Identified considerable continuity in attachment organization from longitudinal studies of children at age 6 who had been classified as 12 months. Those with originally avoidant classifications tended to divert attention to toys or activities during a reunion with the attachment figure.  They also displayed difficulty in discussing feelings or coping strategies in response to separations. Originally secure children showed “free-ranging access to affect, memory and plans relevant to attachment” (p. 950). They also notes that attachment to the mother at age one is a better predictor of social behavior at age six than is attachment to the father. They also proposed a fourth attachment style.

They suggested that individual differences in attachment style can be seen as “differences in the mental representation of the self in relation to attachment [and] the secure versus various types of insecure attachment organisations can best be understood as terms referring to particular types of internal working models of relationships, models that direct not only feelings and behaviour but also attention, memory and cognition.” (p. 67).