Dans L'érotisme ou le mensonge de Freud, j'ai écrit (p. 18) que l’approche scientifique stricte ne semble pas encore prête à se prononcer sur un rejet inné et universel des odeurs fécales. J'ai ensuite cité ce passage d'un article publié en 1987 par France Bellisle et Patrick Even, Les raisons du corps, dans le n° 92 (Odeurs) de la revue Autrement, série Mutations :
L’étude du cerveau a confirmé que l’information gustative et surtout olfactive est transmise par des régions du cerveau qui sont impliquées dans l’émotion. Pourtant, jusqu’à présent, on n’a pas réussi à mettre en évidence un comportement d’acceptation ou de rejet inné et universel pour des odeurs.

Mais mon livre venait à peine d'être envoyé chez l'imprimeur que je tombais sur ces deux articles :

 

http://tubeaessai.blogs.nouvelobs.com/archive/2009/01/18/le-delicat-parfum-du-cheeseburger.html

NouvelObs.com

18.01.2009

Le délicat parfum du cheeseburger

Les souris et les humains ont un point commun: les deux espèces sont attirées (ou repoussées) par les mêmes odeurs. Telle est la conclusion d’une étude publiée le 16 janvier dans la revue «PLoS ONE», par l’équipe de Nathalie Mandairon (CNRS, Université Lyon 1). Des souris de laboratoire d’une part, et des étudiants de l’autre, ont été soumis à une longue série de stimuli olfactifs. Pour les souris, on mesurait le temps accordé à l’exploration de la senteur proposée; les cobayes humains, eux, étaient priés d’attribuer une note de 1 (la sensation la plus déplaisante) jusqu’à 9 (la plus agréable). Or les souris sont tombées parfaitement d’accord avec les humains – avec par exemple un clair plébiscite du géraniol (senteur florale), et une détestation commune du gaïacol (odeur de brûlé). Les chercheurs en concluent que les préférences olfactives, en partie innées, sont déterminées par la structure chimique des molécules odorantes. Et surtout (horreur!) que ceci «permet d’envisager des applications, dans l’industrie agro-alimentaire par exemple».

Autrement dit, nos futurs aliments risquent fort d’être testés sur les souris, ce qui coûtera moins cher que les jurys de bipèdes. Or, en dépit du film «Ratatouille», on est en droit d’exprimer quelques doutes sur le sens gastronomique des rongeurs. Mais le plus inquiétant, c’est que la découverte lyonnaise va fournir de nouveaux outils aux spécialistes du «marketing olfactif» - la technique qui consiste à manipuler les consommateurs en parfumant les magasins. Déjà en 1957, le regretté Gaston Lenôtre avait imaginé un dispositif diffusant une attirante odeur de pain frais aux abords de sa toute première boutique – au risque d’attirer aussi les souris... Mais au moins s’agissait-il de la véritable odeur du pain, en provenance des fours. Or depuis, la méthode a fait énormément de progrès, notamment grâce aux travaux de son gourou américain, Eric Spangenberg, de la Washington State University.

Ainsi, nous apprenait récemment le magazine «The Economist», on sait aujourd’hui mener les consommateurs par le bout du nez, avec des odeurs incitant à l’achat - même si elles n’ont aucun rapport avec les objets proposés. Par exemple, un discret parfum de chocolat améliore les ventes de chaussures pour dames. Et l’odeur du foin coupé celles des meubles de jardin. Des expériences concluantes ont été menées à Londres, chez Harrods: en somme, désormais, dans les magasins, les humains sont traités comme des souris de laboratoire. La chaîne de «néfaste food» Burger King est allée encore plus loin: sous le nom de «Flame», elle a vendu en décembre – sous forme d’aérosols de 5 ml vendus 3,99$ pièce – «un parfum de séduction, avec une note de viande grillée». Enveloppé d’une pareille fragrance, on ne s’étonnera pas d’être poursuivi par les souris… et par les chiens.

