Test-retest reliability of visual and self-motion cue combination during navigation: Accuracy, variability, and cue weighting.

Multisensory cue combination during navigation is examined for the relative use and integration of different sensory spatial cues, but the reliability of cue combination metrics has not yet been characterized. We quantify the test-retest reliability of three metrics (accuracy, variability, predicted cue-weighting) used in studies of sensory cue combination for spatial updating. Participants completed a triangular homing task in immersive virtual reality comprising three cue conditions - visual cues-only, self-motion cues-only, and visual + self-motion cues - in two sessions separated by an average 11-day interval. We found strong test-retest reliability of homing accuracy with two sensory cues but poor reliability of accuracy when only one sensory cue was available. Homing variability and predicted cue-weighting showed poor test-retest reliability across all conditions.