Unloaded Minds - How Prolonged Unloading Shapes Human Performance in Lunar Gravity

The study combines two spaceflight analogues in sequence: a 14-day Unilateral Lower Limb Suspension (ULLS) protocol followed by a parabolic flight campaign simulating lunar gravity (0.16G). This design allows investigation of both the neuromuscular consequences of unloading and the acute reactivation of the system upon first weight-bearing locomotion under partial gravity.

During ULLS, participants keep one lower limb (randomized in a counterbalanced manner) unloaded continuously using a shoe with a 10-cm thick sole and short-length crutches with forearm support. The suspended limb remains in a near-straight position and swings freely from the hip during all upright and ambulatory activity. Compliance is verified daily via telephone interviews, continuous midcalf skin temperature monitoring, and accelerometer data. Fragmin 5000 IU is administered subcutaneously once daily throughout the ULLS period as thromboprophylaxis.

Following the 14-day ULLS period, participants travel to Bordeaux where they continue the ULLS protocol until initiating the first steps on a treadmill during the lunar gravity phases of a parabolic flight. Each participant is allocated 15 parabolas (~20 seconds of 0.16G each): parabolas 1-8 at 3 km/h and parabolas 9-15 at 5 km/h.

Gait analysis is performed on a treadmill at 3 and 5 km/h at baseline (pre-ULLS) and during the parabolic flight. Kinematic data (knee and ankle joint angles) are recorded bilaterally using 2D electronic goniometers (Biometrics, 500 Hz) in the frontal and sagittal planes. Ground reaction forces and spatiotemporal parameters (step and stride length, cadence, step width, gait cycle duration and variability) are recorded using plantar pressure insoles (200 Hz). All signals are synchronized to allow left/right (loaded/unloaded) differentiation.

EEG is recorded during all gait sessions using a 64-channel cap (EasyCap GmbH) with electrodes placed per the international 10-20 system (reference: FCz; ground: AFz). Data are processed in BrainVision Analyzer 2.2 using visual inspection, automated artefact rejection (gradient threshold >60 µV), ICA, and bandpass filtering (0.5-40 Hz). Cortical current density in the motor cortex (M1, Brodmann area 4) is computed using the sLORETA module.

Surface EMG is recorded continuously during gait from eight bilateral lower limb muscles (rectus femoris, lateral hamstrings, gastrocnemius lateralis, tibialis anterior) using a wireless system (Delsys, USA; 2000 Hz), with electrode placement per SENIAM guidelines. Data are bandpass filtered (12-500 Hz), full-wave rectified, and lowpass filtered (7 Hz) to derive linear activation envelopes.

Motor Unit Number Estimation (MUNE) is performed using the MScanFit method, which applies incremental electrical stimuli to the peripheral nerve at 0.6-second intervals across varying intensities, recording the full stimulus-response curve via surface EMG to estimate the number of functional motor units innervating lower leg muscles.

MRI is performed before and after ULLS using a 3-Tesla system. T2-weighted images are acquired to quantify thigh and calf muscle volumes bilaterally. Participants rest supine for ~30 minutes prior to each scan to minimize fluid shift artefacts.

Neuromuscular performance and balance are assessed before ULLS and immediately after the parabolic flight using dual force plates (MuscleLab, 200 Hz). Measures include countermovement jump height, peak power, rate of force development, and centre-of-pressure-based postural stability variables.

Statistical Analysis Data will be assessed for normality using the Shapiro-Wilk test. Continuous outcome variables will be analysed using linear mixed-effects models (LMMs) with Time (pre, post), Leg (unloaded, control), and their interaction (Time × Leg) as fixed effects, and Subject as a random intercept. This approach appropriately handles the repeated-measures, within-subject structure of the data and accommodates missing observations without listwise deletion. Degrees of freedom will be estimated using the Satterthwaite approximation. Where a significant interaction is detected, pairwise post-hoc comparisons will be performed with Bonferroni correction. Effect sizes will be reported as Cohen's d.

For the parabolic flight gait analysis, outcomes recorded at 1G (pre-ULLS) and 0.16G (during flight) will be compared using LMMs with Gravity Condition (1G, 0.16G), Leg (unloaded, control), and Speed (3 km/h, 5 km/h) as fixed effects, and Subject as a random intercept.

Associations between central (EEG-derived cortical current density, MUNE) and peripheral (muscle volume, jump performance, EMG) outcomes will be examined using Pearson or Spearman correlation coefficients, depending on data distribution.

A sample size of 12 participants provides 95% power to detect a moderate effect size (f = 0.6) at a significance level of α = 0.05 for a within-subject design comparing outcomes across gravity conditions (G*Power; λ = 17.3, critical F = 4.85). Twenty-four participants will be recruited to account for potential dropouts (including motion sickness during the parabolic flight, which may be a common, impossible-to-predict cause of dropout) and ensure the required sample is available for analysis. Statistical analyses will be performed in R (R Foundation for Statistical Computing). The significance threshold is set at p < 0.05.