Malta

Physiology & Behavior, Vol. 65, No. 2, pp. 311–320, 1998 © 1998 Elsevier Science Inc. All rights reserved. Printed in the U.S.A. 0031-9384/98 $19.00 .00

PII S0031-9384(98)00168 -1

The Influence of Odor Type on the Discrimination and Identification of Odorants in Multicomponent Odor Mixtures
ANDREW LIVERMORE*1 AND D. G. LAING† *School of Social Sciences & Liberal Studies, Charles SturtUniversity - Mitchell, Bathurst NSW 2795; and †Faculty of Science and Technology, University of Western Sydney, Bourke Street, Richmond, NSW 2753, Australia Received 7 May 1996; Accepted 4 June 1998
LIVERMORE, A. AND D. G. LAING. The influence of odor type on the discrimination and identification of odorants within multicomponent odor mixtures. PHYSIOL BEHAV 65(2) 311 – 320, 1998.—Using a limited set ofodorants, previous studies have indicated that the ability of humans to discriminate and identify the components of olfactory mixtures is limited to approximately four (24,25,30). However, the ability to generalize these results may have been limited by specific neural or cognitive interactions among the particular odorants used. In the present experiment, 41 subjects examined the influence of odortype (different individual odorants), from two very different odor sets, on the perception of the components of complex mixtures. One set contained odors that were selected by an expert panel to blend well in mixtures (good blenders), whereas the other contained odors that blended poorly in mixtures (poor blenders). The stimuli were common, dissimilar odorants of equivalent, moderate intensity,each of which was a single chemical. A computer-controlled air dilution olfactometer delivered a single odorant or a mixture containing up to eight odorants. Although the poor blenders were more easily discriminated, this superiority was displayed within a narrow range, and the ability of subjects to identify mixture components with either odor set was limited to approximately four. The resultsindicate that, whereas odor type can alter which odorants will be perceived in a mixture, the limited capacity to discriminate mixture components is independent of the type of odorants. These findings are discussed in terms of their implications for olfactory coding. © 1998 Elsevier Science Inc. Olfaction Odor Mixture Human Quality Discrimination Identification

MOST psychophysical studies of odorantmixtures have focused on interactions within simple binary mixtures [e.g., (4 – 6, 26 –28)]. However, few have explored more complex stimuli. Although using two-odor mixtures simplifies the interpretation of the results, it may oversimplify the task that the olfactory system can perform. The odor environment that humans and animals live in contains hundreds of odors, each of which may itself be acomplex mixture of hundreds of components. To test the working limits of the system, and to observe the way that it processes and selects olfactory information, it is necessary to investigate multicomponent odor stimuli. The results of previous studies (24,25,30) have indicated that humans have great difficulty discriminating and identifying more than three components in olfactory mixtures. Withthree-, fourand five-component mixtures, the most common number of odors selected, and the most common number correctly identified, was three. Furthermore, the study by Livermore and Laing (30) indicated that, irrespective of the extent of training, the number of years of professional experience in olfactory discrimination, and the amount of cognitive assistance given, the capacity of the olfactorysystem to identify mixture components was limited to
1

four. However, it is unclear whether this limited capacity was due to physiological, stimulus, or methodological factors. A recent experiment (25) that used similar, but not identical, odorants to that of Laing and Francis (24) and Livermore and Laing (30), established that the limited discrimination of mixture components found in previous...