Sitting is the most functional posture for play and exploration in developing children; usually attained by 6-7 months of age. However, children with cerebral palsy (CP) have significant deficits in sitting attainment, and hence are restricted in interaction. We compared two interventions to promote sitting postural control in children with severe CP, for effects on behavioral and kinetic measures. Gross Motor Function Measure (GMFM), was used to code behavioral changes pre/post intervention; when COP data were also collected. Ten children with moderate or severe CP were randomized into 2 intervention groups. One group received a perceptual motor intervention, the other group received the same intervention plus stochastic noise at the support surface. Repeated measures 2X2 ANOVA (treatment X time)assessed GMFM and COP; p<0.1. There was a significant effect of time for GMFM, but no difference between groups. There was a significant interaction effect for RMS (variability of COP path) in the anterior posterior (AP) direction (p=0.1), and Approximate Entropy in the AP (p=0.08). In both cases, the group with stochastic noise intervention developed in the same direction as expected for typically developing infants in sitting. We conclude that while both interventions produced behavioral changes in sitting skill, the group with stochastic noise added during intervention showed specific improvement in variability and complexity of the COP, which mirrored developmental changes in typical infants’ sitting postural control. Results indicate potential for greater change using the stochastic noise intervention, and thus greater function as sitting skill progresses.