Dynamic Aspects of Central Olfactory Processing

Abstract

Perhaps because the sense of smell involves the discrimination of molecules by neural tissue, most neurophysiological studies of vertebrate olfaction have used purified compounds as stimuli and have been directed at relating neural response parameters to the structural and/or physicochemical properties of molecules. In studies of odor responses by individual neurons (single units), a commonly employed paradigm has been one in which individual members from a preselected battery of “odors” (volatile molecules in otherwise highly purified air) are puffed sequentially onto the nasal mucosa and an attempt is made to derive the relevant structural and/or physicochemical parameters of odor from the relative response spectra of the units recorded. When employed in studies of peripheral olfactory neurons (cf. Gesteland, Lettvin and Pitts, 1965), this paradigm is consistent with the assumption that these neurons contain receptor sites for odoriferous molecules and in effect is aimed at a preliminary answer to the two interrelated questions of how specific are the receptor sites and do the peripheral neurons have more than one type of receptor site. This general paradigm also has been used in studies of central olfactory neurons to test hypotheses about the physical dimensions of odors. For example, Higashino, Takeuchi and Amoore (1969) attempted to verify a stereochemical theory of odor quality by using a battery of pure chemical stimuli and comparing the response spectra of single units recorded in the olfactory bulbs with generalization gradients derived from psychophysical studies.