Discrimination of individual odours by hamsters (Mesocricetus auratus) varies with the location of those odours

ANIMAL BEHAVIOUR, 2002, 64, 269–281 doi:10.1006/anbe.2002.3053, available online at http://www.idealibrary.com on

Discrimination of individual odours by hamsters (Mesocricetus auratus) varies with the location of those odours DARRYL J. MAYEAUX & ROBERT E. JOHNSTON

Department of Psychology, Cornell University (Received 19 July 2001; initial acceptance 27 September 2001; final acceptance 22 January 2002; MS. number: A9121)

When animals perceive social signals, information about the identity and the location of the signaller can be important determinants of a response by the perceiver. An unfamiliar individual often elicits a greater response than does a familiar individual. Similarly, a signal from an unexpected location may elicit a greater response than if it came from an expected location. For example, in field experiments on vocal communication in birds, an unexpected location has been many metres away from the expected one. Laboratory experiments on the responses of voles and hamsters to scent overmarks and on the habituation of hamsters to social scents suggest that much smaller differences in the location of odours may be salient. To explore this further, we examined the influence of changes in spatial location of familiar and novel male scents on responses of female golden hamsters, M. auratus. The spatial changes were about 9 cm, less than three-fourths of the body length of our subjects. The decline in females’ investigation of the same male’s flank odour across four habituation trials was not affected by changing the location of the odour. During test trials, however, changes in location did influence the results. The expected higher level of investigation of a novel scent versus that of a familiar one was observed primarily when the novel scent occupied a novel location. Such increases in investigation were usually not seen when only one of these variables was changed (individual or location). Thus, small changes in spatial location influence the salience of conspecific odours in this species. 

2002 The Association for the Study of Animal Behaviour. Published by Elsevier Science Ltd. All rights reserved.

In the course of every-day social life, individuals of most species interact repeatedly with conspecifics that live in the same area. Repeated interactions between neighbours no doubt change the quantity and quality of the social interactions that occur. Hostilities during initial encounters may decline over time and eventually become nonaggressive; even the attention paid to another individual may wane substantially. For example, field experiments indicate that individuals of many species tend to habituate to the presence of familiar conspecifics, or their cues, as indicated by declining aggressive or investigatory responses to repeated presentations of the same songs or calls (birds: Lambrechts & Dhont 1995; amphibians: Bee & Gerhardt 2001; mammals: Fischer et al. 2001). In contrast, a signal from a novel individual, or a familiar signal from a novel location, usually results in an increase in responses. For example, after habituation to the playback of an individual’s alarm calls indicating a predator, or contact calls indicating separation from group members, animals typically increase their responses to the same type of call from a different Correspondence: D. J. Mayeaux, Department of Psychology, Uris Hall, Cornell University, Ithaca, NY 14853, U.S.A. (email: [email protected]). 0003–3472/02/$35.00/0



individual (Richardson’s ground squirrel, Spermophilus richardsonii, Hare 1998; rhesus monkey, Macaca mulatta, Rendall et al. 1996). In addition, after a simple change in the location of either a familiar individual (three-spined stickleback, Gasterosteus aculeatus, Peeke & Veno 1973), a speaker producing song playbacks (great tit, Parus major, Krebs 1976), or a branch containing a scent mark (ringtailed lemur, Lemur catta, Mertl 1977), animals increase their aggressive responses to or investigation of these familiar stimuli in the new spatial context. The magnitude of the change in location sufficient for such reinvigorated responding typically has ranged between a few body lengths of the subject and 50 m. These findings indicate that both individual identity and spatial context influence responses to social signals in nature. Some issues remain unresolved. First, how large of a change in location is necessary to elicit an increase in responses to a familiar stimulus? After ringtailed lemurs habituated to a territorial scent mark on a branch, they increased investigation of that same odour when the branch was moved 60 cm (Mertl 1977), a distance only about 1.5 times the body length of an adult lemur (Napier & Napier 1985). It seems likely that if animals deposit territorial scent marks in traditionally scent-marked

269 2002 The Association for the Study of Animal Behaviour. Published by Elsevier Science Ltd. All rights reserved.

270

ANIMAL BEHAVIOUR, 64, 2

locations, the location of new marks with respect to previous ones may vary, even if only by a few centimetres. These small variations in location of a social signal may be salient. Second, when both the individual providing the cues and the location of those cues are novel, are the effects of these two types of novelty additive, indicating something more salient than either type of novelty alone? Habituation–discrimination paradigms are used widely in the laboratory to examine discrimination between odours of individuals (Halpin 1986; Johnston 1993). In previous experiments we used a habituation method to examine the effect of brain lesions on discrimination of odours by female golden hamsters, Mesocricetus auratus (unpublished data). Our results suggest that the response to a novel individual’s scent depended on the location of this scent. The habituation–discrimination procedure we used involved repeated, brief presentations (four trials, 3 min each) of the flank odour of one male and a subsequent test of discrimination between that male’s odour and the flank odour of another male (one trial, 3 min). Females generally increased their investigation of an odour from a novel male, thus revealing which stimuli are meaningfully different to them (Nelson & Marler 1989). We did not train subjects to respond to differences between stimuli that we designated as important. Because our experiments indicated that hamsters might be responding to both scent identity and to scent location, we explored this phenomenon in more detail. We reasoned that this method could provide an interesting laboratory model of complex memories composed of social identity and of spatial location of odours. Golden hamsters provide an excellent model for examining how small changes in the location of social cues (odours in scent marks) affect responses to familiar and novel scents. Like other rodents, hamsters make extensive use of olfactory cues in regulating social interactions (Johnston 1990). Female hamsters deposit vaginal secretions, and both sexes deposit oily secretions from sebaceous glands on their flanks as they vigorously rub the length of their bodies along vertical surfaces. In laboratory enclosures, individuals deposit flank marks around their nest areas that they defend for their exclusive use (Johnston 1975, 1985). When two hamsters meet they investigate each other intensively (Johnston 1985), appearing to concentrate for some of that time around the other’s flank gland (personal observation). Using these odours, hamsters can discriminate, based on body odours, conspecifics from heterospecifics, males from females, kin from nonkin, and one individual from another (Johnston 1985, 1993; Mateo & Johnston 2000). Hamsters distinguish individual odours from at least five different scent sources on the body: urine, faeces, and secretions from the vaginas of females and from the ear and flank glands of both sexes (Johnston et al. 1993). We examined how a small change in the location of an odour (two-thirds to three-fourths of a hamster’s body length, or about 9 cm) affected investigation and discrimination of those odours in a habituation– discrimination task. This paradigm allowed standardization of odour presentation across trials and across

subjects. We addressed two fundamental questions. First, how does varying the location of flank odour of one male, termed ‘familiar’ by virtue of its repeated presentation, affect investigation over a series of trials? Second, after females have habituated to one male’s odour, how does varying the location of odours from ‘novel’ and ‘familiar’ males on the test trial influence the investigation of these odours? Specifically, does novelty of the individual and of spatial location of the odour combine to produce greater investigation than just one type of change? GENERAL METHODS

