Learning and exposure affect environmental perception less than evolutionary navigation costs.

Most behaviors are conditional upon successful navigation of the environment, which depends upon distance perception learned over repeated trials. Unfortunately, we understand little about how learning affects distance perception-especially in the most common human navigational scenario, that of adult navigation in familiar environments. Further, dominant theories predict mutually exclusive effects of learning on distance perception, especially when the risks or costs of navigation differ. We tested these competing predictions in four experiments in which we also presented evolutionarily relevant navigation costs. Methods included within- and between-subjects comparisons and longitudinal designs in laboratory and real-world settings. Data suggested that adult distance estimation rapidly reflects evolutionarily relevant navigation costs and repeated exposure does little to change this. Human distance perception may have evolved to reflect navigation costs quickly and reliably in order to provide a stable signal to other behaviors and with little regard for objective accuracy.