OLFACTORY MODELS

The role of olfaction in navigation was an essential component of the argument for parallel maps in the hippocampus. The olfactory spatial hypothesis took this one step farther, proposing that the main olfactory system also maps space using a parallel map architecture and is an essential part of the hippocampal navigation system in most vertebrates. The navigational nose hypothesis offers the first olfactory function hypothesis for the human external nose. The PROUST hypothesis introduces the concept of embodied olfactory cognition.

To a navigator, odors, like visual landmarks, can be classified by their utility, i.e., their affordance to the navigator.

Odor plumes can be classified as directional cues, whose function is primarily to provide direction and not position (or location). In contrast, a discrete visual object, which defines a point in space with precision, can be classified as a positional cue. In the parallel map theory, Françoise Schenk and I proposed that these two classes of cues can be employed independently or integrated, by different locations (subfields) in the hippocampus. Thus we proposed that odor plumes are encoded differently from other landmarks.

We proposed that spatial navigation in mammals was dissociable into two parallel processes, subserved by different hippocampal subfields. The ancestral process (the bearing map) encoded space using distributed stimuli, such as odor gradients, which we termed directional cues. We proposed that the more derived hippocampal function was the encoding of space in sketch maps, using high spatial resolution input from the perception of discrete objects, which we termed positional cues. Corralling data from diverse disciplines, we proposed that the bearing map creates a scaffold upon which the spatial relationships among positional cues can be associated. From this, it is possible to infer possible novel shortcuts among known positional cues, by extrapolating gradients among them, a function we termed the integrated map.