In order to systematically investigate the covariation between the wall movements and postural sway, the data were transformed to the frequency domain with a fast Fourier transform. The power spectrum of these transformed data was calculated to analyze the percent variance attributed to different frequencies. In Figure 3, panels (c) and (d) show the power spectrum for the wall movement and postural sway on a given trial. As can be observed, the peak frequency of the sway movement is 0.6 Hz and accounts for 28% of the variance; likewise the peak frequency of the wall movement is 0.6 Hz and accounts for 70% of the variance.
The results from the power spectrum were used to test whether the driving frequency of the moving walls entrained infants to sway at that same frequency. In this analysis, we compared the magnitude (percent variance) of sway at the driving frequency (Same frequency) against the magnitude of sway measured at that same frequency but from another trial where the walls were moving at the opposite frequency (Different frequency). As such, the Same frequency condition represents a measure of entrainment, whereas the Different frequency condition represents a control for spontaneous swaying at the driving frequency.
[FIGURE 4] As can be seen in this next figure, the magnitude of sway in the Same frequency condition increased as a function of age, whereas the magnitude of sway in the Different frequency condition remained constant and not different from chance. These results suggest that sensory-motor integration in the control of a sitting posture increases during precisely the time that infants are learning to sit without support.
Figure 4: Covariation between wall movement and postural sway.
As can be seen, the results from the preceding and other related linear systems analyses are quite informative. Still, two questions are left unresolved. The first concerns whether the developmental improvement with age reflects a change in the underlying mechanisms or more simply an improvement with practice and experience. The second question concerns how to explain all of the variance not accounted for by the linear analyses. Is this variance simply a measure of the instability of the system, or does it reflect the contribution of other unspecified variables.