j ourna l homepage: www.e lsev ie r.com/ locate /phb
Perception of odor blending mixtures in the newborn rabbit
Gérard Coureaud a,⁎, Thierry Thomas-Danguin b, Elodie Le Berre b, Benoist Schaal a
a Centre Européen des Sciences du Goût, Equipe d'Ethologie et de Psychobiologie Sensorielle, UMR 5170 CNRS/UB/INRA, Dijon 21000, Franceb UMR 1129 FLAVIC, ENESAD, INRA, UB, Dijon 21000, France
essing of complex odor mixtures is elemental or configural. Here, we challengedthese processes in newborn rabbits, in evaluating their perception of a binary odor mixture for whichperceptual blending occurs in humans. This model of newborn animal was interesting since generalquestions remain on how odor cues are processed in immature organisms, and since rabbit pups presentabilities of rapid odor learning. In the present study, we first demonstrated (Exp. 1) that rabbit pups rapidlyacquired the odor of the binary mixture through associative conditioning (when the mammary pheromonewas used as unconditioned stimulus). Then, we compared how they responded to the mixture, itscomponents and the mammary pheromone, after they had learned the mixture or one of its constituents.After they had learned the odor of the mixture, they responded to its odor and the odor of its constituents(Exp. 2). However, after they had learned one constituent's odor, they responded to this odor but not to themixture's odor (Exp. 3). The response to the mixture appeared nevertheless when pups successively acquiredthe odor of the two components (Exp. 4). Therefore, both elemental and configural processing of the mixtureseem to be displayed by rabbit pups, suggesting that neonatal perception of a simple odor mixture mayinvolve more than the perception of its constituents.
In the animal kingdom, odors are important vectors of informationlikely to elicit behavioral decisions (e.g., interaction with conspecifics,competitors and predators, selection of habitats, preys and food; e.g.[1,2]). Nevertheless, the odors of the surroundings are rarelycomposed of single odorants, and animals have usually to processodor mixtures, i.e. to detect and discriminate odorants in backgroundsof other odors [3]. Such processing of odors varying in their chemicalcomplexity occurs from the earliest developmental stage. Thus,mammalian fetuses are exposed to odorants in amniotic fluid, amixture containing for instance more than 390 distinct compounds inhumans [4]. At birth, newborn mammals are exposed to the odor ofmilk, a fluid including more than 150 volatile compounds in the rabbit[5]. Despite the complexity of these fluids, fetuses and newbornsdetect, respond, and often learn some of the odorants carried in thefluids [6,7] (review in [8]). However, how odor mixtures are perceivedby fetuses and newborns, and how they influence their behavior,remain largely unexplored.
In adult mammals, the perception of an odor mixture has beensuggested to be elemental (analytical) or configural (configurational,synthetical) [9,10]. On one hand, elemental processing of odor mix-tures consists in the perception of the specific odor of each mixture's
+33 3 80 68 16 01.).
l rights reserved.
component, themixture smelling thus like its constituents [11–14]. Onthe other hand, configural processing of odor mixtures occurs whenthe odor quality of the mixture differs from that of its components. Inthat case, themixture gives rise to a novel perceptual odor quality [15–18]. Moreover, a mixture may be represented by a configural elementunique to this mixture in addition to the individual elements thatcompose each of its constituent [19]. Discussing these mechanisms,Kay et al. [20] suggested that configural processing may be weak orrobust depending on whether the odor of the whole mixture smellspartially as the odors of the mixture's constituents, or does not smellat all as the constituents. In humans, the quality of odor mixtures hasbeen investigated [21,22] and evidence exists for a configural processof certain mixtures [23]. A recent study revealed that the odortypicality of mixtures including two to six components was sig-nificantly higher than the odor typicality of their components [24].These findings confirm that perceptual blending could occur inmixture of odorants. This perceptual blending may thus give rise tothe perception of a novel odor in addition to, or in place of, the odor ofeach component.
