Individual recognition of social rank and social memory performance depends on a functional circadian system

Accepted Manuscript Title: Individual recognition of social rank and social memory performance depends on a functional circadian system Author: L. Muller ¨ D. Weinert PII: DOI: Reference:

S0376-6357(16)30292-3 http://dx.doi.org/doi:10.1016/j.beproc.2016.10.007 BEPROC 3317

To appear in:

Behavioural Processes

Received date: Revised date: Accepted date:

8-7-2016 22-9-2016 11-10-2016

Please cite this article as: Muller, ¨ L., Weinert, D., Individual recognition of social rank and social memory performance depends on a functional circadian system.Behavioural Processes http://dx.doi.org/10.1016/j.beproc.2016.10.007 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Individual recognition of social rank and social memory performance depends on a functional circadian system. L. Müller, D. Weinert1

Institute of Biology/ Zoology, Martin Luther University Halle-Wittenberg, Germany

Running title: Social recognition

1Address

for correspondence:

D. Weinert Martin-Luther-Universität Halle-Wittenberg Institut für Biologie/Zoologie Domplatz 4 D-06108 Halle Phone: +49-345-5526464; Fax: +49-345-5527152 [email protected] Highlights 1. Circadian rhythms are an inherent property of all living systems and essential for their health and wellbeing. Accordingly, disruptions may have adverse consequences for animals´ fitness.

1

2. Arrhythmic or non-circadian Djungarian hamsters reveal normal sociability but cannot realize the social rank of conspecifics and their own rank. 3. Arrhythmic hamsters cannot recognize a familiar conspecific after a short time interval of 60 minutes already. 4. Subordinate animals have better cognitive abilities in a social context because of their higher motivation.

2

ABSTRACT In a natural environment, social abilities of an animal are important for its survival. Particularly, it must recognize its own social rank and the social rank of a conspecific and have a good social memory. While the role of the circadian system for object and spatial recognition and memory is well known, the impact of the social rank and circadian disruptions on social recognition and memory were not investigated so far. In the present study, individual recognition of social rank and social memory performance of Djungarian hamsters revealing different circadian phenotypes were investigated. Wild type (WT) animals show a clear and wellsynchronized daily activity rhythm, whereas in arrhythmic (AR) hamsters, the suprachiasmatic nuclei (SCN) do not generate a circadian signal. The aim of the study was to investigate putative consequences of these deteriorations in the circadian system for animals´ cognitive abilities. Hamsters were bred and kept under standardized housing conditions with food and water ad libitum and a 14 L/10 D lighting regimen. Experimental animals were assigned to different groups (WT and AR) according to their activity pattern obtained by means of infrared motion sensors. Before the experiments, the animals were given to develop a dominant-subordinate relationship in a dyadic encounter. Experiment 1 dealt with individual recognition of social rank. Subordinate and dominant hamsters were tested in an open arena for their behavioral responses towards a familiar (known from the agonistic encounters) or an unfamiliar hamster (from another agonistic encounter) which had the same or an opposite social rank. The investigation time depended on the social rank of the WT subject hamster and its familiarity with the stimulus animal. Both subordinate and dominant WT hamsters preferred an unfamiliar subordinate stimulus animal. In contrast, neither subordinate nor dominant AR hamsters preferred any of the stimulus animals. Thus, disruptions in circadian system result in an impaired individual recognition of social rank. A social recognition/discrimination task was used in Experiment 2 to quantify social memory performance. In a training session, the hamsters were confronted with two unfamiliar stimulus animals. In the test session, one of the two animals was replaced. The training-test interval was 2 min or 24 h. The times animals did explore the novel and the familiar stimulus animal were recorded, and the discrimination index as a measure of cognitive performance was calculated. Behavioral tests revealed that after 2 min both subordinate and dominant WT hamsters were able to 3

discriminate between familiar and novel stimulus animals but after 24 h only the subordinate animals. On contrary in AR hamsters, only subordinates were able to perform the social recognition/discrimination task and only after a training-test interval of 2 min. The results show that the social rank and the circadian system have an impact on the cognitive abilities of Djungarian hamsters. Disruptions of circadian rhythms impair individual recognition and social memory performance. Keywords Circadian rhythms, arrhythmicity, social rank, individual recognition, social memory, Djungarian hamster

