Changes in attachment during the third year: Consequences and predictions

Crittenden has proposed a system for classifying attachment during the preschool years. Ninety-six boys and girls were tested in the Strange Situation at 18 months and 30 months of age. Eighteen-month-olds were coded using the Ainsworth coding system, and the 30-month-olds were coded using Crittenden's preschool attachment coding system. When children changed classification categories, it was not random movement but was in accordance with Crittenden's predictions of reorganization of the child's capacities. Mothers and children were also observed in the home at 18 and 30 months of age, and children were observed in playgroups from 18 to 30 months of age. The children were rated by their teachers on problem behaviors and peer relations and given achievement tests at age 7 years. Parenting of the children and child behaviors at 30 months differed by attachment classification and predicted both children's performance on achievement tests and teacher reports of children's behavior at 7 years. The findings are discussed in terms of reorganization of working models of attachment on the basis of changing child competencies.