detection of an expected event is a global property of the array, and that it is not manifested in the elements where "hits" occur, nor is that information contained in the output.
Conclusion. The dynamics of this model indicate that the operation of the olfactory system depends heavily on centrifugal controls, i.e. signals carried by neural fiber bundles that run from the brain to the olfactory bulb and cortex in the direction opposite to the main olfactory information flow. These signals can be expected to give nonspecific commands to the bulb, such as "turn on", "form template to whatever is there", "activate template to whatever is there", "suppress all activated templates", "habituate to whatever is there", "suppress all habituation", and so on. There are known anatomically to be ten or more centrifugal paths to the bulb and yet others to the olfactory cortex, although their functions have not yet been physiologically sorted out.
According to the model, when an animal is aroused by a novel motivating stimulus in any sensory modality, it will orient to the on-going odor input and thereby form a compound template to a test odor in association with background odors. Presentation of background odors subsequently may suffice for selective activation of the compound template, and selective habituation "to whatever is there" may suppress the background components, leaving the component of the template for the test odor as a matched filter for that odor. The neural active state manifested in the EEG can be viewed as a neural search image for an expected odor, or as a hypothesis to be tested against sensory input, or as the objective substrate of a mental image or recollection of the odor.
The applicability of this model to the visual somesthetic and auditory systems has not yet been determined; too little is known about their central physiology, although the requisite gating and carrier frequencies are abundantly manifest in their EEG's. Certainly neocortical dynamics are more complex than the mechanisms of the olfactory forebrain, and the olfactory perceptual apparatus together with the archicortex phylogenetically preceded the emergence of neocortex in mammals. The manner of growth and the need for commonality of coding among the several modalities give reason to believe that neocortical mechanisms may be adaptations and elaborations of the perceptual processes that first appeared in the sphere of olfaction, such that the findings reported here may have a certain generality.
Amoore J. E., 1971. Olfactory genetics and anosmia, "Handbook of Sensory Physiology. IV", Chemical Senses, Part 1, Olfaction ( L. M. Beidler, ed.), Springer-Verlag, Berlin, Chapter 11.
Babloyantz A. and L. K. Kaczmarek, 1979. Self-organization in biological system with multiple cellular contacts, Bull. Math. Biol. 41, 193-201.
Freeman W. J., 1975. Mass action in the nervous system, Academic Press, New York.
_____, 1978a. Spatial properties of an EEG event in the olfactory bulb and cortex, "Electroencephalogr. Clin. Neurophysiol. 44", 586-605.