Subjects and Scent Donors All animals used in these experiments were adult golden hamsters at least 2 months of age that were born in our laboratory colony. The colony was descended from Charles River (Wilmington, Massachusetts, U.S.A.) stock and occasionally infused with new stock from the supplier. Each adult hamster lived individually in a clear plastic cage (303516 cm) with a solid bottom covered by Sani-Chips bedding (P. J. Murphy Forest Products, Montville, New Jersey, U.S.A.) and with a stainless steel wire top. Pups spent the first 4 weeks of life with their mother and were then separated and housed individually. Hamsters had continuous access to water and food (Agway Prolab); cages were cleaned every 2 weeks. Each colony room contained male and female hamsters and was on a 10:14 h dark:light cycle with lights off at 0800 hours Eastern Standard Time, the start of the active period for these nocturnal animals. Constraints on the number of hamsters available required that some individuals be used in one, two or all three experiments described here; there was an interval of at least 6 weeks between use in different experiments. Subjects were 109 sexually nai¨ve females. Sixty-seven females were used in only one experiment, 24 were used in two experiments, and 18 were used in all three experiments. Fifty-four females had no other experimental testing experience before being used here. Fifty-five females had experimental exposure to odours of another hamster, 39 of these through direct interaction with that individual. Only one female was tested previously in a habituation procedure. Use of hamsters in multiple behavioural experiments, when possible, is typical in our laboratory to promote varied, and presumably more realistic, experiences with social odours. Flank scent donors were 61 males who had not had a sexual or aggressive encounter for at least 1 month before being used. Forty-one were used in only one experiment, 11 were used in two experiments, and 9 were used in all three. Donors were not related within two generations to the subject who encountered their odour.

Procedure Testing occurred during the early dark phase in a room that was separate from the colony room and dimly lit by

MAYEAUX & JOHNSTON: ODOUR DISCRIMINATION AND SPACE

Habituation trial

Zero

One

Two

Three

1 2 3 4 Figure 1. Sequential pattern of odour locations during habituation trials in experiment 1. Diagonal lines represent location of odour on each trial; open areas were clean. We presented the patterns as shown above to approximately half of the subjects and as the mirror image of those patterns to the other half.

two 40-W desk lamps. Female hamsters show sexual receptivity, behavioural oestrus, every fourth active period (dark phase of the light cycle). We tested them during the active period before oestrus, pro-oestrus. We transported female hamsters in their living cages to the testing room; they remained in their cage during testing. We also transported male scent donors to the same room; their cages were approximately 3 m away from those of the females. For odour stimuli, we simulated a male’s flank mark by gently grasping a male scent donor and rubbing his flank gland area against a glass plate (7.617.8 cm) in 10 elliptical motions. Approximately 1 min later, we placed the scented plate on the floor of the female subject’s cage with the longer dimension of the plate parallel to and centred near the front wall of the cage. This trial lasted 3 min, after which we removed the plate. There were four of these habituation trials, with 15 min between each. Subsequently, on the test trial (also 3 min long) we presented the flank odour of a second, ‘novel’, male. In some experiments we presented the novel male’s scent alone; in others, we presented the novel and familiar males’ scents simultaneously on the test trial. The donor of the novel odour for one subject was the donor of the familiar odour for another subject tested concurrently. Thus, on the test trial, the donors of novel and familiar odours had been similarly handled and equally used. Throughout this paper, ‘familiar’ denotes the odour of a male donor that is presented repeatedly to a female subject, and ‘novel’ indicates the odour of another male presented to her only once. Female subjects had not interacted with either of the scent donors. We wore plastic gloves to minimize transfer of human scent to plates and used each scented plate only once. An observer who was unaware of the position of the stimulus on the plate and of the hypothesis being tested recorded the duration of investigation during each 3-min trial on a laptop computer using The Observer 3.0 (Noldus Information Technology, Sterling, Virginia, U.S.A.). The criterion for investigation was that the subject’s nose was less than 1 cm directly above the glass plate. This included sniffing and occasional licking of the plate but excluded gnawing on, standing on, or pushing it with the nose or head.

EXPERIMENT 1 When tested in a habituation–discrimination paradigm, hamsters discriminate the flank odours of different individuals (Johnston 1993; Heth et al. 1999; Petrulis et al. 2000). Methodological details, such as consistency or variation in location of odours across trials have differed among studies. One goal of this experiment was to change the location of a male’s scent across successive presentations during the habituation trials and measure its impact on a female’s investigation of that odour. A second goal was to determine whether a female’s response to a novel scent differed depending on its location (i.e. in a familiar or novel location).

Method Subjects and scent donors Fifty-six females were subjects, each used once in experiment 1. Thirty-three males were donors of flank scent, 12 males were used for only one session of five trials, 19 males were used for two sessions, and two males were used for three sessions. Although some donors were used several times, each female had a unique pair of scent donors. Males were used as donors in only one series of trials per day.

Design During the habituation phase, we varied the location of the familiar scent across habituation trials following four patterns (Fig. 1) to evaluate the effect on responses to the same odour. In the first of these patterns (Fig. 1, pattern Zero) the familiar scent was in the centre of the glass plate on all four trials. In the remaining three patterns (Fig. 1, patterns One, Two and Three) the scent was twice on the left and the right sides of the plate. The names of the four patterns indicate the number of times that the location of the odour changed. Half the surface of the plate on each of these trials was unscented so that we could obtain a measure of investigation of this unscented half of the plate. During the test (fifth) trial, we varied the location of the test scents with respect to the location of the scent on the previous (fourth) habituation trial (top of Fig. 2) to

271

ANIMAL BEHAVIOUR, 64, 2

(a)

Trial

(b)

(c)

(d)

4

Test

20 Investigation time (s)

272

*

10

0 Test trial Unscented

Familiar

Novel

Figure 2. Top: The four test conditions of experiment 1. Each condition consisted of five trials. In the first three conditions (a–c), the familiar odour on trials 1–4 was presented twice on the left and twice on the right half of the plate. In the fourth condition (d) the familiar odour on trials 1–4 was in the centre. The conditions differed in the location of the familiar and novel odours on the test trial (trial 5) relative to the location of the familiar odour on the last habituation trial (trial 4). We presented the patterns as shown above to approximately half of the subjects in each condition and as the mirror image of those patterns to the other half. Bottom: Mean±SE investigation time (s) on the test trial in experiment 1. *Columns differed, P≤0.05.

