Search

The cortical motor areas are activated not only during contralateral limb movements but also during ipsilateral limb movements. Although these ipsilateral activities have been observed in several brain imaging studies, their functional role is poorly understood. Due to its high temporal resolution and low susceptibility to artifacts from body movements, the electrocorticogram (ECoG) is an advantageous measurement method for assessing the human brain function of motor behaviors. Here, we demonstrate that contra- and ipsilateral movements share a similarity in the high-frequency band of human ECoG signals. The ECoG signals were measured from the unilateral sensorimotor cortex while patients conducted self-paced movements of different body parts, contra- or ipsilateral to the measurement side. The movement categories (wrist, shoulder, or ankle) of ipsilateral movements were decoded as accurately as those of contralateral movements from spatial patterns of the high-frequency band of the precentral motor area (the primary motor and premotor areas). The decoder, trained in the high-frequency band of ipsilateral movements generalized to contralateral movements, and vice versa, confirmed that the activity patterns related to ipsilateral limb movements were similar to contralateral ones in the precentral motor area. Our results suggest that the high-frequency band activity patterns of ipsilateral and contralateral movements might be functionally coupled to control limbs, even during unilateral movements.

Tongue movements contribute to oral functions including swallowing, vocalizing, and breathing. Fine tongue movements are regulated through efferent and afferent connections between the cortex and tongue. It has been demonstrated that cortico-muscular coherence (CMC) is reflected at two frequency bands during isometric tongue protrusions: the beta (β) band at 15-35Hz and the low-frequency band at 2-10Hz. The CMC at the β band (β-CMC) reflects motor commands from the primary motor cortex (M1) to the tongue muscles through hypoglossal motoneuron pools. However, the generator mechanism of the CMC at the low-frequency band (low-CMC) remains unknown. Here, we evaluated the mechanism of low-CMC during isometric tongue protrusion using magnetoencephalography (MEG). Somatosensory evoked fields (SEFs) were also recorded following electrical tongue stimulation. Significant low-CMC and β-CMC were observed over both hemispheres for each side of the tongue. Time-domain analysis showed that the MEG signal followed the electromyography signal for low-CMC, which was contrary to the finding that the MEG signal preceded the electromyography signal for β-CMC. The mean conduction time from the tongue to the cortex was not significantly different between the low-CMC (mean, 80.9ms) and SEFs (mean, 71.1ms). The cortical sources of low-CMC were located significantly posterior (mean, 10.1mm) to the sources of β-CMC in M1, but were in the same area as tongue SEFs in the primary somatosensory cortex (S1). These results reveal that the low-CMC may be driven by proprioceptive afferents from the tongue muscles to S1, and that the oscillatory interaction was derived from each side of the tongue to both hemispheres. Oscillatory proprioceptive feedback from the tongue muscles may aid in the coordination of sophisticated tongue movements in humans.

Sophisticated tongue movements contribute to speech and mastication. These movements are regulated by communication between the bilateral cortex and each tongue side. The functional connection between the cortex and tongue was investigated using oscillatory interactions between whole-head magnetoencephalographic (MEG) signals and electromyographic (EMG) signals from both tongue sides during human tongue protrusion compared to thumb data. MEG-EMG coherence was observed at 14-36 Hz and 2-10 Hz over both hemispheres for each tongue side. EMG-EMG coherence between tongue sides was also detected at the same frequency bands. Thumb coherence was detected at 15-33 Hz over the contralateral hemisphere. Tongue coherence at 14-36 Hz was larger over the contralateral vs. ipsilateral hemisphere for both tongue sides. Tongue cortical sources were located in the lower part of the central sulcus and were anterior and inferior to the thumb areas, agreeing with the classical homunculus. Cross-correlogram analysis showed the MEG signal preceded the EMG signal. The cortex-tongue time lag was shorter than the cortex-thumb time lag. The cortex-muscle time lag decreased systematically with distance. These results suggest that during tongue protrusions, descending motor commands are modulated by bilateral cortical oscillations, and each tongue side is dominated by the contralateral hemisphere.

Contribution of the subcortical nuclei to the coordination of human behavior is dependent on the existence of appropriate anatomical architecture. Interpretations of available data have led to opposing ‘information funneling’ and ‘parallel processing’ hypotheses. Using motor circuit as a model, we examined whether cortico-subcortical circuits, especially cortico-basal ganglia circuits, are funneled or parallel in the control of volitional movement. Twenty-five healthy subjects underwent functional magnetic resonance imaging (fMRI). Activated clusters during self-initiated, sequential finger-to-thumb opposition movements of the left hand were identified in the bilateral supplementary motor area (SMA), right lateral premotor cortex (PM) and primary motor cortex (M1), and in the right striatum and thalamus. These functionally defined clusters were applied to probabilistic tractography based on diffusion-weighted MRI to examine patterns of connectivity. Striatal and thalamic sub-regions with high probabilities of connection to the motor cortices partially overlapped, with connection to the two premotor areas outspreading rostrally relative to M1. We suggest that, on a macroscopic anatomical level, there is overlap as well as segregation among connections of the motor cortices with the striatum and thalamus. This supports the notion that neuronal information of the motor cortices is funneled, and parallel processing is not an exclusive principle in the basal ganglia.

Mirror therapy is one of the promising rehabilitation therapeutic interventions but the neural basis of the therapeutic effect remains unknown. It has been reported that the 20-Hz rhythmic activity is induced in the primary motor cortex after median nerve stimulation and the amount of the induced activity is decreased when the primary motor cortex is activated. In the present study, to elucidate the neural mechanisms underlying mirror therapy, we investigated whether the mirror reflection of a hand holding a pencil modulates the stimulus-induced 20-Hz activity. Neuromagnetic brain activities were recorded from 11 healthy right-handed subjects while they were viewing their hand holding a pencil or its mirror reflection. The right median nerve was stimulated and the stimulus-induced 20-Hz activity over the left rolandic cortex dominantly innervating right-hand movements was quantified. The stimulus-induced 20-Hz activity was strongly suppressed when subjects viewed the right hand holding a pencil or the mirror reflection of the left hand looking like the right hand holding a pencil, compared with when subjects viewed the left hand holding a pencil or the mirror reflection of the right hand looking like the left hand holding a pencil. These results suggest that the human left primary motor cortex is strongly activated when the subjects view not only the right hand holding a pencil but also the mirror reflection of the left hand looking like the right hand holding a pencil. This may be one of the neural mechanisms responsible for the therapeutic effect of mirror therapy.