How do we learn to navigate? Competing or coexisting navigation strategies in human

Kinga Igloi, LPPA Collège de France CNRS

Abstract
Learning to navigate in a complex environment involves encoding and organizing multisensorial information of the environment. Two main strategies have been described: the allocentric (map-based) strategy and the sequential egocentric (route-based) strategy. However, questions on temporal arousal and hierarchy between strategies are still debated. To address this issue we investigated spontaneous choice and switches between these two spatial strategies, both relying on the integration of multisensorial information but known to depend on different memory systems. We developed the virtual version of the Starmaze, adapted from rodent studies. Subjects had to learn to navigate to a hidden location using an allocentric and/or a sequential egocentric strategy, through a probe test simultaneously arising and equally performing strategies were characterized. Our results showed bi-directional strategy switches, suggesting no temporal or performance-related hierarchy between strategies.
Thus, we argue that albeit spontaneously performed distinct navigation strategies, all sensorial information are encoded in parallel during learning, leading to coexisting strategies that are used, separately or conjunctly, according to their accuracy to solve the task. Spontaneous decision of the strategy used does not seem to depend on the practice of the task but rather on the awareness of the environmental multisensorial information.

Not available

Back to Abstract