Learning to recalibrate the role of dead reckoning and visual cues in spatial navigation

Abstract

In Experiment 1, hamsters started from their permanent home at the periphery of a circular arena and headed to a food source at the center. They then returned, fully laden with food, along a direct path to their home. On control trials, in which no manipulation takes place, visual cues outside the arena and dead reckoning (i.e., updated internal references generated during the outward journey to the food source) controlled the return journey. On experimental trials, the arena, with the hamster in its nest, was rotated by 90°, putting dead reckoning at variance with the distal visual environment. The animals were rewarded for going with dead reckoning. At first, they favored the distal cues, but later most of the subjects switched to using dead reckoning. Thus, hamsters are flexible enough to recalibrate the relative weight that they normally attribute to different sets of spatial cues. In Experiment 2, the reliance on dead-reckoning was greatly enhanced when a cue card at the nest entrance was rotated along with the arena, pitting one proximal cue plus dead reckoning against distal cues. Hence, dead reckoning and external cues seem to reinforce each other through their mutual correlation.