4

INTRODUCTION Social abilities are essential for animals´ survival and fitness, as they play a crucial role in pair bonding and reproduction, in territoriality and hierarchies (Alcock 2005). If two or more animals of the same species interact with each other, usually a dominant-subordinate relationship develops. The dominant individual shows threat and fighting behavior whereas the subordinate individual shows submissive behavior, which is characterized by humility and defensive reactions (Drews 1993). Dominant animals have a stronger immune system (Devoino et al. 1993; Raab et al. 1986). As a consequence they are more resistant to diseases and may have a longer lifespan than subordinate ones. The social rank also influences the fitness of an animal. For dominant mice (Mus musculus), a higher sperm activity and density was described (Koyama and Kamimura 2000). In golden hamsters kept in a seminatural environment, dominant males were more attractive to females (Lisk and Baron 1983) and had twice as much offspring compared to subordinate hamsters (Huck et al. 1986). Stable social relationships can be established only if the individuals can recognize each other. Individual recognition is defined as recognition of an animal on the basis of a previous encounter (Lai et al. 2005). It allows to distinguish between members of the same or another species (Vasilieva et al. 2001), to differentiate between relatives and non-relatives (Mateo 2003) and to identify the social rank (Lai and Johnston 2002; Lai, Chen, and Johnston 2004; Petrulis, Weidner, and Johnston 2004) of individuals. The information acquired during an encounter concerning particularly the social rank of the opponent can be used in a re-encounter to adapt behavior. This way, aggressive conflicts can be avoided, what lowers energy costs and the risk of injury (Drews 1993). Nocturnal rodents do use mainly olfactory cues (Johnston et al. 1993). A lesion of the vomeronasal organ or a chemical disruption of the sensory system (Johnston and Peng 2000; Liebenauer and Slotnick 1996; Matochik 1988) impairs or even prevents individual recognition. Intact animals can distinguish between dominant and subordinate males by odour cues as the chemical composition of their scent marks is different (Krames, Carr, and Bergman 1969; Apps, Rasa, and Viljoen 1988). During social encounters, animals perceive the smell of urine, feces or substances from special scent glands of the opponent. This information is stored in an olfactory based social memory and recalled during a subsequent encounter (Thor 5

and Holloway 1982; Lai et al. 2005). To quantify social memory performance, Engelmann and co-workeres developed a social recognition/discrimination task (Engelmann, Wotjak, and Landgraf 1995). It is based on the innate drive of rodents to seek and explore novelty. In a training session, the experimental animal is given the opportunity to explore a stimulus animal. After a break, it will be confronted not only with the well-known animal, but at the same time also with an unknown animal (test session). If the experimental animal is able to remember the familiar animal, it will longer explore the novel one. This task represents a sensitive measure of social recognition and social memory and can be used to investigate the underlying neuroendocrinologic processes (Choleris et al. 2009). Very few studies were aimed to investigate the effect of social rank on individual recognition and social memory. Subordinate deer mice (Peromyscus maniculatus) display better social learning abilities than their dominant conspecifics (Kavaliers et al., 2005). Also in crabs (Chasmagnathus granulatus) subordinate animals show a better social memory than dominant ones (Kaczer, Pedetta, and Maldonado 2007). The role of circadian rhythms and particularly the putative effect of rhythm disruption have not been investigated so far. Circadian rhythms are an inherent property of all living systems. In mammals, they are generated by an internal clock which is localized in the suprachiasmatic nuclei (SCN) of the hypothalamus (Reppert and Weaver 2002). Numerous physiological and behavioural processes are regulated rhythmically. As a consequence, disruptions of the circadian system may have adverse consequences for the animals. As the circadian system is also involved in cognitive processes (for references, see (Müller, Fritzsche, and Weinert 2015), disrupted circadian rhythms do cause cognitive impairments. This has been shown in male rats (R. norvegicus) and golden hamsters (Mesocricetus auratus)(Fekete et al. 1985; Antoniadis et al. 2000). In arrhythmic Djungarian hamsters (Phodopus sungorus), the cognitive performance in a novel object recognition test (Müller, Fritzsche, and Weinert 2015; Ruby et al. 2008) and also the spatial memory (Ruby et al. 2013) are diminished. Disrupted circadian rhythms have an effect also on the emotional behavior of Djungarian hamsters, particularly depressive- and anxiety-like behaviors (Prendergast et al. 2014). The present study was performed to investigate the influence of circadian rhythms on both individual recognition and social memory performance considering 6

