Neural Kinesthetic Contribution to Motor Imagery of Body Parts: Tongue, Hands, and Feet

Abstract

Motor imagery (MI) is assimilated to a perception-action process, which is mentally represented. Although several models suggest that MI, and its equivalent motor execution, engage very similar brain areas, the mechanisms underlying MI and their associated components are still under investigation today. Using 22 Ag/AgCl EEG electrodes, 19 healthy participants (nine males and 10 females) with an average age of 25.8 years old (sd = 3.5 years) were required to imagine moving several parts of their body (i.e., first-person perspective) one by one: left and right hand, tongue, and feet. Network connectivity analysis based on graph theory, together with a correlational analysis, were performed on the data. The findings suggest evidence for motor and somesthetic neural synchronization and underline the role of the parietofrontal network for the tongue imagery task only. At both unilateral and bilateral cortical levels, only the tongue imagery task appears to be associated with motor and somatosensory representations, that is, kinesthetic representations, which might contribute to verbal actions. As such, the present findings suggest the idea that imagined tongue movements, involving segmentary kinesthetic actions, could be the prerequisite of language.

Keywords: body parts; connectivity; kinesthetic representations; motor mental imagery; verbal actions.

Figure 1

Timing presentation of the four motor imagery experimental paradigm. The start of each trial was signified by the presence of a short “beep tone”. This corresponds to t = 0. Immediately, a fixation cross appeared in the middle of the screen for 2 s, i.e., t = 1, fixation cross. The 2 s duration time constitutes the baseline. After these 2 s, an arrow indicating either left, right, up, or down direction appeared for 1.25 s on the screen. The arrow is the visual cue, i.e., t = 2. By definition, each direction of the arrow corresponds to a body part, as follows: left arrow ← left hand (i.e., task 1), right arrow → right hand (i.e., task 2), up arrow ↑ tongue (i.e., task 3), and down arrow ↓ feet (i.e., task 4). Each arrow required the participant to perform the corresponding MI task. Once the arrow disappeared from the screen, each participant, facing the black screen, performed a MI task, one at a time, for 2 s, i.e., t = 3. At the end of each motor imagery task, each participant was allowed to take a break and relax. The inter-trial interval was around 2 s.

Figure 2

Functional connectivity matrices obtained from the comparison of EEG data between baseline and imagery tasks. They were ordered according to the motor imagery task: left hand, right hand, feet, and tongue. Information flows from the vertices marked on the vertical axis to the vertices marked on the horizontal axis: 22 vertices × 22 vertices (i.e., 22 electrodes × 22 electrodes). Each element of the correlation matrix shows z-scores corresponding to each node comparison. Statistically significant results were only obtained for tongue motor imagery (p < 0.05).

Figure 3

Graphical presentation on the relationship between baseline and tongue imagery task (A) 19 vertices significant correlations (B) configuration of z-scores. Brain activity is reduced when participants performed the tongue imaginative task. Note that vertices correlations were statistically significant for the anterior and central brain areas (p < 0.05).

Figure 4

Low-resolution electromagnetic tomography (LORETA) source analysis during (A) baseline (B) tongue imagery. On average, the mirrored intensity of the brain activity was reduced for the tongue imagery in premotor, motor, parietal and visual areas bilaterally (p < 0.01). The orange/yellow scale is a positive one, i.e., the more high, the more intense. Orange/Yellow shades indicate decreased sources. Abbreviations: L: left; R: right; A: anterior; P: posterior; Sag: Sagittal; Cor: Coronal.

Figure 5

Brain maps of coherence connectivity at (A) 0.7 threshold (B) 0.8 threshold and (C) 0.9 threshold in relation to tongue imagery. A coherence method followed by complex demodulation with time and frequency sampling 100 ms and 0.5 Hz respectively was applied in BESA Connectivity 1.0 software (BESA^®). When the 0.7 threshold is considered, the connections across the vertices were significant for the baseline vs. tongue imagery (p < 0.05) only. The vertices represent the significant brain areas reported by the source space analysis. The edges are loaded and colored according to the connectivity strength. They indicate significant higher (intense color) and lower (soft color) areas of synchronization both bilaterally and unilaterally.