Background: Synergistic recruitment of muscular activities is a generally accepted mechanism for motor function control, and motor dysfunction, such as cerebral palsy (CP), destroyed the synergistic electromyography activities of muscle group for limb movement. However, very little is known how motor dysfunction of CP affects the organization of the myoelectric frequency components due to the abnormal motor unit recruiting patterns.Objectives: Exploring whether the myoelectric activity can be represented with synergistic recruitment of surface electromyography (sEMG) frequency components; evaluating the effect of CP motor dysfunction on the synergistic recruitment of sEMG oscillations.Methods: Twelve CP infants and 17 typically developed (TD) infants are recruited for self-paced crawling on hands and knees. sEMG signals have been recorded from bilateral biceps brachii (BB) and triceps brachii (TB) muscles. Multi-scale oscillations are extracted via multivariate empirical mode decomposition (MEMD), and non-negative matrix factorization (NMF) method is employed to obtain synergistic pattern of these sEMG oscillations. The coefficient curve of sEMG oscillation synergies are adopted to quantify the time-varying recruitment of BB and TB myoelectric activity during infants crawling.Results: Three patterns of sEMG oscillation synergies with specific frequency ranges are extracted in BB and TB of CP or TD infants. The contribution of low-frequency oscillation synergy of BB in CP group is significantly less than that in TD group (p < 0.05) during forward swing phase for slow contraction; however, this low-frequency oscillation synergy keep higher level during the backward swing phase crawling. For the myoelectric activities of TB, there is not enough high-frequency oscillation recruitment of sEMG for the fast contraction in propulsive phase of CP infants crawling.Conclusion: Our results reveal that, the myoelectric activities of a muscle can be manifested as sEMG oscillation synergies, and motor dysfunction of CP degrade the synergistic recruitment of sEMG oscillations due to the impaired CNS regulation and destroyed MU/muscle fiber. Our preliminary work suggests that time-varying coefficient curve of sEMG oscillation synergies is a potential index to evaluate the abnormal recruitment of electromyography activities affected by CP disorders.