evaluate the effect of these manipulations on discrimination of the test odours. In the first two conditions (Fig. 2a, b) familiar and novel scents were present on the test trial. In the first condition (Fig. 2a, N=15) the novel scent was on the side of the plate that previously contained the familiar one, and the familiar scent was on the side of the plate that was previously unscented. In the second condition (Fig. 2b, N=14) the pattern of scents was opposite that of the first: the novel scent was on the previously unscented side and the familiar scent remained in the same place as on the last habituation trial. These two conditions evaluated the hypothesis that preferential investigation of the novel scent on the test trial occurred only when it was in a previously unscented location. In the third condition (Fig. 2c, N=13) only the familiar scent was present, and it was on both left and right sides of the plate. This condition examined the hypothesis that the hamsters might investigate the familiar odour more when it was in a novel location. Approximately one-third of the subjects in each of the first three conditions received each of the habituation patterns One, Two and Three (see Fig. 1 for these patterns). In the fourth condition (Fig. 2d, N=14), unlike the other three, the familiar scent was in the centre of the plate during habituation trials; on the test trial the familiar and novel scents were on opposite halves of the plate. This evaluated the hypothesis that hamsters would investigate a novel scent more than a familiar one when there was no spatial bias during habituation or test trials.

Analysis During the four habituation trials the familiar scent changed locations either zero, one, two or three times. We compared duration of investigation across trials for these four groups to determine whether the patterns of presentation affected investigation of the scented and clean halves of the plate. When the familiar scent was in the centre, investigation of the extreme left and right quarters of the plate was summed and this number used for investigation of the clean ‘half’. We used a 4 (habituation pattern)4 (trial)2 (side of plate) mixed design analysis of variance (ANOVA). For the test trial of each of the four test conditions, we compared the duration of investigation of the two sides of the plate with repeated measures ANOVA. In all statistical tests, P≤0.05 indicated significance.

Results During the habituation trials, females investigated the side of the plate containing the male flank scent more than the clean side, indicating that they detected and were interested in the odour (compare Fig. 3a and b; ANOVA: side of plate: F1,52 =227.6, P<0.001). Investigation declined across trials, and this decline was more dramatic on the scented side compared with the clean side, indicating the hamsters habituated to the odour and not just to the presence of the glass plate in their cage (trial*side of plate interaction: F3,156 =62.13, P<0.001).

MAYEAUX & JOHNSTON: ODOUR DISCRIMINATION AND SPACE

Investigation time (s)

(b)

(a)

Zero One Two Three

30

20

10

0

1

2

3

4

1

2

3

4

Habituation trial Figure 3. Mean±SE investigation time (s) of the familiar odour (a) and the clean side of the plate (b) during four habituation trials in experiment 1. The four groups are named for the number of times the scented and clean sides of the plate changed location across the four trials (see Fig. 1). See text for statistics.

Regardless of the number of times the odour changed location, this pattern of decline in investigation did not differ between the scented and clean sides of the plate (trial*side of plate*habituation pattern interaction: F9,156 =0.32, P=0.32). Thus, varying the location of the familiar odour did not affect investigation of it or habituation to it. On the test trial, however, discrimination depended on an odour’s novelty as well as its location. If the novel scent was on the same side that the familiar scent had occupied on the last habituation trial, females’ investigation of the novel and familiar scents did not differ (Fig. 2a; ANOVA: side of plate: F1,14 =0.87, P=0.37). If the novel scent was on the side of the plate that was previously clean, however, females investigated it more than the familiar scent (Fig. 2b; F1,13 =5.59, P=0.03). When the familiar odour was on both sides of the plate, females investigated the familiar odour on the previously clean side and the familiar odour on the previously scented side for similar amounts of time (Fig. 2c; F1,12 =0.43, P=0.52). This indicates that a relatively novel location of an odour, by itself, did not result in increased investigation of this odour. Finally, when both the familiar and the novel males’ odours were both in a relatively novel position on the test trial, females did not differentiate between the familiar and novel scents (Fig. 2d; F1,13 =0.61, P=0.45).

Discussion Although changing the location of a familiar scent across habituation trials did not influence the duration of investigation, changing the location of the novel scent on the test trial did. Female hamsters investigated a novel scent more than a familiar one only when the location of the novel odour on the test trial was clean on the previous trial (Fig. 2b). This suggests that the differential investigation of the two odours was based on a spatial-

olfactory discrimination; it depended on an odour being both novel and in a previously clean location. To our surprise, presenting the familiar scent in the centre of the glass plate during habituation trials, thus rendering novel and familiar odours spatially equivalent on the test trial, did not produce differential investigation of the two scents (Fig. 2d). Some previous papers report using a habituation– discrimination paradigm with novel and familiar odours present simultaneously, as we did, to assess discrimination between odours of different individuals. The methods used in several of these studies avoided the potential spatial confound during the test trial by placing the novel and familiar odours equidistant from and not overlapping the location of the familiar odour during habituation trials (Mongolian gerbil, Meriones unguiculatus, Halpin 1974; chipmunk, Tamias striatus, Keevin et al. 1981; prairie vole, Microtus ochrogaster, Newman & Halpin 1988; golden hamster, Tang-Martinez et al. 1993; blind subterranean mole-rat, Spalax ehrenbergi, Todrank & Heth 1996). One study avoided a spatial confound by randomizing the location of odours across trials (cavie, Cavia aperea, Martin & Beauchamp 1982), and a second study avoided the confound by presenting the familiar odour in two locations and then replacing one of them with the novel odour (domestic rat, Rattus norvegicus, Krames 1970). In contrast to these studies, which controlled for the spatial confound, a recent set of studies with female golden hamsters used a method similar to ours except that novel scent in those studies was always in a location that was unscented on the previous trial (Petrulis et al. 1998, 1999a, b, 2000; Petrulis & Johnston 1999; A. Petrulis, personal communication). This method is similar to the one condition in our experiment (Fig. 2b) that produced discrimination of novel and familiar scents. Such a method appears to yield an increase in investigation because of the combination of a novel scent with its placement in a relatively novel position.

273

ANIMAL BEHAVIOUR, 64, 2

(a)

Trial

(b)

(c)

(d)

4

Test

20 Investigation time (s)

274

*

*

10

0 Test trial Unscented

Familiar

Novel

Figure 4. Top: The four test conditions of experiment 2. Each condition consisted of five trials. In the first three conditions (a–c), the familiar odour on trials 1–4 was on the same half of the plate. In the fourth condition (d), the familiar odour on trials 1–4 was on both halves of the plate. The conditions differed in the location of the familiar and novel odours on the test trial (trial 5) relative to the location of the familiar odour on the last habituation trial (trial 4). We presented the patterns as shown above to approximately half of the subjects in each condition and as the mirror image of those patterns to the other half. Bottom: Mean±SE investigation time (s) on the test trial in experiment 2. *Columns differed, P≤0.05.

