Developmental Changes in the Relative Weighting of Geometric and Experience-Dependent Location Cues

sponses are biased toward spatial prototypes, and these biases increase under condi-tions of uncertainty. Consistent with the model, 6- and 11-year-olds ’ biases toward prototypes increased across delays, especially for locations far from prototypes. Re-sponse biases also varied systematically with target frequency; however, responses were not always biased toward prototypes. In Experiment 1, 6-year-olds ’ responses to an infrequent target near the category boundary were biased toward a frequent tar-get in an adjacent category. In Experiment 2, biases toward a frequent target were evi-dent near prototypes. Both categorical information and children’s experience with lo-cations influence location estimates. Moreover, children’s selective weighting of competing location cues changes between 6 and 11 years. Spatial cognition has played a central role in developmental psychology since Piaget’s early explorations of children’s understanding of space (Piaget & Inhelder, 1956; Piaget, Inhelder, & Szeminska, 1960). Despite this long history of research, the field lacks a unifying theory of the development of spatial skills. Re-cently, Newcombe, Huttenlocher, and colleagues (Newcombe & Huttenlocher, 2000; Newcombe, Huttenlocher, Drummey, & Wiley, 1998) proposed a frame-work that may unify decades of spatial memory research. According to their ac-count, children encode locations using four coding systems: response learning, dead reckoning, cue learning, and place learning. The response-learning and dead-reckoning systems encode locations in relation to the self. In contrast, the cue-learning and place-learning systems encode locations relative to landmarks in