movements.

Event-related desynchronizations are detected by comparison with brain activity during a specific event with that during baseline (nonevent) period in the event-related design, instead of comparison with a control task. ERDs have been reported in motor and cognitive tasks without control experiments (Pfurtscheller and Lopes da Silva, 1999; Li et al., 2018; Spadone et al., 2020; Xie et al., 2021). According to the previous works, we detected the brain activity during swallow movements by subtracting that during rest. The ERDs in work emerged at the immediately preceding time of the swallow movements and disappeared around the end time of the swallow movements (the end of the submental EMG activities). The ERDs showed prior to movement onset over the contralateral sensorimotor region and ended after the motor execution in the hand and foot movements in the previous report (Pfurtscheller and Lopes da Silva, 1999). The temporal property of the ERDs during the swallow was similar with the previous finding.

We did not investigate relationships between the ERD and swallow movements because they are beyond the scope of this EEG descriptive work. It is the limitation of this work. The next step would be necessary to investigate relationships between the ERD and swallow movements using correlation, decoding, or validation methods.

We have found that the significant CMC existed in all nine channels covering the lateral and medial parts of the sensorimotor area in theta, alpha, and beta frequency bands in accordance with the previous works (Maezawa et al., 2014, 2016; Maezawa, 2017). For the finger muscles, CMC was not consistently observed in the theta and alpha frequency bands (Mima and Hallett, 1999). However, the previous MEG works reported the CMC in the beta band at 15–35 Hz and the low-frequency band at 2–10 Hz and its contralateral side-dominance in both sides of the tongue during the 2-min tongue protrusion mainly contracting genioglossus muscle (Maezawa et al., 2014, 2016). The CMC in the low-frequency band was suggested to be oscillatory proprioceptive feedback from the tongue muscles to the primary sensory cortex (Maezawa et al., 2016). Swallow consists of a complex series of the pharyngeal and tongue movements. The CMC during swallow may have been partly produced by tongue motion because the submental group muscles included the suprahyoid muscles attached to the hyoid although the muscles working during the swallow are different from those during the tongue protrusion.

As for the temporal property of the CMC, the CMC in low-frequency band was dominant in the early swallow stage with low activity of the submental group muscles and the CMC in beta band frequency was shown with the high activity of the submental group muscle in the late part of the swallow. Previous works showed that the CMC in the beta band frequency reflected the cortical motor commands during a steady tonic muscle contraction and that it did not appear during the initial parts of the movement before the steady contraction (Kilner et al., 1999; Mima and Hallett, 1999). It suggests that the CMC in the beta band may indicate the neural state during a few hundred milliseconds after the pharyngeal movements reached the maximum, that is, the peak of laryngeal elevation. Moreover, it may reflect sensory afferents to motor related areas (Baker et al., 2006). Therefore, the CMC in the beta band frequency might reflect both of motor commands and of sensory afferents in the late part of the swallow. The CMC in the low-frequency band was shown in the early part of the swallow. The previous MEG works showed that the CMC in the low-frequency band reflected the sensory feedback to M1 area with a delay of about 80 ms after CMC in the beta frequency band. However, the CMC in the low-frequency band was only shown without the CMC in the beta frequency band in the early part of the swallow. It might suggest that different networks activate in the early part and in the late part of the swallow. Swallow is a complex motion with multiple muscles working at various timings. Therefore, it is difficult to differentiate precisely the timing of cortical commands with that of the muscle contraction unlike previous MEG works. A future work would be necessary to reveal it.

Corticomuscular coherence does not inform us of the directionality of signals between the EEG and the EMG (Liu et al., 2019). Although it is beyond the scope of this EEG descriptive work, the next step would be to investigate the directionality using causality analyses such as Granger causality analysis (Witham et al., 2011).

The relationship between the submental EMG activities and the phases of the swallow movement was reported by using electroglottography (EGG) in the previous work (Nederkoorn et al., 1999; Ding et al., 2002). Although the submental EMG activities indicate the final oral and pharyngeal phases, it is difficult to differentiate the end of the oral phase from the start of the pharyngeal phase only by the EMG activity. In future, the relationship between swallow phases and EEG activity is to be revealed with the concurrent EEG and EGG recordings.

In conclusion, this work reported that EEG recordings with a small number of electrodes can detect ERDs in the bilateral sensorimotor cortices and oscillatory interaction between the cortex and pharyngeal muscles during volitional swallow in humans. The EEG is an easy and economical equipment for the clinical use compared with MEG, which is available in community hospitals. It might be a useful technology for the evaluation of cortical function during swallow in both healthy subjects and patients with dysphagia.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Committee of Medical Ethics of Dokkyo Medical University. The participants provided their written informed consent to participate in this study.

SK designed the work, collected and interpreted the data, and wrote the initial draft of the manuscript. TMima contributed to interpretation of data and assisted in the preparation of the manuscript. All other authors have contributed to data collection and interpretation and critically reviewed the manuscript. All authors approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.