Participants. Thirty 3-year-olds (15 boys; mean age 36 months, range 2–7 to 3–4) with attachment objects were recruited via the local health authority’s birth records. Two additional children failed to complete the experiment and were excluded. The children were all native English speakers with normal vision. Most came from white, middle-class backgrounds. Parents who had indicated that their child had an attachment object (in answer to a routine set of questions asked during initial recruitment) were contacted by a follow-up screening call with a further set of criterion questions to assess suitability. These are detailed in Table 1. Our criterion for inclusion was that children used their object for self-soothing when going to sleep and that they had owned it for at least a third of their life. Of the parents contacted, approximately half had children whose attachment objects were toys, and the remainder typically reported that the attachment object was a blanket. Children were excluded if they had multiple attachment objects, did not use the object for sleeping, or did not get upset when it was mislaid. Twenty-three additional children (aged 2–10 to 3–5) were excluded because they did
not meet all of these criteria.