ABSTRACT:It is often assumed that navigation implies the use, by animals, of landmarks indicating the location of the goal. However, many animals (including humans) are able to return to the starting point of a journey, or to other goal sites, by relying on self-motion cues only. This process is known as path integration, and it allows an agent to calculate a route without making use of landmarks. We review the current literature on path integration and its interaction with external, location-based cues. Special importance is given to the correlation between observable behavior and the activity pattern of particular neural cell populations that implement the internal representation of space. In mammals, the latter may well be the first high-level cognitive representation to be understood at the neural level.

During short foraging excursions away from their home, central place foragers update their position relative to their point of departure by processing signals generated by locomotion. They therefore can home along a self-generated vector without using learned references. In rodents and other mammals, this path integration process (dead reckoning) can occur on the basis of purely internal signals, such as vestibular or proprioceptive (re)afferences. We report here that hamsters are also capable of proceeding to a previously learned feeding site through vector information from locomotion only. The subjects compute the direction and distance to the goal by subtracting their current-position vector from the stored nest-to-goal vector. This computation pertains to locations per se and therefore occurs in absolute space, independently of landmark objects. If available, prominent visual cues merely serve to confirm the path planned through the addition of self-generated vectors, whereas visual as well as nonvisual references confirm that the subject has arrived at the goal site.

No abstract

During short excursions away from home, some mammals are known to update their position with respect to their point of departure through path integration (dead reckoning) by processing internal (idiothetic) signals generated by rotations and translations. Path integration (PI) is a continuously ongoing process in which errors accumulate. To remain functional over longer excursions, PI needs to be reset through position information from stable external references. We tested the homing behaviour of golden hamsters (Mesocricetus auratus W.) during hoarding excursions following a rotation of the arena and nest. In continuous darkness, the hamsters returned to their point of departure at the rotated nest, and therefore depended on PI only. In other trials, the animals were briefly presented with visual room cues during or at the end of the outward trip, visual cues being pitted by 67°or 98°against the animal's current selfgenerated position vector. After a fix, the animals headed for the usual (unrotated) nest location, as defined by room cues, independent of the timing of the fix. These results were obtained in two different geometrical settings and showed that, after the fix, the animals update their position, and not merely their head direction or internal compass, in a new reference frame. Thus, episodic fixes on familiar external references reset the PI and therefore greatly enhance the functional signification of navigation that is based on feedback information from locomotion.

No abstract