 

*********************************************

 

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0004209

Humans and Mice Express Similar Olfactory Preferences

Nathalie Mandairon, Johan Poncelet, Moustafa Bensafi, Anne Didier

Université Lyon 1, CNRS, UMR5020, Neurosciences Sensorielles, Comportement, Cognition, Lyon, France

Abstract

In humans, the pleasantness of odors is a major contributor to social relationships and food intake. Smells evoke attraction and repulsion responses, reflecting the hedonic value of the odorant. While olfactory preferences are known to be strongly modulated by experience and learning, it has been recently suggested that, in humans, the pleasantness of odors may be partly explained by the physicochemical properties of the odorant molecules themselves. If odor hedonic value is indeed predetermined by odorant structure, then it could be hypothesized that other species will show similar odor preferences to humans. Combining behavioral and psychophysical approaches, we here show that odorants rated as pleasant by humans were also those which, behaviorally, mice investigated longer and human subjects sniffed longer, thereby revealing for the first time a component of olfactory hedonic perception conserved across species. Consistent with this, we further show that odor pleasantness rating in humans and investigation time in mice were both correlated with the physicochemical properties of the molecules, suggesting that olfactory preferences are indeed partly engraved in the physicochemical structure of the odorant. That odor preferences are shared between mammal species and are guided by physicochemical features of odorant stimuli strengthens the view that odor preference is partially predetermined. These findings open up new perspectives for the study of the neural mechanisms of hedonic perception.

 

Citation: Mandairon N, Poncelet J, Bensafi M, Didier A (2009) Humans and Mice Express Similar Olfactory Preferences. PLoS ONE 4(1): e4209. doi:10.1371/journal.pone.0004209

Editor: Leslie B. Vosshall, The Rockefeller University, United States of America

Received: October 24, 2008; Accepted: December 4, 2008; Published: January 16, 2009

Copyright: © 2009 Mandairon et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This research was supported by the Marie Curie Foundation, MOIF-CT-2005-51474, and the CNRS. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

These authors contributed equally to this work.

 

Introduction

Olfaction is of great importance to mammals' survival, influencing a variety of social activities, including recognition, mate selection, fear responses to predator odors, and food intake [1]–[3]. Of the various aspects of olfactory perception, pleasantness is particularly fundamental, and dominates odor perception [4]. Most odors we encounter induce attraction or repulsion behavior. However, it is unclear what it is that makes a given component pleasant or unpleasant. While, in the visual and auditory modalities, perception can be predicted from the physical properties of the stimuli, the rules that govern the relationship between perception and chemical structure in olfaction are largely unknown, making it difficult to predict the perceptual properties of novel odorants. In a recent study, Khan and collaborators built a mathematical model to predict the hedonic valence of molecules in humans on the basis of their physicochemical properties. While it is well established that odor hedonic perception is strongly influenced by experience and learning [5], [6] and that its representations are characterized by a high level of plasticity [7]–[9], these authors suggest that it nevertheless remains partially dependent on the odorants' physicochemical properties [10].

This, if true, suggests a predetermination of odor preferences [10], and it might consequently be hypothesized that other species may show similar odor preferences to humans. The present study found that the same odorants were similarly attractive to mice and humans, revealing for the first time a component of olfactory preference conserved across the two species. Consistent with this, behavioral responses to the odorants (hedonic rating and sniffing in humans; investigation time in mice) were found to correlate with the physicochemical properties of the molecules, suggesting that olfactory preferences are indeed partially engraved in the structure of the odorant molecule. Our data support the view that odor preferences are partially predetermined, in contrast to the more common view of them as predominantly shaped by experience.

 

Results

Our hypothesis was that humans and mice exhibit similar preferences towards the same odorants. To test it, we first assessed odor preference in humans (through odor pleasantness ratings) and in mice (through investigation time) and then investigated the relationship between these odor preferences and odorant structures.