EXPERIMENT 2 Another feature that differed between most of the conditions in experiment 1 and the experiments by Petrulis and colleagues (Petrulis et al. 1998, 1999a, b, 2000; Petrulis & Johnston 1999; A. Petrulis, personal communication) is that in their experiments the odour presented on the habituation trials was always on the same side of the glass plate, whereas in experiment 1 it was usually moved from side to side (Fig. 1). The salience of location could, therefore, be greater in experiment 1 than in previous studies. Therefore, in this experiment we kept the location of the familiar scent consistent across habituation trials and varied the location of the scents on the test trial. This design allowed us to examine whether consistency of location of the familiar odour affected investigation of the novel or familiar odours on the test trial.

Method Subjects and scent donors Fifty-six females were subjects; each was used only once in experiment 2. Twenty-nine of these females were used in experiment 1 (minimum of 7 weeks between use in the two experiments). Twenty-nine males were flank scent donors, 13 males were used for only one session of five trials, eight males were used for two sessions, five males were used for three sessions, and three males were used for four sessions. Although some donors were used several times, females had unique pairs of scent donors.

Males were used as donors in only one series of trials per day.

Design In this experiment we kept the familiar scent in the same location across the four habituation trials. There were four test conditions, three with familiar scent on only one side of the plate and one with familiar scent on both sides. We varied the location of the novel and familiar scents on the test trial (top of Fig. 4) to evaluate the effect of position of these odours on discrimination of the scents of two individuals. In the first condition (Fig. 4a, N=14) the novel scent was on the side that previously contained familiar scent on all four trials. In the second condition (Fig. 4b, N=14) the novel scent was on the side that was previously clean on all four trials. In the third condition (Fig. 4c, N=14) only familiar scent was present on both sides of the plate on the test trial. In the fourth condition (Fig. 4d, N=14) familiar odour was on both sides of the plate during all four habituation trials, and on the test trial both familiar and novel scents were present. This arrangement in the fourth condition rendered odours on the test trial spatially equivalent because both were on a portion of the plate that had previously contained familiar odour.

Analysis For habituation trials we compared duration of investigation on both sides of the plate and across trials with a 2

MAYEAUX & JOHNSTON: ODOUR DISCRIMINATION AND SPACE

(side of plate)4 (trial) repeated measures ANOVA. For the test trial of each condition we assessed discrimination of the two sides of the plate with a repeated measures ANOVA. In all statistical tests P≤0.05 indicated significance.

Results

EXPERIMENT 3 The results of experiments 1 and 2 indicated that discrimination of males’ flank odours by female hamsters was very sensitive to the sequence of locations of these odours across the last two trials of the habituation– discrimination task. To determine whether this effect was dependent on the presence of two odours during the test trial, or would also occur using a single odour on the test trial we conducted the following experiment.

As in experiment 1 females tested in the first three conditions (Fig. 4a–c) decreased their investigation of the familiar odour across the four habituation trials (all Fs3,39d5.71, all Psc0.002). To save space we do not present those data here, although we present them as part of a larger comparison in experiment 3. During habituation trials of the fourth condition (familiar scent on both sides of the plate), females’ investigation of the two scented sides did not differ (ANOVA: side of plate: F1,13 =1.21, P=0.29), but total investigation did decline across trials (F3,39 =20.78, P<0.001). MeanSE investigation time(s) for the two sides of the plate were: trial 1, 12.71.8 and 13.8 1.6; trial 2, 5.71.4 and 6.41.0; trial 3, 5.21.0 and 6.61.5; trial 4, 4.70.9 and 6.01.3. On the test trial, the location of familiar and novel scents, relative to the last habituation trial, influenced investigation of these two odours. If the novel scent was on the same side that the familiar scent had occupied on the last habituation trial, the duration of investigation of novel and familiar scents did not differ (Fig. 4a; ANOVA: side of plate: F1,13 =1.08, P=0.32). If the novel scent was on the side of the plate that was previously unscented, however, females investigated it more than the familiar scent (Fig. 4b; F1,13 =6.03, P=0.03). When the familiar odour was on both sides of the plate, females did not investigate differently the familiar odour on the previously clean side as compared to the familiar odour on the previously scented side (Fig. 4c; F1,13 =1.17, P=0.29), suggesting that a change in location alone was not sufficient to result in increased investigation. When the spatial bias on the test trial was eliminated by presenting the familiar odour on both sides of the plate on habituation trials, females investigated the novel scent more than the familiar scent (Fig. 4d; F1,13 =24.41, P<0.01).

We again used a habituation–discrimination paradigm. As in experiment 2, the location of the familiar scent was the same across habituation trials. Unlike experiment 2, however, there was only one scent present on the test trial. There were four test conditions. On the test trial, two conditions used a novel scent and two used the familiar scent (top of Fig. 5). In the first condition (Fig. 5a, N=14) the novel odour was on the same side of the plate that the familiar odour occupied on the habituation trials. In the second condition (Fig. 5b, N=14) the novel odour occupied the side of the plate that was clean on the habituation trials. In the third condition (Fig. 5c, N=15) the familiar scent was on the same side of the plate that it occupied on the habituation trials. In the fourth condition (Fig. 5d, N=14) the familiar scent was on the opposite side of the plate as on the habituation trials.

Discussion

Analysis

These results are similar to those of experiment 1 and confirm that increased investigation of a novel male’s flank odour occurred when it was in a previously clean location. This increase tended not to occur when the novel male’s scent was in a location that previously contained another male’s scent. This effect of the familiar location occurred whether the familiar odour was in this location for one or two previous trials (experiment 1) or all four previous trials (experiment 2). In an interesting contrast, females investigated the novel scent more than the familiar one when both scents were in a location that previously contained the familiar scent (Fig. 4d). This suggests that in some situations a change in odour quality alone increases investigation.

For the test trial of each condition, we examined whether hamsters increased investigation of the odour as compared to investigation of the familiar odour on the last habituation trial, with repeated measures ANOVA. In an additional analysis across all three experiments, we pooled the data from experiments 2 and 3 for conditions in which the odour was on the same side of the plate in all four habituation trials (side biased, N=50) and compared them with the data from the three conditions of experiment 1 in which the familiar odour was on each side of the plate twice (side balanced, N=42). Because some females were used in more than one experiment, we included in this analysis only the data from the first experiment in which each female was used. We compared

Method Subjects and scent donors Fifty-seven females were subjects, each used only once in experiment 3. Thirteen were subjects in one of the previous experiments, and 18 were used in both experiments (minimum of 6 weeks between use in experiments 2 and 3). Twenty-eight males were flank scent donors, eight males were used for only one session of five trials, 13 males were used for two sessions, five males were used for three sessions, and two males were used for four sessions. Although some donors were used several times, females had unique pairs of scent donors. Males were used as donors in only one series of trials per day.

