Over the last two decades, the number of studies on motor imagery in children has witnessed a large expansion. Most studies used the hand laterality judgment paradigm or the mental chronometry paradigm to examine motor imagery ability. The main objective of the current review is to collate these studies to provide a more comprehensive insight in children's motor imagery development and its age of onset. Motor imagery is a form of motor cognition and aligns with forward (or predictive) models of motor control. Studying age-related differences in motor imagery ability in children therefore provides insight in underlying processes of motor development during childhood. Another motivation for studying age-related differences in motor imagery is that in order to effectively apply motor imagery training in children (with motor impairments), it is pertinent to first establish the age at which children are actually able to perform motor imagery. Overall, performance in the imagery tasks develops between 5 and 12 years of age. The age of motor imagery onset, however, remains equivocal, as some studies indicate that children of 5 to 7 years old can already enlist motor imagery in an implicit motor imagery task, whereas other studies using explicit instructions revealed that children do not use motor imagery before the age of 10. From the findings of the current study, we can conclude that motor imagery training is potentially a feasible method for pediatric rehabilitation in children from 5 years on. We suggest that younger children are most likely to benefit from motor imagery training that is presented in an implicit way. Action observation training might be a beneficial adjunct to implicit motor imagery training. From 10 years of age, more explicit forms of motor imagery training can be effectively used.