Afferents to Action: Cortical Proprioceptive Processing Assessed with Corticokinematic Coherence Specifically Relates to Gross Motor Skills

Abstract

Voluntary motor control is thought to be predicated on the ability to efficiently integrate and process somatosensory afferent information. However, current approaches in the field of motor control have not factored in objective markers of how the brain tracks incoming somatosensory information. Here, we asked whether motor performance relates to such markers obtained with an analysis of the coupling between peripheral kinematics and cortical oscillations during continuous movements, best known as corticokinematic coherence (CKC). Motor performance was evaluated by measuring both gross and fine motor skills using the Box and Blocks Test (BBT) and the Purdue Pegboard Test (PPT), respectively, and with a biomechanics measure of coordination. A total of 61 participants completed the BBT, while equipped with electroencephalography and electromyography, and the PPT. We evaluated CKC, from the signals collected during the BBT, as the coherence between movement rhythmicity and brain activity, and coordination as the cross-correlation between muscle activity. CKC at movements' first harmonic was positively associated with BBT scores (r = 0.41, p = 0.001), and alone showed no relationship with PPT scores (r = 0.07, p = 0.60), but in synergy with BBT scores, participants with lower PPT scores had higher CKC than expected based on their BBT score. Coordination was not associated with motor performance or CKC (p > 0.05). These findings demonstrate that cortical somatosensory processing in the form of strengthened brain-peripheral coupling is specifically associated with better gross motor skills and thus may be considered as a valuable addition to classical tests of proprioception acuity.

Keywords: EEG; gross motor skill; motor control; proprioception; sensorimotor cortex.

Figure 1.

Experimental setup. Participants performed the BBT (left) while equipped with a 64-channel EEG cap and bipolar EMG electrodes on the FDI, biceps, and deltoid. Once a block is selected and picked up, the participant moves their hand over the barrier and releases the block. Participants performed the PPT (right) unequipped. Pegs were transferred from the starting position to the next available small hole.

Figure 2.

Processing of EMG signals. A, Excerpt of pre-processed EMG signals for the FDI, biceps, and deltoid from a representative participant. From each of the rectified EMG signals (dark gray traces) were extracted a fast (blue traces) and a slow (red traces) envelope. B, Muscle recruitment traces for each of the muscles, obtained as the ratio between the fast and slow envelopes. C, Cross-correlations between pairs of muscle recruitment traces. Their maximal amplitude was taken as an estimate of motor coordination. D, Auto correlations for the same muscle recruitment traces. The amplitude of their side peaks was taken as a measure of regularity.

Figure 3.

Relationship between PPT scores (pegs inserted in 30 s) and BBT scores (blocks transferred in 1 min). Circles indicate individual values, and their linear regression line is in red. The correlation value and associated significance level are indicated in the bottom right corner.

Figure 4.

Spectrum and scalp distribution of CKC for a representative individual. The CKC spectrum presents one trace for each EEG electrode, and a horizontal red line indicates the level of statistical significance. Coherence peaked at F0 and F1, corresponding to ∼1.5 and 3 Hz, respectively, for this particular individual. Scalp distributions are mostly compatible with tangential sources in the left SM1 cortex, although other sources might have contributed (Bourguignon et al., 2012).

Figure 5.

Relation of CKC at F1 with BBT scores (A), PPT scores (B), and EEG SNR corrected CKC at F1 with BBT scores (C) and PPT scores (D). Plots are as in Figure 3.

Figure 6.

Fine-grained relationship of CKC with motor skills. A, Information structure of BBT and PPT scores predicting CKC. Ellipses outline the decomposition of mutual information into unique information brought by BBT scores (purple), unique information brought by PPT scores (beige), redundant information present in both scores, and synergistic information not present in either score but emerging from them together. Their significance level is indicated in each partition. B, Relationship between standardized CKC and standardized BBT scores, where individual values are color-coded for their standardized PPT scores. C, Relationship between standardized BBT scores subtracted from standardized CKC as a function of PPT scores. Plots are as in Figure 3.