5 Stability of attachment into later childhood
Predictions of subsequent development from Strange Situation attachment classifications are far from perfect, regardless of which attachment type is concerned (Lamb et al., 1985; Thompson, 1998). Rather, the predictive relationship between Strange Situation behaviour in infancy and subsequent child behaviour is found only when there is stability in caregiving arrangements and family circumstances, which maintain stability in patterns of parent–child interaction. This raises the interesting question of whether the prediction over time is attributable to individual differences in the quality of early infant–parent attachments or instead to the continuing quality of child–parent interactions over time? The latter would imply that the quality of early relationships was predictively valuable; not because it caused later differences directly, but because it presaged later differences in the quality of relationships that in turn support continuing differences in the child’s behaviour. Such a pattern of findings would suggest that stability in attachment is a consequence of stability in parent–child interactions and some evidence has been found that supports this (Belsky and Fearon, 2002).