Design

275

ANIMAL BEHAVIOUR, 64, 2

(b)

(a)

Trial

(c)

(d)

4

Test

20 Investigation time (s)

276

*

10

0

4

Test

4

Test

4

Test

4

Test

Trial Unscented

Familiar

Novel

Figure 5. Top: The four test conditions of experiment 3. Each condition consisted of five trials. In all four conditions (a–d), the familiar odour on trials 1–4 was on the same half of the plate. The conditions differed in the familiarity and the location of the single odour present on the test trial (trial 5) relative to the last habituation trial (trial 4). We presented the patterns as shown above to approximately half of the subjects in each condition and as the mirror image of those patterns to the other half. Bottom: Mean±SE investigation time (s) on the last habituation (trial 4) and on the test trial in experiment 3. *Columns differed, P≤0.05.

duration of investigation for the side-biased and sidebalanced groups across the four habituation trials to determine whether the pattern of scent location affected the level of investigation of the scented and clean halves of the plate. We used a 2 (side biased or side balanced)4 (trial)2 (side of plate) mixed design ANOVA, followed by separate 2 (side balanced or side biased)4 (trial) mixed design ANOVAs for the scented and clean sides. Finally, for each trial, a between-groups ANOVA compared investigation by subjects receiving side-balanced and side-biased series. In all statistical tests P≤0.05 indicated significance.

Results If the novel scent was on the same side that the familiar scent occupied on the last habituation trial, the duration of investigation did not increase (Fig. 5a; ANOVA: trial 4 versus test: F1,13 =0.48, P=0.49). If the novel scent was on the side of the plate that was previously clean, however, females investigated it more than the familiar scent on the previous trial (Fig. 5b; F1,13 =13.04, P=0.003). If the familiar scent was present on the test trial, females did not increase their investigation of it, regardless of its location (previously scented side, Fig. 5c; F1,14 =0.0004, P=0.98; previously clean side, Fig. 5d; F1,13 =0.27, P=0.61). Analysis of the data from the habituation trials of all experiments indicated that placing the familiar scent consistently on one side of the plate (side biased) did not affect the decline in investigation across trials as com-

pared to changing the location of the familiar scent (twice on each side, side balanced, Fig. 6). Females investigated the side of the plate containing the male flank scent more than the clean side, indicating that they detected the odour (compare Fig. 6a and b; ANOVA: side of plate: F1,90 =329.26, P<0.001). Investigation declined across trials and this decline was more dramatic on the scented side of the plate compared with the clean side, indicating the hamsters habituated to the odour and not just to the presence of the glass plate in their cage (trial*side of plate interaction: F3,270 =75.31, P<0.001). Although there was a three-way interaction (trial * side of plate * balanced or biased: F3,270 =3.84, P=0.01), this resulted in part from an apparently spurious difference on trial 1 between the two groups in investigation of the scented side of the plate (Fig. 6; ANOVA: trial 1: F1,90 =4.37, P=0.039). On subsequent trials, investigation of the scented side of the plate did not differ between groups (trial 2: F1,90 =0.00, P=1.00; trial 3: F1,90 =2.58, P=0.11; trial 4: F1,90 =0.19, P=0.67). On the fourth habituation trial investigation of the clean side of the plate by the side-balanced group was greater than that by the side-biased group (Fig. 6b; trial 4: F1,90 =4.41, P=0.038).

Discussion Females increased their investigation of a novel scent only when it was in a previously clean location. This was consistent with the major pattern from experiments 1 and 2 in which there were two scents, novel and familiar,

MAYEAUX & JOHNSTON: ODOUR DISCRIMINATION AND SPACE

(b)

(a) 30

Investigation time (s)

Side balanced Side biased

20

10

0

1

2

3

4

1

2

3

4

Habituation trial Figure 6. Mean±SE investigation time (s) by females of the familiar odour (a) and the clean side of the plate (b) during four habituation trials in experiments 1, 2 and 3. Side balanced refers to the three series of experiment 1 in which the familiar odour was presented twice on each side of the plate while the other side was clean. Side biased refers to three conditions in experiment 2 and all four conditions in experiment 3 in which the familiar odour was on the same side of the plate on all habituation trials.

present during the test trial. Thus, the interaction of spatial- and olfactory-quality cues in producing an increase in investigation of the novel male’s odour described for paired-scent presentations in experiments 1 and 2 also applies to single-scent presentations. GENERAL DISCUSSION The three experiments presented here demonstrate that the relationship between the location of an odour during habituation trials and the location of familiar and novel odours on the test trial can affect the relative amount of investigation of these odours on the test trial. That is, the location of odours can affect the behaviours that indicate discrimination between the odours. Although this relationship has not been systematically investigated before now, our results are consistent with the results of similar habituation studies in which the novel odour was presented on the side of the plate that was clean on at least the last habituation trial (Petrulis et al. 1998, 1999a, b, 2000; Todrank et al. 1998, 1999; Heth et al. 1999; Petrulis & Johnston 1999; A. Petrulis, personal communication; J. Todrank, personal communication). Specifically, our experiments demonstrated that female hamsters generally did not increase their investigation of a novel individual’s scent when it was in the location that previously contained the familiar odour on the last habituation trial. One exception to this pattern occurred in a test condition in experiment 2 (Fig. 4d) in which familiar odour was on both sides of the plate during habituation trials. On the test trials of that condition, females did discriminate between the novel and familiar individuals’ scents, perhaps because the test trial of this condition differed from other conditions in these experiments by combining two features. First, both the

novel and familiar odours were present rather than just one odour. Second, both novel and familiar odours were in locations that contained odour on the habituation trials rather than one or both odours being located on a portion of the plate that was previously either partially or completely clean. The results in this experiment suggest that although a change in location adds to the probability of observing discrimination between odours of individuals, such a change is not always necessary. This conclusion is supported by experiments with other species (Singh et at. 1987; Brown et al. 1987; Brown 1988; Hunter & Murray 1989; Schellinck et al. 1992, 1995) and with hamsters (R. E. Johnston, unpublished data). We did not evaluate whether these same effects would occur if hamsters were tested with odours of other species or with odours from other sources, such as plant odours, food odours, or laboratory chemicals. Thus, we do not know whether the importance of odour location is a general phenomenon or whether it is specific to odours used for communication between individuals of this species. None the less, the results of these experiments indicate that the sequence of stimulus locations should be taken into account in similar laboratory studies with hamsters and other species and probably also in field experiments using similar methods. The effects that we observed, however, contrast with results of other studies on olfactory discrimination between odours of individuals, perhaps due to important methodological differences. First, many studies with golden hamsters have used male subjects and male donors (Johnston 1993), and it is possible that there are differences between the sexes in the salience of the location of conspecific odours. Second, species may differ in the degree to which the precise location of an odour is important. Although responses to odours