In newborn mammals, the contribution of elemental and/orconfigural processes in the perception of odor blending mixturesremained to be explicitly addressed. However, one may suggest theirinvolvement in the neonatal perception of environmental stimuli,sincememory and neural/behavioral plasticity, which are suggested toplay a critical role in odor discrimination and (at least) configuralprocesses [25], are early functional during development [26,27]. Here,we seized the opportunity to address the question of odor processing
in the newborn of amammal species, the European rabbit (Oryctolaguscuniculus). As in other mammals, rabbit newborns trace the nipples infollowing odor cues emitted by the mother. During the nursing(occurring once per day, only; [28]), rabbit pups are exposed to amosaic of odors carried by the mother's abdomen and milk [29–32],among which is at least one pheromone, the mammary pheromone(MP). This mono-molecular compound, immediately significant atbirth, strongly releases the typical head searching-oral graspingresponses usually displayed by pups during nursing [5,33]. Moreover,the MP is sufficient by itself to engage, as an unconditioned stimulus,the associative learning of any novel odorant (conditioned stimulus)by rabbit pups. This learning is rapid (single-trial) and appetitive [34].Here, in Experiment 1, we used this reinforcing ability of the MP toevaluate whether rabbit newborns show MP-induced conditioningnot only to a single odorant but also to a binary odormixture (previousstudies have demonstrated in other experimental conditions thatrabbit pups can learn complex odor mixtures [35,36]). We voluntarilychose to use a binary mixture previously demonstrated as carryingblending property in human adults [24]. We indeed hypothesized thatthis mixture could engage configural process also in newborns, andeven in newborns from another species (common mechanismsinvolved in the perception of odor mixtures have been suggested indistinct species, e.g. in adult rats and humans [37]). In Experiment 2, toassess this hypothesis, we compared the response of rabbit pups to theodors of the mixture and of its components, after their initial learningof the mixture. If configural process effectively occurred, the pupsshould respond to the mixture but not to its components. InExperiments 3 and 4, knowing that prior exposure to one componentof a non-blending mixture can influence the later discrimination ofthe mixture from its components (in adult rats [38]), we also testedthe behavior displayed to the mixture and to its components by pupshaving learned one of the mixture's component, or both componentssuccessively. Again, if perceptual blending (configural process) wasinvolved in the perception of the mixture, the previous learning ofone or two of its constituents should not induce the response to themixture.
2. General methods
2.1. Animals and housing conditions
The animals were New-Zealand rabbits (Charles River strain,France) originating from the breeding colony of the Centre deZootechnie (Université de Bourgogne, Dijon). Females and maleswere kept in individual cages, and a nest-box (0.39×0.25×0.32m)wasadded on the outside of the pregnant females' cages 2 days beforedelivery (the day of delivery was considered as day 0). To equalize thenursing experience of the pups, all the females had access to their nestbetween 11:30 and 11:45 a.m. This procedure allowed to follow thebrief (3–4 min) daily nursing of the species [28]. The animals werekept under a constant 12:12 light:dark cycle (light on at 7:00 a.m.)with ambient air temperature maintained at 21–22 °C. Water andpelleted food (Lapin Elevage 110, Safe, France) were provided adlibitum. A total of 268 newborns (1 to 3-day-old) born from55 femaleswere used in the study.
Regarding the care and experimental use of animals, the local,institutional and national rules (French Ministries of Agriculture, andof Research & Technology) were strictly followed.
2.2. Odorants
The odorants consisted of the Mammary Pheromone (MP; 2-methylbut-2-enal), Ethyl isobutyrate (odorant A), Ethyl maltol(odorant B), Ethyl caproate (odorant C), the AB binary mixture, andthe MP–A, MP–B and MP–AB blends. All the compounds werepurchased from Sigma-Aldrich (France).
The MP was used as an unconditioned stimulus to induce thelearning of other odorants (conditioned stimuli), or as a control toensure that the pups were awake and at the similar state of generalresponsiveness at the behavioral test time. In the latter case, the MPwas diluted in a solvent constituted by hydroalcoholic solution [0.1%ethanol (anhydrous, Carlo Erba, France) in purified water (MilliQSystem, Millipore®, France)], except in one occasion detailed inExperiment 1 (purified water only). Alone, the MP was always used ata concentration of 10−5 g/ml, a level known to release high orocephalicresponses in 1 to 3-day-old rabbit pups [39].
For the MP-induced conditioning sessions, the MP–A and MP–Bblends were prepared in solvent at a final concentration of 10−5 g/mlof each constituent (the MP-induced learning of a single odorant ishighly efficient at this concentration [34]). The MP–AB blend wasprepared in solvent at a final concentration of 10−5 g/ml, where ABincluded 0.3×10−5 g/ml of odorant A and 0.7×10−5 g/ml of odorant B(this 30/70 v/v ratio of A/B was chosen since it generates theperception of pineapple odor in human adults due to odor-blending[24,40]). Behavioral assays were run with the same solutions thanthose prepared for the conditioning.
Ethyl caproate (odorant C) was used as a control odorant (at 10−5 g/ml, diluted in solvent) in all experiments. It offered the advantage tocarry an odor quality in humans (pineapple) similar to the one ofthe AB mixture [24,40]. Here, we used it to assess the selectivity ofthe responses displayed by pups to the binary mixture (the fact thatrabbit pups perceive a pineapple odor in ethyl caproate is howeverhypothetical).
Finally, the solvent (0.1% ethanol in purified water) was also usedfor control experiments, either alone or associated to the MP.
