Pretend play has been considered to be effective for the development of children because it is related to their language and social cognitive skills. In particular, pretend play involving role-play requires understanding of theory of mind (ToM) and communication skills. Although previous studies have discussed the advantages of pretend play, the majority has been based on comments from caregivers or questionnaires. Experimental studies remain few to date. In addition, pretend play has been proposed as a possible factor in promoting the emotional understanding of others. However, experimental studies that directly examined ToM are lacking. To address this research gap, the present study aimed to examine the effect of pretend play on children with/without a doll on the understanding of ToM. The present study specifically focused on the function of inference of others' mental state in a pretend play and the relation observed during the play (i.e., dyadic or triadic). The study consisted of two experiments, which were conducted on girls approximately 3 years old. At this age, they are yet to attain ToM. Experiment 1 examined the change in children's understanding of ToM after different types of interventions through pretend play. The results indicated that only pretend play using a specific script that verbalizes mental states with a triadic relationship (i.e., child, mother, and a doll) significantly changed children's performance in a ToM task. Importantly, a group using the same script but with dyadic relationship did not exhibit a significant change, suggesting the crucial role of a triadic relationship with a doll. Experiment 2, which focused on the influence of children's experience of pretend play with a doll, found no difference in ToM performance due to difference in accumulated experiences with doll play. The differences between the two experimental results were discussed with a view of quality of pretend play at home in relation to the verbalization of mental states and relationship in pretend play.

論文