277

278

ANIMAL BEHAVIOUR, 64, 2

in a habituation–discrimination paradigm have been examined in other species, no systematic studies on the influence of odour location have been done with these species (ringtailed lemur, Mertl 1975; Harrington 1976; guinea pig, Cavia porcellus, Beauchamp & Wellington 1984; domestic rat, Brown et al. 1987; Brown 1988; Hunter & Murray 1989; Schellinck et al. 1992, 1995; kangaroo rat, Dipodomys merriami, Randall 1991; Dipodomys ingens, Murdock & Randall 2001). Third, only one other study reported using observers who were blind to the location and identity of the stimuli the subjects were investigating (Gouat et al. 1998). Although in our study the effect of location on investigation was statistically significant, these differences were small. In one case mean investigation of the novel scent mark was only 4 s longer than investigation of the familiar one (9.7 and 5.7 s, respectively, experiment 2; Fig. 4d, test trial). An observer’s expectancy of differences in particular directions could easily sway results (Rosenthal 1976). For that reason, data collectors in our experiments were unaware of the stimulus arrangement, the hypothesis under investigation, and the cumulative investigation time they recorded on the laptop computer for each trial. Interobserver agreement in recording duration of investigation in hamsters was 0.8 or higher (unpublished data), indicating that relatively clear and unambiguous criteria for recording investigation were established. The importance of ‘blind’ observers would be even greater for species in which interobserver agreement is less robust; this might occur because each bout of investigation is shorter or behavioural indications of investigation are less obvious than in golden hamsters. It is unclear why female hamsters did not always discriminate the novel from the familiar odour when the novel odour occupied a location that previously contained familiar scent (Figs 2a, d, 4a and 5a). In habituation–discrimination tasks, an increase in investigation of a novel stimulus implies the subjects perceived the two stimuli as different. The lack of a difference in investigation of the familiar and the novel stimulus could imply either that the animal did not perceive a difference (Heth et al. 1999) or that the difference was not salient enough to prompt differential investigation. It is difficult to interpret a lack of difference in investigation of two stimuli; it is also difficult to demonstrate experimentally that animals perceived two stimuli as similar (Gamboa et al. 1991; Stoddard 1996). Accordingly, we are reluctant to conclude that hamsters in our study perceived novel and familiar odours as similar in some spatial arrangements (Figs 2a, 4a and 5a) and as different in others (Figs 2b, 4b and 5b). First, lack of discrimination could reflect a lack of motivation of the subjects to investigate odours for a reason unrelated to stimulus similarity or difference. If we had trained subjects to discriminate odours (Schellinck et al. 1991; Gheusi et al. 1997), they probably would have discriminated between the two individuals’ odours regardless of the spatial arrangement. Second, other experiments in our laboratory indicate that the relationship between odours of individuals and spatial location of the odours that was shown here can vary with the interval between odour presentations.

Although in experiment 1 female hamsters did not show discrimination when the novel odour was on the side of the plate that contained familiar odour 15 min earlier (Fig. 2a), they did discriminate novel from familiar when we reduced the intertrial interval to 20 s (unpublished data). Thus, it is possible that in some situations hamsters could recognize the difference between individual odours but not show this discrimination by differences in duration of investigation (Gamboa et al. 1991). Caution is necessary when interpreting results indicating no discrimination between stimuli (Stoddard 1996). Changes in methodology might facilitate odour discrimination in similar habituation tasks. For example, our subjects had not interacted with the scent donors. If subjects were familiar with the donors, they may have discriminated the odours of different donors in a wider range of test conditions. When hamsters have the opportunity to interact, they appear to form complex multiodour representations of other individuals. Evidence for this conclusion, for example, comes from male hamsters that, after habituation to the vaginal secretions of one female, investigated the flank odour of a second female more than the flank odour from the donor of the vaginal secretions (Johnston & Jernigan 1994; Johnston & Bullock 2001). Males made such a discrimination only if they had interacted with both donor females on previous days and presumably had made associations between several odours of one individual. If the males had relied simply on common chemical components in different sources of odour from one individual, they should have generalized between different odour sources whether or not they knew the donors. Thus, hamsters appear to have multicomponent representations of individuals and to recognize individuals as independent entities rather than merely discriminate between single sources of odours that differ between individuals. How odour location may affect this recognition ability is not known, but presumably it should be independent of spatial context. Why should hamsters pay attention to and remember the location of the scents of individuals? One reason might be because the relative location of each individual’s scent allows evaluation of these individuals by third parties. When hamsters or meadow voles encounter scent overmarks (the scent mark of one individual partially overlapping that of a second individual), they later respond differentially to these two individuals or to their whole-body odour (as found, for example, in bedding material; Johnston et al. 1994, 1995; Wilcox & Johnston 1995; Johnston & Bhorade 1998; Ferkin et al. 1999; Johnston & Bullock 2001). In one type of experiment, male hamsters and female voles, after being exposed to overmarks of opposite-sexed conspecifics, later preferred the top-scent to the bottom-scent individual (Johnston et al. 1997a, b; R. E. Johnston, R. Wilcox & C. Lee, unpublished data). The authors suggest that these preferences reflect mate preferences, and, if so, it may be because the ability of an individual to keep its scent on top of the rival’s scent most of the time may reflect the relative quality of the two individuals (Johnston 1999). It is not fully understood how hamsters

MAYEAUX & JOHNSTON: ODOUR DISCRIMINATION AND SPACE

or voles can determine which individual’s scent is on top, but they may do so partially by analysis of the spatial configuration of scent marks (Johnston & Bhorade 1998; Cohen et al. 2001). That is, if they obtain a detailed representation or ‘picture’ of the precise layouts of two individual marks, they may be able to determine which is on top by determining which one is interrupted and which one is continuous. In order to obtain such a picture they would have to remember the location and time of encounter of an odour as they move around in the environment and investigate overmarks. This ability may underlie the results in the present experiments; females were apparently paying close attention to which individual’s odour was in which location, and remembered these locations over both a very short timescale (seconds, as they investigated) and a longer one (15 min, between two trials). It is not obvious whether the sensitivity to small-scale changes in scent locations that we found in our experiments are related to the interpretation of scent marks of others in an individual’s home range in nature. Presumably golden hamsters have representations of particular individuals’ odours linked with particular locations in the environment (as the patterns of activity and scent marking by dwarf hamsters, Phodopus campbelli, suggests, Wynne-Edwards et al. 1992), but, obviously, the spatial scale in the present experiments is smaller than in the natural environment (Murphy 1977). Observations from mammalian and avian species suggest that locationspecific representations of individuals is widespread (Harrington 1974; Stoddard 1996), but there have been few experiments that test specific hypotheses about this type of knowledge. It is clear that the location of social signals affects the amount and intensity of responses to familiar and unfamiliar signals, as shown by field and laboratory experiments with a range of species (Peeke & Veno 1973; Krebs 1976; Mertl 1977; Sliwa & Richardson 1998). In some species smelling an odour or hearing a song in a new location may indicate a situation of instability, such as a change in territorial boundaries. If such a change constitutes a threat, a resident must allocate some energetic resources to countering the threat, resources that it would otherwise conserve if the location of the signals remained constant. Thus, identifying signals of particular individuals with particular locations may make a large contribution to maintaining stable patterns of space use without frequent and costlyaggressive interactions. Acknowledgments This work was supported by NIH grants T32MN19389 and R01MH58001. J. M. Mateo provided assistance and stimulating discussion at many stages of this project. S. Zornetzer, P. Agrawal, O. Fitch, W. S. Lai, R. Sugarman and T. Winkfield assisted with data collection. D. Leger and two anonymous referees provided valuable comments that improved this manuscript. The research presented here was evaluated and approved by the Institutional Animal Care and Use Committee of Cornell University (Protocol No. 93-120).

