One Hidden Object,Two Spatial Codes: Young Children's Use of Relational and Vector Coding

An important characteristic of mature spatial cognition is the ability to encode spatial locations in terms of relations among landmarks as well as in terms of vectors that include distance and direction. In this study, we examined children's use of the relation middle to code the location of a hidden toy, using a procedure adapted from prior work on spatial cognition in gerbils (Collett, Cartwright, & Smith, 1986). Children of 4 and 5 years searched for a hidden toy in a large-scale environment. They were trained to find the toy with either 2 or 1 landmark present. On subsequent trials we altered the number and locations of the landmarks to determine how children represented the location of the toy. With 2 landmarks present during the initial training trial, the children coded both the middle location and the distance and direction from the toy to the landmarks. With 1 landmark present during the training trial, the children coded the location in terms of distance and direction to the single landmark. Our results shed light on seemingly inconsistent prior findings in both human and nonhuman species and indicate that both relational and vector coding are present in young children.