| Abstrakt: |
The aim of this study was to evaluate the characteristics of infants' spontaneous movements using dynamical systems analysis. The participants were 4 healthy full-term newborn infants, I month of age. A tri-axial accelerometer was used to measure limb acceleration in three-dimensional space. Acceleration signals were recorded on the right wrist when the infant was in an active alert state, lying supine. The recording time was 200 seconds, based on recommendations for the characteristics of this accelerometer. Digitized data were transferred to a PC for subsequent processing using analysis software. The time-series data, analyzed by linear and non-linear methods using the software package TISEAN, provided some evidence for a non-linear deterministic structure. Fast Fourier transform (FFT) analysis suggested it to be largely a linear random process. Non-linear analysis (recurrence plot, maximal Lyapunov exponent and surrogate data), which facilitates hypothesis testing of dynamical measures, provided evidence that the infants' spontaneous movements had non-linear chaotic dynamic characteristics with 5 or 6 optimal embedding dimensions in structuring. The findings support the hypothesis that chaotic behavior dominates physiological systems, including neuronal networks. [ABSTRACT FROM AUTHOR] |
