RMSERoot-mean-square error AIM Motor learning is enhanced with practice and feedback. This cohort control study investigated the effect of different relative feedback frequencies during skill acquisition in children with cerebral palsy (CP) and children with typical development.METHOD Nineteen children with spastic hemiplegic CP (nine males, 10 females; mean age 11y 7mo; range 8-16y) and 20 children with typical development (12 males, eight females; mean age 10y 8mo; range 8-14y) were assigned to 100% or reduced (62%) feedback subgroups as they practised 200 trials of a discrete arm movement with specific spatiotemporal parameters. Children with CP used their less involved hand. Learning was inferred by delayed (24h) retention and reacquisition tests.RESULTS All children improved in accuracy and consistency. Children with typical development demonstrated significantly greater accuracy than children with CP during acquisition (p=0.001), retention (p=0.031), and reacquisition (p=0.001), and greater consistency during retention (p=0.038). The typically developing group who received 100% feedback performed with significantly less error than the 62% feedback group during acquisition (p=0.001), and with greater retention (p=0.017). No statistically significant difference was found between feedback subgroups of children with CP, although the 100% feedback group consistently demonstrated less error.