2.3. Odor conditioning and behavioral assay
The conditioning sessions and behavioral assays were run in anexperimental room isolated from the breeding room. The pups from asame litter were transferred into a box lined with nest materials andmaintained at room temperature for the duration of the conditioningor assay (10 min maximum).
The MP-induced conditioning was carried out in 1 or 2-day-old rabbit pups, following a procedure previously described [34].Immediately before the conditioning session, 2 ml of the MP-single odorant or MP-AB blends were pipetted on a pad (19×14 cm,100% cotton) in a standardized manner. The pad was then held withgloved hand 2 cm above the litter for 5 min. Five minutes after the endof the conditioning, the pups were individually marked (withscentless ink) and returned to their nest. The box containing the litterwas rinsed with alcohol and distilled water after each conditioningsession.
The behavioral assay consisted in an oral activation test [5,7,34]during which the pup was immobilized in one gloved hand of theexperimenter, its head being left free, and the test odor was presentedfor 10 s with a glass-stick 0.5 cm in front of the nares. A test wasconsidered positive when the stimulus elicited head searchingmovements (vigorous, low amplitude horizontal and vertical scanningmovements of the head, displayed after stretching of the neck towardthe stick) followed by oral grasping movements (labial seizing of thestick extremity). On the contrary, non responding pups displayed noresponse but sniffing to the stimulus. Each pup participated in onlyone experiment, but was successively tested for its responses to 2 to 4stimuli (depending on the experiment; no more than 4 stimuli werepresented to avoid tiredness or habituation). The successive testingconsisted in the presentation of a first stimulus to all the pups from asame litter, then a second stimulus and so on, with an inter-trialinterval of 120 s. If a pup responded to a stimulus, its nose was softlydried with absorbing paper before the next stimulation. The order ofstimuli presentation was counterbalanced, except that the MP wasalways applied last, as a control.
196 G. Coureaud et al. / Physiology & Behavior 95 (2008) 194–199
To minimize litter effects, the experimental groups were drawnfrom 4 to 10 litters, with a maximum of 5 pups conditioned and/ortested per litter and per group. The conditioning and testing werealways run in the morning, 1–2 h before the daily nursing, to equalizethe pups' motivational state and limit the impact of satiation on theresponse [41].
2.4. Statistics
The frequencies of pups responding in the behavioral test werecompared using the χ2 test of Pearson (with Yates correction whennecessary) when the groups of pups were independent (i.e., distinctgroups tested for their response to a same stimulus), or the Cochran'sQ test or χ2 test of McNemar when the groups were dependent (i.e.,pups from a same group tested for their response to several stimuli).All analyses were conducted using the Systat software (Evanston,USA).
3. Experiment 1: pheromone-induced learning of a binaryodor mixture
To investigate whether the pups showed MP-induced conditioningto the ABmixture, 20 pups (from 4 litters) were exposed to theMP–ABblend and tested the day after for their response to AB+ (conditionedstimulus) and MP (unconditioned stimulus). As a result, 90% of thepups responded to AB+, a rate similar to that displayed in response tothe MP (95%; χ2=0.5, pN0.05). This activity of the mixture could notbe due to its spontaneous saliency for rabbit pups, nor to the saliencyof its components, since the single odorant A, B and the AB mixturereleased no response but sniffing when tested with 30 independentpups (from 6 litters; comparison of AB+ vs AB activity: χ2=42.2,pb0.001; Fig. 1a). Thus, the results seemed to confirm that the MPfunctioned as an unconditioned stimulus able to promote the learningof a binary mixture. However, two supplementary conditions neededto be assessed to assert this conclusion: 1) the fact that the MP wasdetected by the pups (and notmasked) in theMP–AB blend; 2) the factthat the pup responses to AB+ did not result from the activity of thesolvent.
To assess whether rabbit pups perceived the MP in the ABmixture,33 pups (from 7 litters) were successively tested for their response tothe MP–AB blend and to the MP in distilled water. As a result, 93.9% ofthe newborns responded to MP–AB, a rate of response similar to thatdisplayed to the single MP (96.9%; χ2=0.5, pN0.05; Fig. 1b). Therefore,
Fig.1. Percentage of 2-day-old rabbit newborns responding in an oral activation test to: a. the(−24 h; n=30 pups, 6 litters) or a day after (+24 h; n=20 pups, 4 litters) the mammary pherotested to the MP); b. the MP added to the binary odor mixture (MP+AB) or presented alone (M−24h (n=24 pups, 5 litters) or +24 h after exposure of the pups to the MP⁎S blend (on +24
the MP appeared detectable by pups, and functional as a releaser ofneonatal response, even when it was in solution with the binarymixture.
