This study aims to investigate the predictive value of lower limb muscle strength and explosive power on exercise-induced prefrontal hemodynamic responses. Using an interpretable machine learning framework (GCAT-Net), the research analyzes how various physical performance indicators-such as isokinetic muscle strength, 1RM leg press, and vertical jump metrics-can predict oxygenated hemoglobin (ΔHbO) changes in the bilateral dorsolateral prefrontal cortex (DLPFC) during moderate-intensity cycling.

Inclusion Criteria:

1. Healthy adults aged between 18.00 and 60.00 years.
2. Must be able to perform standardized lower-limb physical assessments, including isokinetic strength testing and explosive power tests (e.g., CMJ, SJ, and 30m sprints).
3. Capability to complete a 6-minute aerobic cycling task at a moderate intensity (60% VO_2max).
4. Physically cleared for exercise as determined by a Physical Activity Readiness Questionnaire (PAR-Q).
5. Willingness to undergo functional near-infrared spectroscopy (fNIRS) monitoring with a Scalp Coupling Index (SCI) ≥ 0.7.

Exclusion Criteria:

1. History of musculoskeletal injuries to the lower limbs within the past 6 months that would impede maximum voluntary contraction (MVC) or high-intensity efforts.
2. Known cardiovascular, respiratory, or neurological conditions that contraindicate maximal exercise testing (VO_2max ramp protocol).
3. Body Mass Index (BMI) or health status that prevents the safe completion of high-impact explosive power tests like the 30cm drop jump (RSI assessment).
4. Presence of excessive motion artifacts or poor fNIRS signal quality, specifically a Signal-to-Noise Ratio (SNR) < 5 dB or motion artifact frame ratio ≥$ 10%.