Cue combination used to update the navigator's self-localization, not the home location.

Abstract

This study investigated when the Bayesian cue combination of piloting and path integration occurs in human homing behaviors. The Bayesian cue combination was hypothesized to occur in estimating the home location or self-localization. In Experiment 1, the participants learned the locations of 5 objects (1 located at the learning position) in the presence of distal landmarks before walking a 2-leg path without viewing the landmarks and objects. At the end of the path, the participants indicated the original locations of the objects in 4 cue conditions: (a) path integration only, (b) landmarks only where the participants were disoriented and the landmarks reappeared, (c) both path integration and the reappearing landmarks, and (d) path integration and conflicting landmarks rotated 45°. The participants' heading, position, and homing estimations were calculated. The ratio of the length of the second leg to that of the first leg was manipulated to be 0.5, 1, or 2. The results showed evidence of the Bayesian cue combination for heading estimates in all leg ratios, and for homing estimates in a small leg ratio (0.5) but not in a large leg ratio (2). The following experiments replicated the results of the Bayesian cue combination for heading but not for homing estimates for the large leg ratio (2) when participants did a typical homing task without learning the locations of objects (Experiment 2) and when proximal landmarks replaced distal landmarks (Experiments 3-4). These findings suggest that the Bayesian cue combination occurs in self-localization prior to homing. (PsycInfo Database Record (c) 2021 APA, all rights reserved).