Finally, to evaluate the hypothesis of a solvent activity, 24 pups(from 5 litters) were exposed to the solvent before any exposure to theMP, and 20 pups (from 4 litters) were exposed to the solvent 24 h afteran exposure to the MP-solvent blend (to assess whether the solventodor might be MP-learned). As a result, no pup responded to thesolvent (both before and after the exposure to the MP-solvent blend),but 95% responded to the MP (comparison solvent vs MP: χ2N17.05,pb0.001; Fig. 1c). This revealed that the solvent was probably notdetected by rabbit pups in our conditions.
To sum up, the response to AB+ displayed 24h after the pups'exposure to theMP–AB blend appeared to result from theMP-inducedlearning of the mixture.
4. Experiment 2: perception of the components when the mixtureis learned
The previous experiment revealed that the MP was sufficient byitself to support the learning of the binary odor mixture chosen for thestudy. Here, we investigated whether rabbit pups were able todiscriminate the mixture from its components after having learnedthe mixture. To that goal, 39 newborns (from 8 litters) wereconditioned to AB by an exposure to the MP–AB blend. The dayafter, the pups were divided in 2 groups (20 and 19 pups from 4 and 4litters, respectively) either tested for their response to A, AB+, C andMP (group 1) or to B, AB+, C and MP (group 2).
Pups from the group 1 strongly responded to the AB+ mixture(90%), a rate that did not differ significantly from those displayed tothe odorant A and to the MP (80% and 90%, respectively; Q=3, df=2,pN0.05). However, they did not respond to the control odorant C(0%; C vs A, AB+ or MP comparisons: χ2N14.1, pb0.001). Pups from thegroup 2 were also highly responsive to AB+ (89.5%), B and MP (84.2and 94.7%, respectively; Q=1.2, df=2, pN0.05). Their level of responseto B did not differ from that displayed to A by pups from the group 1(χ2b0.1, pN0.05). Their response to C was also almost nil (5.2%; C vs B,AB+ or MP comparisons: χ2N10.6, pb0.01). These results are resumedin Fig. 2a (responses displayed by both groups of pups to AB+, C andMP are pooled).
To sum up, when rabbit pups had learned the binary mixture, theylater responded to it but not to a novel odor similar (for humans) inquality. Their response seemed to depend more on the chemical
constituents (A and B) of the ABmixture, and themixture itself (AB or AB+), a day beforemone-induced conditioning (MP⁎AB) of the AB mixture (on +24 h, the pups were alsoP; n=33 pups, from 7 litters); and c. the solvent (S) inwhich the odorants were diluted,h, the pups were also tested to the MP).
Fig. 2. Percentage of 2-day-old rabbit newborns responding in an oral activation test to the constituents of the binary mixture (A and B), the ABmixture, a control odorant carrying anodor quality close to that of the mixture (C), and the mammary pheromone (MP), 24 h after MP-induced conditioning to: a. the AB mixture (n=20 and 19 pups, from 4 and 4 litters,respectively tested to A or B, and all tested to AB+, C and MP); b. the component A of the mixture (n=22 and 25 pups, from 5 and 5 litters, respectively tested to B or AB, and all testedto A+, C and MP); and c. the component B of the mixture (n=20 and 25 pups, from 4 and 5 litters, respectively tested to A or AB, and all tested to B+, C and MP). Responses are pooledbetween the two groups of pups from a same condition.
compounds than on the odor quality of themixture that they acquired.Moreover, having learned the mixture, they responded to itsconstituents, demonstrating their ability to detect and learn alsothese latter during the conditioning.
5. Experiment 3: perception of the mixture when one of thecomponents is learned
The previous experiment showed that newborn rabbits detect andlearn the odor carried by each constituent of the mixture, when theyare conditioned to the mixture. Here, we evaluated whether themixture may also carry an additional odor quality, resulting fromperceptual blending of the constituents. Thus, we tested the capacityof the pups to respond, or not, to themixture after having learned onlyone of its components. For that goal, 47 pups (from 10 litters) wereconditioned to odorant A by associationwith the MP, among which 22and 25 pups were respectively tested the day after for their responseto A+, B, C and MP, or A+, AB, C and MP. Conversely, 45 other pups(from 9 litters) were conditioned to odorant B by exposure to the MP–
Fig. 3. Percentage of 3-day-old rabbit newborns responding in an oral activation test to thpheromone (MP), respectively 48 h and 24 h after their successive MP-induced conditioningb. the components B then A of the binary mixture (n=20 pups, from 4 litters).
B blend, among which 25 and 20 pups were respectively tested fortheir response to B+, AB, C, and MP, or B+, A, C and MP the day after.
