Despite the lack of age-related variation in maternal scaffolding behaviors, our results showed that during both the natural and experimental play conditions, mothers and children engaged in “a spiralof mutual regulation”. During the ME condition, where mothers were instructed to engage in pretend play with their children, maternal pretense and verbalizations increased and child imitation of maternal pretense increased, in comparison to the MN condition. Our findings also showed that maternal imitation predicted child pretense during both MN and ME, and maternal pretense predicted child imitation during ME. Additionally, maternal verbalizations predicted child pretense in the ME condition. These results reveal that mother and child appear to take turns such that children’s pretense is followed by maternal imitation and verbalization, and maternal pretense is followed by children’s imitation. Moreover, in the ME condition, childand maternal pretense seemed to exclude each other. Previous research showed that children’s pretense behaviors were more likely to follow maternal social referencing behaviors. Although we did not apply sequential analyses to our data, as Nishida and Lillard did, our findings also suggest that pretend play is a sequence where pretense acts of one partner follow and are followed by behaviors that streng then and elaborate on the preceding pretense by the other partner.