J2 · University of Sharjah · 2026

How posture changes brain–muscle coupling during balance

Corticomuscular coherence in people with and without forward head posture, as balance gets harder.

Status: Manuscript under review
Role: Data analyst
Signals: EEG · EMG: tibialis anterior and gastrocnemius, both legs
Methods: Coherence · spectral power

Research question

Does spinal alignment change how strongly the brain and muscles couple when balance becomes difficult?

My role

I was the data analyst on the study, working with Dr. Ibrahim M. Moustafa, Dr. Alham Al-Sharman and Dr. Meeyoung Kim.

Acquisition. EEG recorded on Nexus (Mind Media), with synchronised surface EMG from tibialis anterior and gastrocnemius on both legs.
Preprocessing. Cleaning and segmenting both signals by balance condition.
Coherence. Corticomuscular coherence between EEG and EMG, plus EEG spectral power by band.
Statistics. Posture variables against neurophysiological outcomes, condition by condition.

Experiment

Firm · eyes openbaseline

Firm · eyes closedno vision

Unstable · eyes openunstable surface

Unstable · eyes closedhardest (D4)

easierharder →

Main result

Pearson correlation heatmap between postural alignment variables and neurophysiological outcomes during the unstable eyes-closed condition, from the manuscript — **In the hardest condition, alignment tracked brain–muscle coupling.** Vertebral rotation showed the strongest association with Peak CMC (r = 0.59). Correlation, not causation.

Across conditions

Coupling stays flat until the hardest condition, then rises. Mean ± 1 SD for Peak and Average CMC across the four conditions, from the manuscript's Table 2. The spread between people is large: SDs are close to the means, and lower whiskers are cut at zero.

Interpretation

During the most challenging balance condition, corticomuscular coherence points to increased cortical engagement with postural malalignment: the brain appears to take a more hands-on role in balance when posture is off. These are associations from a cross-sectional design, so they describe how the measures move together rather than what causes what.

What I learned

Coherence is simple to compute and easy to over-read. Most of the work is upstream: getting two different signals cleanly aligned, choosing segments that mean the same thing across participants, and resisting the urge to report every band that crosses a threshold. The graded-difficulty design earned its keep, because a difference that grows with difficulty is far more convincing than a single contrast.

Tools & signals

Nexus (Mind Media)
EEG
Surface EMG
Coherence
Python
MATLAB