As a result, pups conditioned to MP–A displayed 24 h later a rate ofresponse to A+ both high (N77%) and similar to that displayed to theMP (N86%; χ2b0.17, pN0.05). This result confirmed the effectivelearning of odorant A by pups. At the same time, these pups did notrespond significantly to odorants B and C (b4.5% of response), a resultrevealing the selectivity of the odor learning engaged by the MP.Moreover, the pups that responded to A+ did not respond significantlyto the binary mixture (8% of response; comparison A+ vs AB: χ2=13.1,pb0.001), demonstrating their ability to discriminate the odor of themixture from the odor of one of its components (Fig. 2b). The resultswere similar for the pups initially conditioned to theMP–B blend: theystrongly responded to B+ and MP (80% and 91.1% of response,respectively; χ2=1.6, pN0.05), but they did not respond to A and C(b4%; B+ vs A or C comparisons: χ2N12.5, pb0.001) nor to AB (0%;comparison B+ vs AB: χ2=16.05, pb0.001; Fig. 2c).
Thus, having learned one constituent of the binary mixture, thepups responded to this constituent selectively, but not to the mixture
e constituents of the binary mixture (A+ and B+), the AB mixture, and the mammaryto: a. the components A then B of the binary mixture (n=20 pups, from 4 litters); and
198 G. Coureaud et al. / Physiology & Behavior 95 (2008) 194–199
including it. This result was generalized for both constituents of themixture.
6. Experiment 4: perception of the mixture when bothcomponents are successively learned
Experiment 3 suggested that the binary mixture differed from itsconstituents in terms of perception, the learning of one constituentbeing not sufficient to support the response to the mixture. Here, weevaluated whether the successive learning of each constituentallowed the final response to the mixture. In other words, weinvestigated whether the perception of the mixture resulted onlyfrom the perception of a novel odor induced by the blending of theconstituents, or if it also depended on the previous learning of bothconstituents. To that goal, 20 pups (from 4 litters) were exposed to theMP–A blend on day 1, and to the MP–B blend on day 2. On day 3, theywere tested for their response to A+, B+, AB and MP. Twenty otherpups (from 4 litters) were involved in a similar procedure, except thatthey were first exposed to the MP–B then the MP–A blends.
Under these conditions, the pups first conditioned to A displayedhigh and similar levels of responses to all stimuli, including the binarymixture (90%, 95%, 85% and 95% in response to A+, B+, AB and MP,respectively; Q=2.53, df=3, pN0.05; Fig. 3a). Similar results wereobtained for the pups first conditioned to B (90%, 95%, 95% and 100%of response to B+, A+, AB and MP, respectively; Q=4, df=3, pN0.05;Fig. 3b).
Thus, the successive learning of each constituent allowed the laterresponse to both components, but also the response to their binarymixture.
7. General discussion and conclusion
The present study investigated how newborn rabbits processbinary odor mixtures. Using a mixture model inducing perceptualblending in humans, the experiments revealed that 1) pups learned torespond to the odor of a binarymixturewhen it was combinedwith anodor signal eliciting unconditional response (Exp. 1); 2) pupsresponded to the mixture's constituents after being conditioned tothe mixture (Exp. 2); 3) pups did not respond to the mixture afterhaving learned only one of the constituents (Exp. 3), but 4) they didafter having learned both constituents successively (Exp. 4). Thus,rabbit pups appeared able to extract odor cues from amixture, but notnecessarily to generalize from a constituent to a mixture carrying it.
To date, the mammary pheromone emitted by rabbit females havebeen demonstrated to promote the rapid learning of any novel,initially neutral chemical odorant. Here (Exp. 1), the MP appears toengage the rapid learning of a binary odor mixture [more than 89% ofthe pups learned to respond to the mixture previously paired with thepheromone, a rate similar to that displayed (N75%) after the MP-induced learning of a single odorant]. This result and those ofCoureaud et al. [34] are in linewith the theory of elemental processingof odor mixture ([42]; review in [9]), since they underline the pupability to discriminate the MP from the odorant, or the mixture ofodorants, to which it is associated during the conditioning session.Indeed, if robust configural perception and learning [20,43] occurredduring the exposure to the MP–AB blend in Exp. 1, the MP wouldcertainly not act as an unconditioned stimulus, and the pups wouldnot learn to respond to the AB odor. In a more ecological point of view,the results suggest that the MP may act for rabbit pups as a reinforcerallowing to learn odor mixtures carried by the maternal body, andthus to improve the pups' skill to orient to the mother and her nipplesover the first and critical postnatal days [44,45].
