Probing Corticospinal Control During Different Locomotor Tasks Using Detailed Time-Frequency Analysis of Electromyograms

Abstract

Locomotion relies on the fine-tuned coordination of different muscles which are controlled by particular neural circuits. Depending on the attendant conditions, walking patterns must be modified to optimally meet the demands of the task. Assessing neuromuscular control during dynamic conditions is methodologically highly challenging and prone to artifacts. Here we aim at assessing corticospinal involvement during different locomotor tasks using non-invasive surface electromyography. Activity in tibialis anterior (TA) and gastrocnemius medialis (GM) muscles was monitored by electromyograms (EMGs) in 27 healthy volunteers (11 female) during regular walking, walking while engaged in simultaneous cognitive dual tasks, walking with partial visual restriction, and skilled, targeted locomotion. Whereas EMG intensity of the TA and GM was considerably altered while walking with partial visual restriction and during targeted locomotion, dual-task walking induced only minor changes in total EMG intensity compared to regular walking. Targeted walking resulted in enhanced EMG intensity of GM in the frequency range associated with Piper rhythm synchronies. Likewise, targeted walking induced enhanced EMG intensity of TA at the Piper rhythm frequency around heelstrike, but not during the swing phase. Our findings indicate task- and phase-dependent modulations of neuromuscular control in distal leg muscles during various locomotor conditions in healthy subjects. Enhanced EMG intensity in the Piper rhythm frequency during targeted walking points toward enforced corticospinal drive during challenging locomotor tasks. These findings indicate that comprehensive time-frequency EMG analysis is able to gauge cortical involvement during different movement programs in a non-invasive manner and might be used as complementary diagnostic tool to assess baseline integrity of the corticospinal tract and to monitor changes in corticospinal drive as induced by neurorehabilitation interventions or during disease progression.

Keywords: corticospinal; electromyography; humans; locomotion; neuromuscular control; walking.

Figures

Figure 1

Total EMG intensity resolved in time and amplitude during various locomotor tasks. Averaged total EMG intensity and area under the curve (AUC) representing EMG intensity of the TA (A,B) and GM muscle (C,D) during various walking conditions. Statistical analysis of total EMG intensity over time was performed using repeated measures 2-way ANOVA with the independent factors time (within the gait cycle) and locomotor conditions. AUC data were analyzed with 1-way ANOVA repeated measures. Detailed effects of locomotor tasks on total EMG intensity (A,C) and AUC (B,D) were examined by Dunnett's post-hoc correction for multiple comparisons. *P < 0.05; **P < 0.01; ***P < 0.001. AUC, area under the curve; con, congruent; incon, incongruent; EMG, electromyogram; GM, gastrocnemius medialis; TA, tibialis anterior.

Figure 2

Spectral analysis of EMG intensity during different walking conditions. Mean EMG frequency of the TA (A) and GM muscle (C) during various locomotor tasks. Detailed spectral analysis of EMG intensity within different frequency bands of the TA (B) and GM muscle (D). Statistical analysis of mean EMG frequencies during various tasks were performed by repeated measures 1-way ANOVA. Analysis of EMG intensity at specific frequency ranges was performed by repeated measures 2-way ANOVA with the independent factors frequency and locomotor conditions. Detailed effects of locomotor tasks on mean EMG frequency (A,C) and EMG intensity at different frequency bands (B,D) were examined by Dunnett's post-hoc correction for multiple comparisons. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. con, congruent; incon, incongruent; EMG, electromyogram; GM, gastrocnemius medialis; TA, tibialis anterior.

Figure 3

Spectral density of EMG intensity in GM during regular and targeted walking. EMG intensity of the GM muscle at different frequencies during regular (black line) and skilled, targeted walking (purple line). Data represent mean ± SEM of 27 healthy controls. Data were analyzed by repeated measures 2-way ANOVA with the independent factors frequency and locomotor conditions. Changes of EMG intensity at particular frequencies were examined by Dunnett's post-hoc test. **P < 0.01. EMG, electromyogram; GM, gastrocnemius medialis.

Figure 4

Phase-dependent modulation of neuromuscular control during different walking conditions. Averaged total EMG intensity and detailed EMG intensity at specific frequency bands over restricted activity periods of the TA [(A,B): 95–15% of GC], [(C,D): 60–85% of GC], as well as of the GM [(E,F): 20–50% of GC]. Detailed analysis of EMG intensity at specific frequency ranges was performed by repeated measures 2-way ANOVA with the independent factors frequency and locomotor conditions. Task-specific changes in EMG intensity at particular frequency bands were examined by Dunnett's post-hoc test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. con, congruent; GC, gait cycle; incon, incongruent; EMG, electromyogram.

Figure 5

Heatmaps resolving time-amplitude-frequency characteristics of TA and GM EMGs during different walking conditions. Averaged EMG intensity over time and at specific frequencies during different locomotor tasks are displayed for the TA muscle (left column) and GM muscle (right column). EMG intensity is colored relative to the maximal EMG intensity measured during regular walking (maximal EMG intensity during regular walking corresponds to 1.0). con, congruent; incon, incongruent; EMG, electromyogram; GM, gastrocnemius medialis; TA, tibialis anterior.

Figure 6

Differential neuromuscular control during slow and comfortable walking speed. Averaged total EMG intensity over time during slow (1 km/h) and half-maximal walking (v50%) speed for the TA (A) and GM muscle (E). Area under the curve (AUC) representing EMG intensity and mean EMG frequency during slow and half-maximal gait velocity for the TA (B,C) and GM muscle (F,G). EMG intensity at specific frequency bands during slow and faster walking for the TA (D) and GM muscle (H). Statistical analysis of total EMG intensity over time and EMG intensity at specific frequency bands was performed using repeated measures 2-way ANOVA with the independent factors locomotor conditions and time (within gait cycle) or frequency bands, respectively. Changes of EMG intensity at particular time points (A,E) and frequencies (D,H) were examined by Bonferroni's post-hoc test. AUC and mean frequency data during slow and faster walking were analyzed with two-tailed, paired t-test. *P < 0.05; **P < 0.01; ***P < 0.001. AUC, area under the curve; EMG, electromyogram; GM, gastrocnemius medialis; TA, tibialis anterior; v50%, half-maximal walking speed.