Attachment categories as reflections of multiple dimensions: comment on Fraley and Spieker (2003).

Ainsworth's system of classifying patterns of attachment behavior has served the field well and will not be easily displaced--not because of a commitment by attachment researchers to the taxonomic status of these categories but rather because of difficulties lying in the way of a dimensional approach. Foremost among these is the large number of dimensions used in making classifications and the need to develop reliable scales to tap them. Other problems include the need to capture how behavior changes across age and across the episodes of the Strange Situation. Others will debate R.C. Fraley and S.J. Spieker's (2003) arguments concerning taxonomic status. However, whether taxa or not, Ainsworth's categories at the least have represented well the multidimensional space underlying attachment behavior in conditions of mild stress. Only when an alternative approach is demonstrated to predict (with equal power, in practice) the vast range of outcomes associated with Ainsworth's categories will it be a candidate for substitution.