The fact that pups respond to the two constituents after havinglearned the binary mixture (Exp. 2) confirms their elemental pro-cessing of the mixture during the conditioning [13,20,46]. Thisobservation is also in line with previous studies reporting that infor-
mation about individual odor components persist in amixture of a fewconstituents. It is usually considered that this persistency results firstfrom the number of odorants: in humans, when less than fourodorants are mixed, odors of the components remained perceptibleeven if the odorants are known to be good blenders [11,47]. Second,this persistency may also results from the perceptual similarity of thecomponents: when odors are dissimilar, they are more likely to beperceived independently [13]. Our results are consistent with theseprevious findings. The mixture we used was indeed binary only, andthe odors of its constituents seemed perceptually distinct, because thepups conditioned to one odorant of the mixture did not respond tothe other one. Therefore, it is likely that newborn rabbits extract theinformation related to the two constituents of the mixture during theconditioning, store this information, and become then able to respondboth to the mixture and the constituents.
In contrast, having learned a single constituent of the mixture,pups do not respond to the mixture (Exp. 3). This result is quitesurprising as pups previously appear to display elemental processingof the binarymixture (Exp. 2). In Experiment 3, onemight suggest thatduring the behavioral test, the odor of one component predominatesin the mixture, as a consequence of odor masking [47–49]. However,the absence of response to the AB mixture after the previous learningof odorant A, as after the learning of odorant B, does not credit sucheffect. Therefore, the pups appear unable to generalize from thecomponent that they have learned to the mixture, a result suggestingtheir configural processing of the mixture during the behavioral assay[13,14,23]. Taken together, Experiments 2 and 3 suggest that whenelemental processing of the binary mixture is engaged during theconditioning, configural processing of the same mixture could bedisplayed during the retention test.
Another conclusion emerging from the absence of response to ABafter the learning of one constituent, is that the binary mixture chosenfor the study is not only, in perceptual terms, the sum of its twodistinct components. The ABmixture, in 30/70 (v/v) proportion, seemsto form a blending mixture (i.e., a mixture inducing perceptualblending) for newborn rabbits, as suspected for this and for othermixtures in adult humans [24,40] and rodents [20]. Due to its blendingproperties, the present AB mixture would therefore engage (at leastpartially, see below) configural processing by rabbit pups. Here, theconfigural processing of this mixture is revealed by behavioral ex-pression and results. More generally, such processing of odor mixturescould find support in neurobiological studies from receptor level[17,50,51] to higher olfactory centers [e.g., olfactory bulb or analogousstructure: 52,53; piriform cortex: 25,54,55].
Finally, onemay note that even though the ABmixture is perceivedby the pups as different from its components (Exp. 3), it does notmeanthat no information about the individual components remainsperceptible in this mixture (as suggested for other mixtures in otheranimals; [13]). Indeed, pups respond to the components of themixture when they have learned the mixture (Exp. 2) and interest-ingly, they respond to the mixture after having learned each of itsconstituents, successively (Exp. 4). Taken together, these results raisetwo hypotheses. First, the blending effect of the AB mixture may beincomplete [56] and its perception could then be supported by a weakconfigural processing [20]. Accordingly, the pups could detect threedistinct stimuli in the AB mixture: the odor of A, the odor of B, and anew percept produced by the AB blend. Then, the quantity of novelinformation carried in the mixture would determine the perceptionthat the pups have of the mixture during the retention test and,consequently, would determine their behavior. Thus, when pups learnthe AB mixture, they may in fact learn the three stimuli carried in themixture, and become then able to respond to each of them,independently. Second, the results may also be interpreted by anovelty effect: when the pups learn only one component of themixture and are later tested for their response to the mixture, thedetection of the unknown odorant in the mixture could restrain their
response. On the contrary, when they successively learn eachconstituent of the binary mixture, they could acquire sufficient salientinformation to respond to the whole mixture.
To conclude, the present study reveals that elemental andconfigural processes may be functional in the neonatal perception ofbinary odor mixtures. These processes may constitute key-mechan-isms underlying olfactory perception and related behaviors through-out ontogenesis. In the newborn, the functionality and synergy ofthese mechanisms may contribute to the optimal reactivity of theyoung to the complex odor mixtures emitted by the mother, andtherefore improve their adaptation to the immediate environment.Here, the binary mixture seems able to give rise to novel perceptualqualities that are not present in the individual components (config-ural), but the individual odors of the components appear to persist inthe mixture (elemental). Therefore, the results tend to confirm in therabbit that the odors of constituents composing a binary mixture mayblend to form a novel perceptible odor, as in humans, even if theblending remains incomplete. The hypothesis of incomplete blendingcharacterizing the rabbit pup perception of our AB mixture shouldnow be further investigated with dedicated experiments, in particularwith the aim to assess the putative interfering impact of novelty.
Acknowledgements
This research could not have been donewithout the cooperation ofValérie Saint-Giorgio, Florent Costilhes and Nicolas Malaty (Centre deZootechnie, Dijon). It was supported by the ANR 2005-CPER-Lorraine-125 and IFR 92.
References
[1] Brown RE. Mammalian social odors: a critical review. Adv Study Behav1979;10:103–62.