Odorant selection was based on a recent study by Khan et al. [10]. Using principal component analysis (PCA, a multivariate statistical method), these authors generated two odor spaces:

  1. A perceptual space generated from a matrix of 144 odorants and 146 verbal labels describing perceptual properties of odors (see Dravnieks [11]). Given the complexity of interpreting such a multidimensional matrix, PCA was applied to reduce this multidimensional space to a small number of principal components (PCs). The PCs are ordered so that each successive PC has the maximal possible variance, the first PC explaining the most variance of the original data set.

  2. A physicochemical space generated from a matrix of 1565 odorants and 1513 physicochemical descriptors. Here again, PCA reduced this multidimensional space to a small number of PCs.

In the present study, the odorants used in Experiment 1 and in Experiment 2 were respectively from Khan's perceptual and physicochemical space.

Mice and humans express similar odor preferences

Experiment in mice and humans using odorants selected according to their perceptual pleasantness (Experiment 1).

Odor preference was recorded using the same odorants in both species. In mice, odorant investigation time, a behavioral measure of the mouse's interest in and attraction for the odorant, was used as an index of odor preference: when smelling an attractive odor, mice spend more time investigating the odorant source than when encountering a less attractive odor [12], [13]. Odorants were presented to mice using a computer-assisted hole-board, and odorant investigation time (nose poking into the hole) was automatically recorded using electronic sensors [14]. Experiments in humans consisted in sniffing odorized vials and rating compound pleasantness, intensity and familiarity on a 9-point scale (“not at all” (1) to “extremely” (9) pleasant, intense or familiar).

In the first experiment, nine odorants were selected from the perceptual space of Khan et al. [10]. Multiple regression analysis (F[3,89] = 4.111, p<.001) revealed that the odorants investigated longest by mice were those rated most pleasant by humans (t(89) = 2.232, p<.003) (Fig. 1A). Mouse investigation time was also compared to other aspects of human olfactory perception represented by intensity and familiarity ratings (see Materials and Methods): no correlation was found, either with intensity (t(89) = 1.048, p>.05) or familiarity rating (t(89) = .340, p>.05). Of the different parameters of human olfactory perception measured here (hedonic, intensity and familiarity ratings), mouse investigation time thus correlated only with human olfactory pleasantness.

 

Figure 1. Mice and humans express similar odor preferences.

A. In experiment 1, a significant and positive correlation was found between odor investigation time in mice and odor hedonic rating in humans. B. In experiment 2, using different odorants, a significant positive correlation was also observed between odor investigation time in mice and odor hedonic rating in humans.

doi:10.1371/journal.pone.0004209.g001

 

In brief, humans and mice showed similar preferences towards the nine odorants tested.

Experiment in mice and humans using odorants selected according to their physicochemical properties (Experiment 2).

To further investigate similarity in odor preference, the above experiment was replicated using ten different odorants, this time selected from the physicochemical rather than the perceptual space of Khan's study [10]. Again, a significant correlation between odor investigation time in mice and odor hedonic response in humans was observed (F[1,149] = 31.190, p<.0001) (Fig. 1B).

Pooling data from both experiments confirmed that odor investigation time in mice correlated positively with human hedonic ratings (F[1,239] = 41.709, p<.0001).

Importantly, in order to compare homologous behaviors in humans and mice, sniffing time, which reflects odor pleasantness [15]–[17], was recorded in humans, and found to correlate with investigation time in mice (F[1,239] = 15.535, p<.0001).

In brief, odor preferences correlated in mice and humans, whether perceptual or physicochemical criteria were used to select the odorants.

Mouse and human odor preferences correlate with odorant structure

A possible link between odor preference in humans or mice and odorant structure was explored using the first principal component (PC1, which explains the most variance) of the physicochemical space in Khan et al.'s study [10]. PC1 was found to correlate positively with both investigation time in mice (F[1,289] = 22.940, p<.0001) (Fig. 2A) and hedonic rating in humans (F[1,239] = 6.186, p<.02) (Fig. 2B). These results clearly link the physicochemical properties of odorants with their attractiveness in the two different species.