also animals´ social rank. The Djungarian hamster is best suited for such studies. Animals of this species live mainly solitary (Flint 1966; Feoktistova 2008). At the same time, they display intense pair bonding (Crawley 1984) and social tolerance towards conspecifics of the same sex (unpublished own results). Furthermore, they are able to discriminate olfactory between members of the same and another species (Vasilieva et al. 2001) (unpublished own results). Djungarian hamsters are also a good model to study circadian rhythms (Steinlechner, Stieglitz, and Ruf 2002; Grone et al. 2011; Schöttner and Weinert 2010; Schöttner et al. 2015; Schöttner, Hauer, and Weinert 2016). In our breeding colony they reveal three rhythmic phenotypes (Weinert and Schöttner 2007). Wild type (WT) hamsters have a well-entrained activity pattern. Their activity onset is in synchrony with ‘‘lights-off’’, the activity offset with ‘‘lights-on’’. The DAO animals show a delayed activity onset. The activity offset, on the other hand, remains coupled to ‘‘lights-on’’. As a consequence, the activity period is compressed. When it goes below a critical value of 3:03 ± 0:02 h, the activity rhythm starts to free-run and breaks down finally. Hamsters then show an arrhythmic activity pattern, being active episodically throughout the 24-h day (AR animals). The AR phenotype can be observed not only under LD conditions but also in constant light or darkness, i.e. it is not a consequence of photic masking. According to Steinlechner and co-authors a collision of the morning and the evening oscillator might have occurred, leading to a zero output of the SCN (Steinlechner et al. 20002). Evidence in favour of this hypothesis comes from results obtained on melatonin and body temperature. In both cases, arrhythmic daily patterns were found (Schöttner, Waterhouse, and Weinert 2011; Schöttner et al. 2011). Obviously, the SCN of these animals do not generate a circadian signal. Hence, these hamsters behave like SCN-lesioned animals (Silver et al. 1996; Wollnik and Turek 1989), and this is a big advantage of our model as no surgery must be performed. An SCN lesion is a heavy intervention which not only abolishes circadian rhythmicity but may compromise other functions as well. Because of the close relationship between the circadian system and both individual recognition and social memory, the aim of the present study was to investigate if WT and AR animals differ in their ability to recognize and remember an individual and its social rank. In addition, a putative effect of animals´ social rank on its cognitive performance was investigated.