[2] Halpin ZTH. Individual odors among mammals: origins and functions. Adv StudyBehav 1986;16:39–70.
[3] Linster C, Smith BH. Generalization between binary odor mixtures and theircomponents in the rat. Physiol Behav 1999;66:701–7.
[4] Antoschechkin AG, Glovkin LA, Bakharev VA. Screening of amniotic fluid metabolitesby gas chromatography–mass spectrometry. J Chromatogr 1989;489:353–8.
[5] Schaal B, Coureaud G, Langlois D, Giniès C, Sémon E, Perrier G. Chemical andbehavioural characterization of the rabbit mammary pheromone. Nature2003;424:68–72.
[6] Menella JA, Johnson A, Beauchamp GK. Garlic ingestion by pregnant women altersthe odor of amniotic fluid. Chem Senses 1995;20:207–9.
[7] Coureaud G, Schaal B, Hudson R, Orgeur P, Coudert P. Transnatal olfactorycontinuity in the rabbit: behavioral evidence and short-term consequence of itsdisruption. Dev Psychobiol 2002;40:372–90.
[8] Schaal B. From amnion to colostrum to milk: odour bridging in early develop-mental transitions. In: Hopkins B, Johnson S, editors. Prenatal development ofpostnatal functions. Westport: Praeger; 2005. p. 52–102.
[9] Harris JA. Elemental representations of stimuli in associative learning. Psychol Rev2006;113:584–605.
[10] McNamara AM, Magidson PD, Linster C. Binary mixture perception is affected byconcentration of odor components. Behav Neurosci 2007;121:1132–6.
[11] Laing DG, Francis GW. The capacity of humans to identify odors in mixtures.Physiol Behav 1989;46:809–14.
[12] Laska M, Hudson R. Discriminating parts from the whole: determinants ofodor mixture perception in squirrel monkeys, Saimiri sciureus. J Comp Physiol A1993;173:249–56.
[13] Wiltrout C, Dogras S, Linster C. Configurational and nonconfigurational interac-tions between odorants in binary mixtures. Behav Neurosci 2003;117:236–45.
[14] Linster C, Cleland TA. Configurational and elemental odor mixture perception canarise from local inhibition. J Comp Neurosci 2004;16:39–47.
[15] Derby C, Huston M, Livermore B, Lynn W. Generalization among related complexodorant mixtures and their components: analysis of olfactory perception in thespiny lobster. Physiol Behav 1996;60:87–95.
[16] Smith BH. The role of attention in learning about odorants. Biol Bull1996;191:76–83.
[17] Jinks A, Laing DG. A limit in the processing of components in odour mixtures.Perception 1999;28:395–404.
[18] Valentincic T, Kralj J, Stenovec M, Koce A, Caprio J. The behavioral detection ofbinary mixtures of amino acids and their individual components by catfish. J ExpBiol 2000;203:3307–17.
[19] Rescorla RA, Wagner AR. A theory of Pavlovian conditioning: Variation in theeffectiveness of reinforcement and nonreinforcement. In: Black AH, Prokasky WF,editors. Classical conditioning: current research and theory, vol. 2. New York:Appleton-Century-Crofts; 1972. p. 64–99.
[20] Kay LM, Crk T, Thorngate J. A redefinition of odor mixture quality. Behav Neurosci2005;119:726–33.
[21] Olsson MJ. An integrated model of intensity and quality of odor mixtures. Ann NYAcad Sci 1998;855:837–40.
[23] Jinks A, Laing DG. The analysis of odor mixtures by humans: evidence for aconfigurational process. Physiol Behav 2001;72:51–63.
[24] Thomas-Danguin T, Le Berre E, Barkat S, Coureaud G, Sicard G. Evidence for odorblending in odorant mixtures. Chem Senses 2007;32:A19.
[25] Wilson DA, Stevenson RJ. Olfactory perceptual learning: the critical role ofmemory in odor discrimination. Neurosci Biobehav Rev 2003;27:307–28.
[26] Johanson IB, Hall WB. Appetitive learning in 1-day-old rat pups. Science1979;205:41–421.
[27] Wilson DA, Sullivan RM. Neurobiology of associative learning in the neonate: earlyolfactory learning. Behav Neural Biol 1994;61:1–18.
[28] ZarrowMX, Denenberg VH, Anderson CO. Rabbit: frequency of suckling in the pup.Science 1965;150:1835–6.
[29] Hudson R, Distel H. Nipple location by newborn rabbits: behavioural evidence forpheromonal guidance. Behaviour 1983;85:260–75.
[30] Keil W, von Stralendorff F, Hudson R. A behavioral bioassay for analysis of rabbitnipple-search pheromone. Physiol Behav 1990;47:525–9.