 

Figure 2. Mouse and human odor preferences are driven by odorant structure.

Physicochemical PC1 correlated positively with investigation time in mice (A) and with hedonic rating in humans (B), indicating that hedonic behavior in both species may be driven in part by the physicochemical properties of the molecules.

doi:10.1371/journal.pone.0004209.g002

 

Two groups of odorants with different physicochemical properties evoke distinct behaviors in mice and humans

Strengthening the above results, we observed differences in PC1 values between the first and the second experiment: the average value of PC1 across odorants in experiment 2 was significantly greater than that in experiment 1 (t-test, p<.05, Fig. 3A). As a positive PC1 value corresponds to increased investigation time in mice and increased odor pleasantness in humans (see Fig. 2), this difference may explain the distinct behaviors observed in response to the two sets of odorants used in experiments 1 and 2 (for both humans and mice): mice investigated longer and humans preferred and sniffed odorants longer in experiment 2 than in experiment 1 (t-test, p<.05, Fig. 3B–D). Taken together, these results support the view that odor preference is driven at least in part by the physicochemical properties of the molecules, accounting for the similar olfactory preferences found in both humans and mice.

 

 

Figure 3. Two groups of odorants with different physicochemical properties evoke distinct behaviors in mice and humans.

A. Physicochemical PC1 significantly differed for the two sets of odorants. B. Mice investigated odorants longer in experiment 2 than in experiment 1 C. Humans sniffed odorants longer in experiment 2 than in experiment 1 D. Humans preferred the odorants of experiment 2 to those of experiment 1. (t-test, *: p<.05, **: p<.001, ***: p<.0001)

doi:10.1371/journal.pone.0004209.g003

 

Discussion

Mammals possess an excellent ability to detect and discriminate odorants. They also exhibit odor preferences that seem to be crucial for survival [13], [18], [19] and represent a central component of the quality of human experience [20]. The present study examined whether odor preferences were similar between humans and mice, and might thus include a predetermined component.

A first result of interest was the positive correlation in odor preferences between mice and humans. Odor preferences result from complex physiological and motivational states characterized by experience-dependent plasticity in both animals [21], [22] and humans [4]. In particular, the hedonic representation of smells is not fixed and may be modified by learning and experience in both animals and humans [23]–[25]. For example, a smell may acquire a novel hedonic valence through an associative learning procedure [5], [26]. Moreover, it is well known in humans that odor pleasantness is modulated with repeated exposure to the same stimulus: i.e., without any apparent mediation of environmental stimuli [27]. Although hedonic representations are plastic, as seen above, the present demonstration that two different species exhibited similar preferences for the same odorants provides evidence for odor representations conserved across mice and humans, which emerge independently of life experience. These findings are in line with reports that human newborns exhibit olfactory preferences as shown by behavioral and physiological responses to chemical stimuli [28], and are able to exhibit behavioral markers of disgust in response to unpleasant odors [29]. Such predisposition in odor preference may be underlain by genetically programmed neural circuits, as has been suggested in the olfactory systems of mammals [13], Drosophila melanogaster [30] and Caenorhabditis elegans [31].

Secondly, the present investigation further showed that olfactory preferences in humans and mice are linked to the physicochemical structure of odorants. The data suggest that physicochemical properties are prominent factors in orchestrating the activation pattern of the predetermined neural code for odor preference. It may be asked why odorant structure should correlate more strongly with odor preference in mice than with hedonic perception in humans (F-values of 22.940 in mice, vs 6.186 in humans). One explanation may be that mouse preferences are very weakly colored by the environment (the laboratory mice were raised in an olfactory poor environment) whereas odor pleasantness in adult humans is likely to be more affected by learning and experience. Nevertheless, despite the strong influence of experience, the physicochemical properties of odorants still played a prominent role in determining odor preferences, strengthening Khan et al.'s model [10]. In other words, even if pleasantness is the result of culture, life experience and learning, the present interspecies comparison shows that there is an initial part of the percept which is innate and engraved in the odorant structure.

