Purpose Understanding how the neuromuscular system adapts to increasing force demands is essential for characterizing compensatory motor control. This study investigated force-dependent reconfiguration of muscle and cortical functional networks during isometric upper-limb tasks. Methods Twelve healthy participants performed isometric elbow flexion at 30%, 50%, and 70% of maximal voluntary contraction (MVC). Surface electromyography (sEMG) from eight upper-limb muscles and electroencephalography (EEG) from 21 scalp electrodes were recorded concurrently. Directed functional connectivity was estimated using generalized partial directed coherence (GPDC), and graph-theoretical metrics—average global efficiency (AGE), average clustering coefficient (ACC), and average path length (APL)—were computed separately for muscle and cortical networks. Results In the muscle network, a significant main effect of force level was observed. Compared with 30% MVC, AGE increased by 12.24% (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">$P = 0.043$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mo>=</mo> <mn>0.043</mn> </mrow> </math></EquationSource> </InlineEquation>) and APL decreased by 17.14% (<InlineEquation ID="IEq2"> <EquationSource Format="TEX">$P = 0.031$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mo>=</mo> <mn>0.031</mn> </mrow> </math></EquationSource> </InlineEquation>) at 70% MVC, while ACC increased by 44.64% (<InlineEquation ID="IEq3"> <EquationSource Format="TEX">$P = 0.018$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mo>=</mo> <mn>0.018</mn> </mrow> </math></EquationSource> </InlineEquation>). In the EEG beta band, AGE increased by 8.12% (<InlineEquation ID="IEq4"> <EquationSource Format="TEX">$P = 0.048$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mo>=</mo> <mn>0.048</mn> </mrow> </math></EquationSource> </InlineEquation>) and APL decreased by 12.34% (<InlineEquation ID="IEq5"> <EquationSource Format="TEX">$P = 0.036$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>P</mi> <mo>=</mo> <mn>0.036</mn> </mrow> </math></EquationSource> </InlineEquation>) at 70% MVC relative to 30% MVC. Gamma band changes were limited or non-significant across conditions. Conclusion These results demonstrate systematic, force-dependent reconfiguration of both muscle and cortical functional networks during isometric force production. Rather than indicating improved performance or neural plasticity, the observed network changes suggest shifts in coordination strategies as force demands increase. The present framework provides quantitative network metrics that can be extended to clinical and longitudinal studies in future work.

Network Reconfiguration Underlies Compensatory Muscle Control Across Force Gradients: Parallel Functional Network Evidence from EEG and sEMG

期刊论文
发表日期：

2026年2月28日

查看全文

Xiaoguang Liu,
Pengyuan Lin,
Yutong Wang,
Tie Liang,
Xiaodong Wang,
Jun Li,
Peng Xiong,
Xiuling Liu

摘要

Purpose

Understanding how the neuromuscular system adapts to increasing force demands is essential for characterizing compensatory motor control. This study investigated force-dependent reconfiguration of muscle and cortical functional networks during isometric upper-limb tasks.

Methods

Twelve healthy participants performed isometric elbow flexion at 30%, 50%, and 70% of maximal voluntary contraction (MVC). Surface electromyography (sEMG) from eight upper-limb muscles and electroencephalography (EEG) from 21 scalp electrodes were recorded concurrently. Directed functional connectivity was estimated using generalized partial directed coherence (GPDC), and graph-theoretical metrics—average global efficiency (AGE), average clustering coefficient (ACC), and average path length (APL)—were computed separately for muscle and cortical networks.

Results

In the muscle network, a significant main effect of force level was observed. Compared with 30% MVC, AGE increased by 12.24% ( $P = 0.043$ $P = 0.043$ ) and APL decreased by 17.14% ( $P = 0.031$ $P = 0.031$ ) at 70% MVC, while ACC increased by 44.64% ( $P = 0.018$ $P = 0.018$ ). In the EEG beta band, AGE increased by 8.12% ( $P = 0.048$ $P = 0.048$ ) and APL decreased by 12.34% ( $P = 0.036$ $P = 0.036$ ) at 70% MVC relative to 30% MVC. Gamma band changes were limited or non-significant across conditions.

Conclusion

These results demonstrate systematic, force-dependent reconfiguration of both muscle and cortical functional networks during isometric force production. Rather than indicating improved performance or neural plasticity, the observed network changes suggest shifts in coordination strategies as force demands increase. The present framework provides quantitative network metrics that can be extended to clinical and longitudinal studies in future work.