Suboptimal Muscle Synergy Activation Patterns Generalize their Motor Function across Postures.

Area of Science:

• Neuroscience
• Biomechanics
• Motor Control

Background:

• Consistent muscle synergy patterns are used to produce functional motor outputs across varying biomechanical conditions, a property termed generalizability.
• Previous studies in cats show similar muscle synergies are employed during reactive postural responses across different limb configurations.
• The underlying basis for this generalizability—whether biomechanical properties or neural selection—remains unclear.

Purpose of the Study:

• To investigate the biomechanical and neural factors contributing to generalizability of muscle synergies.
• To explore feasible muscle activation patterns that produce experimental synergy force vectors and test their generalizability across postures.

Main Methods:

• Utilized a detailed cat hindlimb musculoskeletal model.
• Selected candidate muscle activation patterns using three methods: random feasible, minimum effort, and generalizable patterns.
• Assessed generalizability by measuring force vector deviations across different postures.

Main Results:

• Generalizable muscle activation patterns exhibited the smallest force angle deviations (<5°), outperforming random (>100°) and minimum-effort (~20°) patterns.
• Generalizable patterns were effort-suboptimal, often exceeding 50% maximum effort compared to ~5% for minimum-effort patterns.
• Imposing generalizability reduced feasible muscle activation ranges by ~45% and recruited muscles less sensitive to postural changes.

Conclusions:

• Generalization of motor function across postures is not solely due to limb biomechanics or a single optimality criterion.
• Muscle synergies may represent learned motor solutions optimized for broad applicability across diverse biomechanical contexts.
• This global tuning for generalizability likely facilitates rapid motor adaptation and robust motor control.