Four experiments were conducted to investigate methods of promoting preschool children's consumption of fruits and vegetables, and some entirely novel foods.
In Experiment 1 a between groups design was employed to examine the effect of
in vivo peer behaviour on children's consumption of a novel food. Children exposed to 'positive' peers who consumed a target novel food tended to consume that food, whilst those exposed to 'negative' peers who rejected the food tended to also reject it. The negative effects persisted despite subsequent exposure to 'positive' peers.
In Experiments 2, 3 and 4, multiple baseline designs were employed to examine
the utility of video-peer modelling with rewards interventions in promoting consumption of fruits and vegetables. Target behaviour was measured in two contexts: a snack-time experimental setting and a lunchtime (generalisation) setting.
The intervention in Experiment 2 targeted individual children's consumption of
fruit and vegetables that they had previously refused. Consumption increased during experimental sessions, with some evidence of the changes in behaviour being maintained, but there was little evidence of generalisation of behaviour to the lunchtime setting.
The interventions in Experiments 3 and 4 targeted children's consumption, at
snack time of fruit (Experiment 3) and fruit and vegetables (Experiment 4) within the group setting of a pre-school nursery classroom. (Some procedures were modified between Experiments 3 and 4).
In Experiment 3 increases in consumption of fruit were observed in both the
snack and lunch settings. Some maintenance of behaviour with half the fruits (the sweetest) was noted.
In Experiment 4, increases in consumption of target fruit and then target
vegetables was observed following the intervention on each food category. The
increases were in evidence at both snack and lunchtimes and were maintained in the absence of the intervention and during follow-ups.
The results showed that peer modelling alone can be effective in promoting food
consumption, but is most powerful when combined with contingent rewards and is implemented in a group setting. These findings are discussed with reference to the role of rule-governed behaviour in promoting and maintaining behaviour change.