7

MATERIAL AND METHODS Animals and housing conditions Investigations were carried out on male Djungarian hamsters (Phodopus sungorus PALLAS 1773) of our own breeding colony (Weinert and Schöttner 2007). Animals were singly housed in standard plastic cages (Macrolon® Type II) with wire mesh tops. Bedding material (Allspan®, Karlsruhe, Germany) was renewed every two weeks. Food (Altromin® 7024; Altromin GmbH, Lage, Germany) and tap water were available ad libitum. Cages were kept in air-conditioned windowless rooms at a temperature of 21 ± 2 °C and a relative humidity of 55–65%. The light/dark condition was 14:10 h, with lights on from 04:00 to 18:00 h Central European Time and light intensities of 145 – 160 versus 0 lux measured at cage bottom. Hamsters were derived from two breeding lines. One line yields almost exclusively WT offspring by pairing the WT hamsters that were most unrelated genetically. By contrast, for the second line unrelated DAO animals are paired what leads to a high percentage of DAO offspring which is characterized by a significant longer free-running period. Part of these DAO hamsters became AR as a consequence of an extreme shortening of their activity period (Weinert and Schöttner 2007). Besides that, no further differences between the animals of the two lines were observed. To increase the sample size, a disruption of the circadian activity rhythm was induced in further animals by light treatment (Schöttner, Limbach, and Weinert 2011). DAO hamsters with an activity onset between 20:00 and 01:00 h CET, were exposed to a light/dark cycle of 16:08 h, with lights on from 02:00 to 18:00 h CET. After two weeks, 50% of DAO hamsters became arrhythmic. Experimental animals were assigned to different groups (WT and AR) on the basis of their daily activity pattern obtained by means of passive infrared motion sensors. The software ‘‘KitAnalyze’’ (Chronobiological Kit®, Stanford University, Stanford, CA) was used to display actograms on computer screen. To verify the absence of a circadian rhythm in AR hamsters, a Chi2-periodogram analysis was performed. Animal number, their mean age and weight (±SEM) at the beginning of the experiments were as follows: WT - n = 24, 198 ± 6 d, 33.9 ± 0.8 g; AR - n = 16, 222 ± 12 d, 43.5 ± 0.9). Furthermore, three unrelated juvenile stimulus animals (mean age: 8

2 months) were used for social memory tests (Experiment 2). Juvenile animals were chosen to avoid any aggressive behavior between test and stimulus animals (Thor and Holloway 1982). To facilitate the formation of a dominant-subordinate relationship during agonistic encounters, unrelated animals with a mean body mass difference 6,6 ± 0,4 g (WT) and 4,9 ± 0,9 g (AR) were paired. All experimental procedures were conducted according to the German law for animal protection and were reviewed and approved by the authorized person for animal protection of the Halle University. Procedures All experiments were performed at the end of light period (14:00 to 17:00 h CET) in a separate room with similar environmental conditions as described above. Thirty minutes before the experiments, animals were transferred to this room. Apparatus and objects The tests were performed in an open-topped arena (LWH = 50 x 50 x 27.5 cm) made of transparent plexiglass, pasted up from outside with white paper to avoid translucency and to enhance contrast for video recording. Light intensity on the arena bottom was between 110 and 120 lux. All behavioral sessions were recorded by a high resolution mini video camera (AL-STWP600, B.Videotechnik) placed 85 cm above arena bottom and the program Ulead VideoStudio® 7 SE DVD (Ulead Systems, Inc., Taipeh, Taiwan). During testing, stimulus animals were covered with perforated plastic boxes to prevent direct physical contact but allowing visual, auditory and olfactory interactions between animals (Litvin et al. 2013). The boxes were placed in the arena at a distance of 8 cm from arena walls allowing the animals to explore them freely and were weighed with water-filled bottles. Experiment 1: Individual Recognition Agonistic encounters On the first day, the two animals of each pair were placed together in the arena in order to develop a dominant-subordinate relationship. One, randomly chosen hamster was marked on its back with a fur cut to enable individual identification. To 9