References Beauchamp, G. K. & Wellington, J. L. 1984. Habituation to individual odors occurs following brief, widely-spaced presentations. Physiology & Behavior, 32, 511–514. Bee, M. A. & Gerhardt, C. 2001. Habituation as a mechanism of reduced aggression between neighboring territorial male bullfrogs (Rana catesbeiana). Journal of Comparative Psychology, 115, 68–82. Brown, R. E. 1988. Individual odors of rats are discriminable independently of changes in gonadal hormone levels. Physiology & Behavior, 43, 359–363. Brown, R. E., Singh, P. B. & Roser, B. 1987. The major histocompatibility complex and the chemosensory recognition of individuality in rats. Physiology & Behavior, 40, 65–73. Cohen, A. B., Johnston, R. E. & Kwon, A. 2001. How hamsters discriminate top from bottom flank scents in over-marks. Journal of Comparative Psychology, 115, 241–247. Ferkin, M. H., Dunsavage, J. & Johnston, R. E. 1999. What kind of information do meadow voles (Microtus pennsylvanicus) use to distinguish between top and bottom scent of an over-mark. Journal of Comparative Psychology, 113, 43–51. Fischer, J., Metz, M., Cheney, D. L. & Seyfarth, R. M. 2001. Baboon responses to graded bark variants. Animal Behaviour, 61, 925–931. doi:10.1006/anbe.2000.1687. Gamboa, G. J., Reeve, H. K. & Holmes, W. G. 1991. Conceptual issues and methodology in kin-recognition research: a critical discussion. Ethology, 88, 109–127. Gheusi, G., Goodall, G. & Dantzer, R. 1997. Individually distinctive odors represent individual conspecifics in rats. Animal Behaviour, 53, 935–944. Gouat, P., Patris, B. & Lalande, C. 1998. Conspecific and heterospecific behavioural discrimination of individual odors by moundbuilding mice. Comptes Rendus de L’Academie des Sciences Serie III-Sciences de la Vie, 321, 571–575. Halpin, Z. T. 1974. Individual differences in the biological odors of the Mongolian gerbil. Behavioral Biology, 11, 253–259. Halpin, Z. T. 1986. Individual odors among mammals: origins and functions. Advances in the Study of Behavior, 16, 39–70. Hare, J. F. 1998. Juvenile Richardson’s ground squirrels, Spermophilus richardsonii, discriminate among individual alarm callers. Animal Behaviour, 55, 451–460. Harrington, J. E. 1974. Recognition of territorial boundaries by olfactory cues in mice (Mus musculus L.). Zeitschrift fu¨r Tierpsychologie, 41, 295–306. Harrington, J. E. 1976. Discrimination between individuals by scent in Lemur fulvus. Animal Behaviour, 24, 207–212. Heth, G., Todrank, J. & Johnston, R. E. 1999. Similarity in the qualities of individual odors among kin and species in Turkish (Mesocricetus brandti) and golden (Mesocricetus auratus) hamsters. Journal of Comparative Psychology, 113, 321–326. Hunter, A. J. & Murray, T. K. 1989. Cholinergic mechanisms in a simple test of olfactory learning in the rat. Psychopharmacology, 99, 270–275. Johnston, R. E. 1975. Scent marking by male golden hamsters (Mesocricetus auratus) III. Behavior in a seminatural environment. Zeitschrift fu¨r Tierpsychologie, 37, 213–221. Johnston, R. E. 1985. Communication. In: The Hamster: Reproduction and Behavior (Ed. by H. I. Siegel), pp. 121–154. New York: Plenum. Johnston, R. E. 1990. Chemical communication in golden hamsters: from behavior to molecules and neural mechanisms. In:

279

280

ANIMAL BEHAVIOUR, 64, 2

Contemporary Issues in Comparative Psychology (Ed. by D. A. Dewsbury), pp. 381–409. Sunderland, Massachusetts: Sinauer. Johnston, R. E. 1993. Memory for individual scents in hamsters (Mesocricetus auratus) as assessed by habituation methods. Journal of Comparative Psychology, 107, 201–207. Johnston, R. E. 1999. How do hamsters know whose scent is on top and why should it matter? In: Advances in Chemical Signals in Vertebrates (Ed. by R. E. Johnston, D. Mu¨ller-Schwarze & P. Sorensen), pp. 227–238. New York: Kluwer Academic/Plenum. Johnston, R. E. & Bhorade, A. 1998. Perception of scent over-marks by golden hamsters (Mesocricetus auratus): novel mechanisms for determining which individual’s mark is on top. Journal of Comparative Psychology, 112, 230–243. Johnston, R. E. & Bullock, T. A. 2001. Individual recognition by use of odours in golden hamsters: the nature of individual representations. Animal Behaviour, 61, 545–557. doi:10.1006/ anbe.2000.1637. Johnston, R. E. & Jernigan, P. 1994. Golden hamsters recognize individuals, not just individual scents. Animal Behaviour, 48, 129–136. Johnston, R. E., Derzie, A., Chiang, G., Jernigan, P. & Lee, H.-C. 1993. Individual scent signatures in golden hamsters: evidence for specialization of function. Animal Behaviour, 45, 1061– 1070. Johnston, R. E., Chiang, G. & Tung, C. 1994. The information in scent over-marks of golden hamsters. Animal Behaviour, 48, 323–330. Johnston, R. E., Munver, R. & Tung, C. 1995. Scent counter marks: selective memory for the top scent by golden hamsters. Animal Behaviour, 49, 1435–1442. Johnston, R. E., Sorokin, E. S. & Ferkin, M. H. 1997a. Female voles discriminate males’ over-marks and prefer top-scent males. Animal Behaviour, 54, 679–690. Johnston, R. E., Sorokin, E. S. & Ferkin, M. H. 1997b. Scent counter-marking by male meadow voles: females prefer the top-scent male. Ethology, 103, 443–453. Keevin, T. M., Halpin, Z. T. & McCurdy, N. 1981. Individual and sex-specific odors in male and female eastern chipmunks. Biology of Behaviour, 6, 329–338. Krames, L. 1970. Responses of female rats to the individual body odors of male rats. Psychonomic Science, 20, 274–275. Krebs, J. R. 1976. Habituation and song repertoires in the great tit. Behavioral Ecology and Sociobiology, 1, 215–227. Lambrechts, M. M. & Dhondt, A. A. 1995. Individual voice discrimination in birds. In: Current Ornithology. Vol. 12 (Ed. by D. M. Power), pp. 115–139. New York: Plenum. Martin, I. G. & Beauchamp, G. K. 1982. Olfactory recognition of individuals by male cavies (Cavia aperea). Journal of Chemical Ecology, 8, 1241–1249. Mateo, J. M. & Johnston, R. E. 2000. Kin recognition and the ‘armpit effect’: evidence of self-referent phenotype matching. Proceedings of the Royal Society of London, Series B, 267, 695–700. Mertl, A. S. 1975. Discrimination of individuals by scent in a primate. Behavioral Biology, 14, 505–509. Mertl, A. S. 1977. Habituation to territorial scent marks in the field by Lemur catta. Behavioral Biology, 21, 500–507. Murdock, H. G. & Randall, J. A. 2001. Olfactory communication and neighbor recognition in giant kangaroo rats. Ethology, 107, 149–160. Murphy, M. R. 1977. Intraspecific sexual preferences of female hamsters. Journal of Comparative and Physiological Psychology, 91, 1337–1346. Napier, J. R. & Napier, P. H. 1985. The Natural History of the Primates. Cambridge, Massachusetts: MIT Press. Nelson, D. A. & Marler, P. 1989. Categorical perception of a natural stimulus continuum: birdsong. Science, 244, 976–978.

