It is widely agreed that imitation drives the acquisition of novel behaviours in children. However, there is less agreement about how this behaviour is established. This thesis explores two areas of research in an attempt to identify the determinants of imitation in young infants.
Chapter 2 investigated the effects of local stimulus enhancement and objects' affordances on imitation tests. Infants aged 6-, 9- and 12-months-old were presented with a puppet wearing a mitten (with a bell inside) and observed either a demonstration of mitten removal (modelling group), the experimenter pointing at the mitten (stimulus enhancement group), the mitten falling off, apparently by itself (affordance demonstration group), or no actions directed at the mitten ( control group). Next all infants were presented with the puppet, and their mitten removal behaviours were recorded. The results indicated that local stimulus enhancement and affordance demonstration were as
effective as full modelling in evoking the target action in 6-month-old infants. The results from the 9- and 12-month-old infants did not replicate this finding. A further study was
conducted with the older infants to investigate the effect of locating the bell in the mitten, but the results were inconclusive.
The behaviour analytic account of imitation ( which explains the development of imitation in terms of trained matching and generalised imitation) was evaluated in Chapter 4. The
determinants of generalised imitation were investigated in a multiple-baseline single-participant design. Infants were trained to match eight baseline gestures; next, four novel gestures that infants did not match in probe trials were selected as target behaviours. Imitation was then tested after (i) repeated modelling, (ii) skills-training, (iii) staggered matching training, and (iv) mixed matching training of the target behaviours. Only the last intervention resulted in matching of all targets in subsequent generalised imitation tests, showing that to establish durable new matching relations infants require extensive matching training. The study found no evidence of untrained matching in generalised
imitation tests in infants under 2-years-old.