Generating variability from motor primitives during infant locomotor development

Area of Science:

• Developmental neuroscience
• Motor control
• Computational neuroscience

Background:

• Motor variability is crucial for motor learning and exploration in developing systems.
• Human infant leg movements are based on fundamental coordination patterns known as locomotor primitives.
• The emergence and developmental trajectory of motor variability from these primitives are not well understood.

Purpose of the Study:

• To investigate how motor variability changes during early human development, from birth to walking onset.
• To determine the relationship between motor primitives and trial-to-trial variability in infant leg movements.
• To understand the developmental mechanisms underlying motor exploration and learning.

Main Methods:

• Longitudinal study of 18 infants from birth (~4 days) to walking onset (~14 months).
• Recorded leg muscle activity during locomotor or rhythmic movements across 2-3 time points.
• Applied unsupervised machine learning to analyze the structure of trial-to-trial motor variability.

Main Results:

• The structure of motor variability significantly changes during early development.
• Neonatal infants exhibit maximal motor variability by variably activating a minimal set of motor primitives.
• Toddlers show reduced variability with more consistent activation of an expanded set of primitives.

Conclusions:

• Human neonates utilize variable activation of basic locomotor primitives for early motor exploration.
• Motor development involves a transition from high variability in early activation to more regular, fractionated primitive activation.
• This developmental shift underlies the refinement of motor control leading to independent walking.