Newman, K. S. & Halpin, Z. T. 1988. Individual odours and mate recognition in the prairie vole, Microtus ochrogaster. Animal Behaviour, 36, 1779–1787. Peeke, H. V. S. & Veno, A. 1973. Stimulus specificity of habituated aggression in the stickleback (Gasterosteus aculeatus). Behavioral Biology, 8, 427–432. Petrulis, A. & Johnston, R. E. 1999. Lesions centered on the medial amygdala impair scent-marking and sex-odor recognition but spare discrimination of individual odors in female golden hamsters. Behavioral Neuroscience, 113, 345–357. Petrulis, A., DeSouza, I., Schiller, M. & Johnston, R. E. 1998. Role of frontal cortex in social odor discrimination and scent marking in female golden hamsters (Mesocricetus auratus). Behavioral Neuroscience, 112, 199–212. Petrulis, A., Peng, M. & Johnston, R. E. 1999a. Effects of vomeronasal organ removal on individual odor discrimination, sex-odor preference, and scent marking by female hamsters. Physiology & Behavior, 66, 73–83. Petrulis, A., Peng, M. & Johnston, R. E. 1999b. Lateral olfactory tract transections impair discrimination of individual odors, sex odor preferences, and scent marking in female golden hamsters (Mesocricetus auratus). In: Advances in Chemical Signals in Vertebrates. Vol. 8 (Ed. by R. E. Johnston, D. Mu¨ller-Schwarze & P. Sorensen), pp. 549–561. New York: Plenum. Petrulis, A., Peng, M. & Johnston, R. E. 2000. The role of the hippocampal system in social odor discrimination and scent-marking in female golden hamsters (Mesocricetus auratus). Behavioral Neuroscience, 114, 184–195. Randall, J. A. 1991. Sandbathing to establish familiarity in the Merriam’s kangaroo rat, Dipodomys merriami. Animal Behaviour, 41, 267–275. Rendall, D., Rodman, P. S. & Emond, R. E. 1996. Vocal recognition of individuals and kin in free-ranging rhesus monkeys. Animal Behaviour, 51, 1007–1015. Rosenthal, R. 1976. Experimenter Effects in Behavioral Research. New York: Irvington. Schellinck, H. M., Brown, R. E. & Slotnick, B. M. 1991. Training rats to discriminate between the odors of individual conspecifics. Animal Learning & Behavior, 19, 223–233. Schellinck, H. M., West, A. M. & Brown, R. E. 1992. Rats can discriminate between the urine odors of genetically identical mice maintained on different diets. Physiology & Behavior, 51, 1079–1082. Schellinck, H. M., Rooney, E. & Brown, R. E. 1995. Odors of individuality of germfree mice are not discriminated by rats in a habituation–dishabituation procedure. Physiology & Behavior, 57, 1005–1008. Singh, P. B., Brown, R. E. & Roser, B. 1987. MHC antigens in urine as olfactory recognition cues. Nature, 327, 161–164. Sliwa, A. & Richardson, P. R. K. 1998. Responses of aardwolves, Proteles cristatus, Sparrman 1783, to translocated scent marks. Animal Behaviour, 56, 137–146. Stoddard, P. K. 1996. Vocal recognition of neighbors by territorial passerines. In: Ecology and Evolution of Acoustic Communication in Birds (Ed. by D. E. Kroodsma & E. H. Miller), pp. 356–374. London: Cornell University Press. Tang-Martinez, Z. L., Mueller, L. & Taylor, G. T. 1993. Individual odors and mating success in the golden hamster, Mesocricetus auratus. Animal Behaviour, 45, 1141–1151. Todrank, J. & Heth, G. 1996. Individual odours in two chromosomal species of blind, subterranean mole rat (Spalax ehrenbergi): conspecific and cross-species discrimination. Ethology, 102, 806–811. Todrank, J., Heth, G. & Johnston, R. E. 1998. Kin recognition in golden hamsters: evidence for kinship odours. Animal Behaviour, 55, 377–386.

MAYEAUX & JOHNSTON: ODOUR DISCRIMINATION AND SPACE

Todrank, J., Heth, G. & Johnston, R. E. 1999. Social interaction is necessary for discrimination between and memory for odours of close relatives in golden hamsters. Ethology, 105, 771– 782. Wilcox, R. M. & Johnston, R. E. 1995. Scent counter-marks: specialized mechanisms of perception and response to individual

odors in golden hamsters (Mesocricetus auratus). Journal of Comparative Psychology, 109, 349–356. Wynne-Edwards, K. E., Surov, A. V. & Telitzina, A. Y. 1992. Field studies of chemical signaling: direct observations of dwarf hamsters (Phodopus) in Soviet Asia. In: Chemical Signals in Vertebrates VI (Ed. by R. L. Doty & D. Mu¨ller-Schwarze), pp. 485–491. New York: Plenum.

281