The progress drive hypothesis: an interpretation of early imitation

Children seem to have a natural tendency to imitate and their interest for particular kinds of imitative behaviour varies greatly with the infant’s age. We argue that different forms of children’s early imitation may be the result of an intrinsic motivation system driving the infant into situations of maximal learning progress. We present a computational model showing how an agent can learn to focus on “progress niches”, situations neither completely predictable nor too difficult to predict given its anticipation capabilities. The existence of such a drive could explain why certain types of imitative behaviour are produced by children at a certain age, and how discriminations between self, others and objects gradually appear.