[31] Coureaud G, Schaal B. Attraction of newborn rabbits to abdominal odors of adultconspecifics differing in sex and physiological state. Dev Psychobiol2000;36:271–81.
[32] Coureaud G, Schaal B, Langlois D, Perrier G. Responsiveness of newborn rabbits tosurface odour cues from females differing in lactational state and to milk. AnimBehav 2001;61:153–62.
[33] Coureaud G, [Olfactory regulation of sucking in newborn rabbit: Ethological andchemical characterization of a pheromonal signal]. PhD Thesis, University Paris 13,2001.
[34] Coureaud G, Moncomble AS, Montigny D, Dewas M, Perrier G, Schaal B.A pheromone that rapidly promotes learning in the newborn. Curr Biol2006;16:1956–61.
[35] Hudson R. Do newborn rabbits learn the odor stimuli releasing nipple-searchbehavior? Dev Psychobiol 1985;18:575–85.
[36] Bilkó A, Hudson R, Distel H, Altbäcker V. Integrity of the olfactory bulb is notnecessary for true odour discrimination. Adv Ethol 1997;32:133.
[37] Bell GA, Laing DG, Panhuber H. Odour mixture suppression: evidence for aperipheral mechanism in human and rat. Brain Res 1987;426:8–18.
[38] Mandairon N, Stack C, Linster C. Olfactory enrichment improves the recognition ofindividual components in mixtures. Physiol Behav 2006;89:379–84.
[39] Coureaud G, Langlois D, Sicard G, Schaal B. Newborn rabbit responsiveness to themammary pheromone is concentration-dependent. Chem Senses 2004;294:341–50.
[40] Le Berre E, Thomas-Danguin T, Béno N, Coureaud G, Etiévant P, Prescott J.Perceptual processing strategy and exposure influence the perception of odormixtures. Chem Senses 2008;33:193–9.
[41] Montigny D, Coureaud G, Schaal B. Newborn rabbit response to the mammarypheromone: from automatism to prandial control. Physiol Behav 2006;89:742–9.
[42] Estes WK. Toward a statistical theory of learning. Psychol Rev 1950;57:94–107.[43] Pearce JM. Similarity and discrimination: a selective review and a connectionist
model. Psychol Rev 1994;101:587–607.[44] Drewett RF, Kendrick KM, Sanders DJ, Trew AM. A quantitative analysis of the
feeding behavior of suckling rabbits. Dev Psychobiol 1982;15:25–32.[45] Coureaud G, Schaal B, Coudert P, Rideaud P, Fortun-Lamothe L, Hudson R, et al.
Immediate postnatal sucking in the rabbit: its influence on pup survival andgrowth. Reprod Nutr Dev 2000;40:19–32.
[46] Livermore A, Hutson M, Ngo V, Hadjisimos R, Derby CD. Elemental and configurallearning and the perception of odorant mixtures by the spiny lobster Panulirusargus. Physiol Behav 1997;62:169–74.
[47] Livermore A, Laing DG. The influence of odor type on the discrimination andidentification of odorants in multicomponent odor mixtures. Physiol Behav1998;65:311–20.
[48] Cain WS, Drexler M. Scope and evaluation of odor counteraction and masking. In:Cain WS, editor. Annals of the New York Academy of Sciences. Odors: evaluation,utilization, and control, vol. 237. New York: New York Academy of Sciences; 1974.p. 427–39.
[49] Sokolic L, Laing DG, McGregor IS. Asymmetric suppression of components inbinary aldehyde mixtures: behavioral studies in the laboratory rat. Chem Senses2007;32:191–9.
[50] Sicard G, Holley A. Receptor cell responses to odorants: similarities and differencesamong odorants. Brain Res 1984;292:283–96.
[51] Kay L, Lowry C, Jacobs H. Receptor contributions to configural and elemental odormixture perception. Behav Neurosci 2003;117:1108–14.
[52] Deisig N, Giurfa M, Lachnit H, Sandoz JC. Neural representation of olfactorymixtures in the honeybee antennal lobe. Eur J Neurosci 2006;24:1161–74.
[53] Lin DY, Shea SD, Katz LC. Representation of natural stimuli in the rodent mainolfactory bulb. Neuron 2006;50:937–49.
[54] Zou Z, Buck LB. Combinatorial effects of odorant mixes in olfactory cortex. Science2006;311:1477–81.
[55] Gottfried JA, Winston JS, Dolan RJ. Dissociable codes of odor quality and odorantstructure in human piriform cortex. Neuron 2006;49:467–79.
[56] Dreumont-Boudreau SE, Dingle RN, Alcolado GM, LoLordo VM. An olfactorybiconditional discrimination in the mouse. Physiol Behav 2006;87:634–40.