minimize stress, the confrontation was aborted if one of the following events was observed: (1) one hamster lay on its back for a minimum of 3 seconds or, (2) both hamsters formed a so-called “rolling fight” (Floody and Pfaff 1977) and one of them stopped to fight and fled from the other one. The hamster which lay on its back side or did escape was designated as subordinate, the other one as dominant. The time from the first contact until one of the hamster lay down on its back or fled was defined as „escape latency“. The encounter procedure was repeated after 10 minutes, so that each pair was confronted two times a day. After each fight, both animals were returned to their home cages and the arena was cleaned with a 60% alcohol solution. In both experiments, the dyadic encounters were performed three times (see below), i.e. in total, the animals were tested 12 times in the course of 6 weeks. Over the entire period, the ranking was stable with only one exception. The pair, whose animals showed a reversal of their social ranks, was excluded from further analyses. Habituation 24 hours after agonistic encounters, the animals were placed individually into the arena, which contained two identical empty plastic boxes, and were allowed to explore for 10 min. This way, the hamsters were familiarized with the experimental setup. In addition, it was possible to estimate the general exploratory behavior and to compare between subordinate and dominant animals (Lai and Johnston 2002). Test Immediately after habituation, each hamster was tested for its behavioral responses toward either a familiar stimulus hamster (known from the agonistic encounters) or an unfamiliar one (from another agonistic encounter) which had the same or an opposite social rank. The stimulus animals were placed in a plastic box, a second box was empty, and the test hamsters were allowed to explore them for 10 minutes as in the habituation sessions. Thereafter, the test and the stimulus hamster were returned to their home cages. The subordinate hamster of each pair was tested first, the dominant hamster after an interval of at least 1 hour. This sequence was chosen as it coincides with the study of Petrulis et al. (Petrulis, Weidner, and Johnston 2004). For each animal, the three tests were performed three times in a random order. The interval between consecutive tests was exactly one week. Before 10

the first hamster and after each test, the surface of the arena and the plastic boxes were cleaned with a 60% alcohol solution to avoid effects of lingering olfactory cues. Experiment 2: Social Memory Performance Agonistic encounters The agonistic encounters were performed as described for Experiment 1. Habituation 24 hours after agonistic encounters, the animals were placed individually into the empty arena, i.e. without plastic boxes, and were allowed to explore for 10 min. Training Immediately after the habituation, the hamster was placed in the arena on the opposite side of two identical plastic boxes, each containing an unknown juvenile animal, and was allowed to explore them for 10 min. Thereafter, the animal was returned to its home cage. Test The tests were performed 2 min or 24 h after training in order to evaluate both the short- and the long-time memory (Taglialatela et al. 2009). Between the tests was a break of one week. In the test sessions, one of the two juvenile stimulus animals (familiar) was the same as in the training session; the other was a novel one. Both were covered with identical plastic boxes. Stimulus animals used as novel or familiar ones and their position (left or right side) were counterbalanced between tests sessions. Each single test was limited to 10 minutes as were the training sessions. The surface of the arena and the plastic boxes were cleaned before the first hamster and after each test with a 60% alcohol solution to avoid effects of lingering olfactory cues. Data acquisition and statistical analyses Video records of habituation, training and test sessions were analysed using the program EthoVision XT® (Noldus Information Technology, Wageningen, The Netherlands). The time spent by the hamsters exploring the plastic boxes was 11

estimated. Exploration was defined as directing the nose to a plastic box at a distance of ≤ 2 cm or touching it with the nose as described elsewhere (Müller et al. 2015). To quantify social memory, the so called discrimination index (DI) was calculated (Akkerman et al. 2012). DI = (time exploring the novel stimulus animal – time exploring the familiar stimulus animal)/total time exploring both stimulus animals * 100. The DI may range from 100 (exclusive exploration of the novel stimulus animal) to -100 (exclusive exploration of the familiar stimulus animal), and positive DIs indicate a preference for the novel stimulus animal, i.e. a good memory. The DI must be significantly different from zero, as DI = 0 indicates that the animal has no preference for any stimulus animal, i.e. it has a poor memory (Akkerman et al. 2012). Data were always presented as arithmetical means ± SEM. For statistical analyses the R software, version 2.15.1 (R Development Core Team, 2012) was used. The level of statistical significance was set at p ≤ 0.05. For details, see Results section.