In conclusion, our phylogenetic heritage includes systems enabling the attribution of a positive or negative value, driving attraction to or avoidance of odorants. This suggests that, upstream of the hedonic plasticity occurring throughout life, we are endowed with a partly predetermined neural basis for these odor hedonic representations–even if the odorant has no biological significance, such as predator, conspecific or spoiled food odors [13], [29], [32]. Perception of the hedonic aspect of odorants is thus a complex process which involves both innate and learned components. Taken as a whole, these results substantially affect our view of olfactory hedonic perception and open up new avenues for the understanding of its neural mechanisms. They also suggest that odor exploration behavior in mice may be used to predict human olfactory preferences.

 

Materials and Methods

Experiments in mice

Adult (8-week old) male C57Bl6/J mice (Charles River Laboratories, L'Arbresle, France) were tested under procedures in accordance with the European Community Council Directive of 11/24th/1986 (86/609/EEC) and the French Ethics Committee.

Upon arrival in the lab, mice were housed in groups of five in standard laboratory cages and were kept on a 12 hr light/dark cycle (constant temperature), with food and water ad libitum. Experiments were conducted in the afternoon (2–5 pm) on a specially designed computer-assisted one-hole-board apparatus (40×40 cm; central hole 3 cm diameter, 4.5 cm deep), with capacitive sensors to detect automatically the beginning of each trial (when the mouse was placed in the starting area facing the hole) and monitor the duration of nose-poking into the hole. A polypropylene swab impregnated with 60 µL of odorant (1 Pa) was placed at the bottom of the hole, under a grid and covered with bedding [14]. The bedding was replaced after each trial. One odorant was presented per day (random order for each animal). Each trial lasted 2 min. Duration of nose-poking into the hole was used as a measure of odor preference. Ten mice were tested in the first experiment and twenty in the second.

Experiments in humans

Respectively ten (mean age, 21.1 yr+/−1.07) and twenty human subjects (mean age, 21.85 yr+/−3.37) recruited from the University of Lyon (France) participated in experiments 1 and 2. Olfactory and/or neurological disease was the exclusion criterion. The study was conducted in accordance with the Declaration of Helsinki.

Testing was performed in an experimental room designed specifically for olfactory experiments. Odorants were presented in 15 ml flasks (opening diameter: 1.7 cm; height: 5.8 cm; filled to 5 ml) and were absorbed on a scentless polypropylene fabric (3×7 cm; 3 M, Valley, NE, USA) to optimize evaporation and air/oil partitioning.

After providing written informed consent to the procedure, which was approved by the “LyonSud-Est2” ethics committee, subjects were taken to the test room, where they sniffed each vial in random order and rated compound pleasantness on a 9-point scale (from 1: “not at all pleasant” to 9: “extremely pleasant”). Subjects rated compound intensity and familiarity (in experiment 1) on similar scales. The instructions given to the subjects were as follows: “You are going to smell several odors one after the other. Your task will be to sniff each vial and then to decide how intense, pleasant or familiar the smell was. To give your estimates, you will rate each odorant on a scale from 1 (not at all intense, familiar or pleasant) to 9 (very intense, familiar or pleasant).” Once the instructions had been read, the experimental session started. Odorants were presented every 45 sec.

Physiological data were acquired using a PROCOMP+ system (Thought Technology, Montreal, Canada; sampling rate, 32 Hz). Sniff duration was measured using an airflow sensor (AWM720, Honeywell, France) connected to nasal cannulae positioned in both nostrils [33] throughout the experimental sessions.

