I. INTRODUCTION
The problem of motor redundancy is well known as Bernstein's problem, named after the scientist who first posed the knotty problem in the early 1930s [1], [2]. At present, it is still a mystery how the central nervous system (CNS) solves the ill-posed problem of motor control. Unlike the usual approach to the inverse problem discussed in robotics [3], physiological studies report unique knowledge for solving the problem. The influential hypotheses are the muscle-synergy hypothesis [1], [4], [5], the population-vector hypothesis [6]–[8], and the convergent-force-fields (CFFs) hypothesis [9]–[11]. These hypotheses are similar in that they combine the motor units of CNS but their assumed primitives are different. The relationships among the hypotheses have remained an open question. This paper then discusses these relationships through our original statistical analysis of lower limb movement during human walking based on the EMG ratio of agonist-antagonist muscle pairs (A-A ratio).