12

RESULTS All hamsters used in the experiments showed clear WT or AR phenotype. Representative actograms are depicted in Figure 1a. WT animals were active mainly in the dark period. AR hamsters on contrary were active episodically throughout the 24-h day. The absence of a circadian component was verified by periodogram analysis (Fig. 1b). Experiment 1: Individual Recognition On the day before each of the three test sessions all hamster pairs underwent two agonistic encounters. The escape latency was similar in both encounters of a day (p > 0.05, Student’s t-test for paired samples). To compare the data obtained in the three successive weeks and those of WT an AR hamsters, a two-way repeated measures ANOVA (two-way RM-ANOVA) followed by posthoc Tukey's test was applied. The escape latency decreased from the first to the third week (Fig. 2), even though not significantly (p > 0.05).The escape latency of AR hamsters was lower compared to WT animals, but the difference was significant only in trial 2 (p ≤ 0.01). For habituation, the WT and AR hamsters were placed individually into the arena, which contained two identical empty plastic boxes. The time the animals did explore the boxes (total investigation time) was estimated in the three successive weeks and analysed by means of RM-ANOVA. No significant difference was obtained (p > 0.05). Moreover, the investigation time did not depend on animals´ social rank (p > 0.05). Subordinate WT hamsters spent 23.0 ± 1.4 %, the dominant WT hamsters 23.2 ± 1.9 % of the ten-minute habituation session investigating the empty boxes. Similar values were obtained in subordinate (21.1 ± 2.7 %) and dominant (20.3 ± 1.9 %) AR hamsters. The difference between WT and AR hamsters was not significant (p > 0.05). Obviously, neither the social rank nor the functionality of the circadian system has an effect on the exploratory behavior. In the test sessions, the behavioral responses of subordinate and dominant hamsters toward either a familiar stimulus hamster (known from the agonistic encounters) or an unfamiliar one which had the same or an opposite social rank (from another agonistic encounter) were investigated. All WT hamsters spent significantly more time investigating the stimulus box than an empty box (Fig. 3). Though, the social rank of both the test and the stimulus animal and familiarity had a 13

significant effect. The longest investigation time was observed if the hamsters were confronted with an unfamiliar subordinate stimulus animal. Dominant hamsters spent more time investigating stimulus animals than subordinate ones. However, this difference was not significant (p > 0.05, Student's t-test for unpaired samples). Also AR hamsters investigated the stimulus boxes longer than the empty boxes (Fig. 4). With one exception, the differences were significant both for subordinate as for dominant animals. However, in contrast to WT hamsters, neither subordinate nor dominant AR hamsters preferred any of the stimulus animals. Also, there was no difference in the investigation time between subordinate and dominant AR hamsters (p > 0.05, RM-ANOVA). Thus in AR hamsters, the investigation time did not depend on the social rank and the familiarity of test and stimulus animals. Experiment 2: Social Memory Performance The social rank and the rhythmic phenotype had an effect on social memory (Figs. 5a and b). A one-sample t-test was used to determine whether DIs were statistically significant from zero. Both subordinate and dominant WT hamsters were able to remember the familiar stimulus animal after a 2-min test interval. By contrast, only subordinate but not dominant AR hamsters did recognize the novel stimulus animal after 2 min. After a 24-h test interval (Fig. 5b), only subordinate WT hamsters did recognize the novel stimulus animals resp. did remember the familiar one. To investigate if the obtained differences in cognitive performance between WT and AR hamsters might have been due to differences in exploratory behavior, total investigation times were analysed. In all training and test sessions, the exploratory behavior was similar and did not depend on circadian phenotype or social rank (p > 0.05; RM-ANOVA).