Odorants

The nine odorants used in the first experiment and the ten used in second were diluted in mineral oil so as to achieve an approximate gas-phase partial pressure of 1 Pa (Table 1). Briefly, vapor pressures of pure odorants were estimated using ACD Chem-Sketch software (Advanced Chemistry Development, Toronto, Ontario, Canada) and variously diluted in mineral oil to concentrations theoretically emitting the same vapor-phase partial pressure for each odorant.

 

Table 1. Odors and their percentage (vol/vol) dilutions (1 Pa).

doi:10.1371/journal.pone.0004209.t001

 

Data analysis

Statistical analysis used SYSTAT software (SSI, Richmond, CA). In Experiment 1, “odor investigation time” in mice on the one hand and “odor pleasantness, intensity and familiarity” in humans on the other hand were compared by multiple regression analysis. In Experiment 2, “odor investigation time” in mice and “odor pleasantness” in humans were compared by simple regression analysis. Human sniff duration was compared to mouse investigation time using a similar analysis. Finally, the relation between “odorant PC1” and “investigation time” in mice (or “odor pleasantness” in humans) was also assessed by simple regression analysis.

Author Contributions

Conceived and designed the experiments: NM MB AD. Performed the experiments: NM JP MB. Analyzed the data: NM MB. Wrote the paper: NM MB AD.

References

  1. Hebb AL, Zacharko RM, Gauthier M, Trudel F, Laforest S, et al. (2004) Brief exposure to predator odor and resultant anxiety enhances mesocorticolimbic activity and enkephalin expression in CD-1 mice. Eur J Neurosci 20: 2415–2429. Find this article online
  2. Small DM, Veldhuizen MG, Felsted J, Mak YE, McGlone F (2008) Separable substrates for anticipatory and consummatory food chemosensation. Neuron 57: 786–797. Find this article online
  3. Vosshall LB (2005) Social signals: the secret language of mice. Curr Biol 15: R255–257. Find this article online
  4. Engen T (1988) The acquisition of odour hedonics. In: Van Toller S, Dodd GH, editors. Perfumery: The psychology and biology of fragrance. Londres: Chapman and Hall. pp. 79–90.
  5. Barkat S, Poncelet J, Landis BN, Rouby C, Bensafi M (2008) Improved smell pleasantness after odor-taste associative learning in humans. Neurosci Lett 434: 108–112. Find this article online
  6. Stagnetto J, Rouby C, Bensafi M (2006) Contextual cues during olfactory learning improve memory for smells in children. European Review of Applied Psychology 56: 253–259. Find this article online
  7. Bensafi M, Rinck F, Schaal B, Rouby C (2007) Verbal cues modulate hedonic perception of odors in 5-year-old children as well as in adults. Chem Senses 32: 855–862. Find this article online
  8. de Araujo IE, Rolls ET, Velazco MI, Margot C, Cayeux I (2005) Cognitive modulation of olfactory processing. Neuron 46: 671–679. Find this article online
  9. Herz RS (2003) The effect of verbal context on olfactory perception. J Exp Psychol Gen 132: 595–606. Find this article online
  10. Khan RM, Luk CH, Flinker A, Aggarwal A, Lapid H, et al. (2007) Predicting odor pleasantness from odorant structure: pleasantness as a reflection of the physical world. J Neurosci 27: 10015–10023. Find this article online
  11. Dravnieks (1985) Atlas of odor character profiles. Philadelphia, PA: ASTM.
  12. Doty RL (2002) Methods for Determining Odor Preferences in Nonhuman Mammals. In: Doty RL, editor. Handbook of olfaction and gustation 2nd Edition. New York: Marcel Dekker. pp. 403–408.
  13. Kobayakawa K, Kobayakawa R, Matsumoto H, Oka Y, Imai T, et al. (2007) Innate versus learned odour processing in the mouse olfactory bulb. Nature 450: 503–508. Find this article online
  14. Mandairon N, Sacquet J, Garcia S, Ravel N, Jourdan F, et al. (2006) Neurogenic correlates of an olfactory discrimination task in the adult olfactory bulb. Eur J Neurosci 24: 3578–3588. Find this article online
  15. Bensafi M, Porter J, Pouliot S, Mainland J, Johnson B, et al. (2003) Olfactomotor activity during imagery mimics that during perception. Nat Neurosci 6: 1142–1144. Find this article online
  16. Bensafi M, Pouliot S, Sobel N (2005) Odorant-specific patterns of sniffing during imagery distinguish ‘bad’ and ‘good’ olfactory imagers. Chem Senses 30: 521–529. Find this article online
  17. Bensafi M, Sobel N, Khan RM (2007) Hedonic-specific activity in piriform cortex during odor imagery mimics that during odor perception. J Neurophysiol 98: 3254–3262. Find this article online
  18. Isles AR, Baum MJ, Ma D, Keverne EB, Allen ND (2001) Urinary odour preferences in mice. Nature 409: 783–784. Find this article online
  19. Levy F, Keller M, Poindron P (2004) Olfactory regulation of maternal behavior in mammals. Horm Behav 46: 284–302. Find this article online
  20. Rolls ET (2000) Precis of The brain and emotion. Behav Brain Sci 23: 177–191; discussion 192–233. Find this article online
  21. Kindermann U, Hudson R, Distel H (1994) Learning of suckling odors by newborn rabbits declines with age and suckling experience. Dev Psychobiol 27: 111–122. Find this article online
  22. Moriceau S, Sullivan RM (2006) Maternal presence serves as a switch between learning fear and attraction in infancy. Nat Neurosci 9: 1004–1006. Find this article online
  23. Brennan PA, Keverne EB (1997) Neural mechanisms of mammalian olfactory learning. Prog Neurobiol 51: 457–481. Find this article online
  24. Kay LM, Laurent G (1999) Odor- and context-dependent modulation of mitral cell activity in behaving rats. Nat Neurosci 2: 1003–1009. Find this article online
  25. Wilson D, Stevenson R (2006) Learning to smell. Baltimore: Johns Hopkins UP.
  26. Sullivan RM, Taborsky-Barba S, Mendoza R, Itano A, Leon M, et al. (1991) Olfactory classical conditioning in neonates. Pediatrics 87: 511–518. Find this article online
  27. Prescott J, Kim H, Kim K (2008) Cognitive Mediation of Hedonic Changes to Odors Following Exposure. Chemosensory Perception in press. Find this article online
  28. Schaal B, Hummel T, Soussignan R (2004) Olfaction in the fetal and premature infant: functional status and clinical implications. Clin Perinatol 31: 261–285.vi–vii Find this article online
  29. Soussignan R, Schaal B, Marlier L, Jiang T (1997) Facial and autonomic responses to biological and artificial olfactory stimuli in human neonates: re-examining early hedonic discrimination of odors. Physiol Behav 62: 745–758. Find this article online
  30. Suh GS, Wong AM, Hergarden AC, Wang JW, Simon AF, et al. (2004) A single population of olfactory sensory neurons mediates an innate avoidance behaviour in Drosophila. Nature 431: 854–859. Find this article online
  31. Troemel ER, Kimmel BE, Bargmann CI (1997) Reprogramming chemotaxis responses: sensory neurons define olfactory preferences in C. elegans. Cell 91: 161–169. Find this article online
  32. Steiner JE, Glaser D, Hawilo ME, Berridge KC (2001) Comparative expression of hedonic impact: affective reactions to taste by human infants and other primates. Neurosci Biobehav Rev 25: 53–74. Find this article online
  33. Frank RA, Dulay MF, Gesteland RC (2003) Assessment of the Sniff Magnitude Test as a clinical test of olfactory function. Physiol Behav 78: 195–204. Find this article online