14

DISCUSSION When Djungarian hamsters were confronted in a dyadic encounter, they developed a dominant-subordinate relationship. Except one WT pair, all animals kept their social rank over the entire experimental period, i.e. over at least 6 weeks. The development of stable dominant-subordinate relationships is based on the “winnerlooser effect” (Chase, Bartolomeo, and Dugatkin 1994). During a subsequent encounter, dominant animals will most probably emerge again as the winner, whereas subordinate animals are more likely to lose again. This results from the assessment of their own fighting ability, so that the animals behave according to the rule: “If I won in the past, be aggressive; if I lost, be cautious.“ (Chase, Bartolomeo, and Dugatkin 1994; Barnard and Burk 1979). No differences in the stability of the social hierarchy were found between WT and AR hamsters. However, the escape latency in dyadic encounters was lower in AR animals, even though not in all cases significantly. Anyway, AR animals seem to be more aggressive than WT animals. when AR and WT hamsters were confronted in a dyadic encounter, the AR animals did show more frequent and longer aggressive behavior (unpublished own results). Similar results were found by Pugh and coworkers on arrhythmic FVB/N mice which showed more offensive behavior than their WT conspecifics (Pugh et al. 2004). During the social investigation test, all hamsters preferred the unfamiliar stimulus animal, independently of their circadian phenotype or social rank. This is in good accordance with animals´ drive to seek novelty (Engelmann, Wotjak, and Landgraf 1995; Müller, Fritzsche, and Weinert 2015). Both subordinate and dominant WT hamsters preferred the subordinate unfamiliar stimulus animal. In an encounter with an inferior animal, it is very likely that they will come off as the winner (Hsu and Wolf 1999). On the other hand, a confrontation with a dominant conspecific yields the risk of injuries (Drews 1993). Moreover, avoiding fights with other dominant animals may help dominant males to maintain their superior social rank. Whereas it seems to be a normal behavior to prefer a subordinate conspecific, the more interesting point is that the hamsters are able to differentiate between dominant and subordinate stimulus animals. This has been shown before for golden hamsters (Lai et al. 2005; Lai, Chen, and Johnston 2004; Petrulis, Weidner, and Johnston 2004; Lai and Johnston 2002) but not yet for Djungarian hamsters. Obviously, animals can either recognize the social rank of their partner or realize their own rank. Evidence in favour of the second 15

point comes from the lower investigation time of subordinate test hamsters. On contrary, neither subordinate nor dominant AR hamsters preferred any of the unfamiliar stimulus animals. Also, there were no significant differences in investigation times between subordinate and dominant test animals. This means that AR hamster were not able to recognize the social rank of stimulus hamsters and to realize their own rank. Thus, they behave like neutral animals, i.e. those not being confronted with conspecifics before. These also show no differences in the investigation time on stimulus animals with different social ranks (Lai and Johnston 2002). From these results, the question came up, why disruptions in the circadian system lead to a decreased ability of individual recognition of social rank. Lai and coworkers (2004) investigated the neuronal mechanisms underlying the individual recognition in golden hamsters (M. auratus). Subordinate males who are confronted again with a familiar dominant animal, have a higher density of activated cells in the anterior dorsal hippocampus, in the dorsal subiculum and in the basolateral amygdala. These brain areas are involved in individual recognition of social rank (Lai, Chen, and Johnston 2004). Also, they are closely related to the SCN and therefore linked to the circadian system (Phan et al. 2011; Lamont et al. 2005). Thus, it can be assumed that the differences between the circadian phenotypes concerning the individual recognition of the social rank probably result from the close linkage of the three mentioned brain areas with the SCN. In the present paper, it was shown for the first time that disruptions of the circadian system lead to reduced individual recognition of social rank. This may have far-reaching consequences for AR animals, because they cannot reflect their own social rank and accordingly cannot adapt their behavior in agonistic encounters. AR animals might not be able to avoid aggressive conflicts in re-encounters, which leads to high energy costs and a high risk of injury and this has an adverse effect on their chances for survival. A social recognition/discrimination task was carried out to investigate social memory performance of subordinate and dominant WT and AR hamsters. After a 2min training-test interval the animals, except the dominant AR hamsters, did remember the familiar test animal resp. did recognize the novel one. After 24 h, only the subordinate WT hamsters did recognize the novel stimulus animal. These results show that, cognitive performance depends on the social rank of the test animal and 16

the functionality of its circadian system. Subordinate hamsters of both rhythmic phenotypes showed a better social recognition ability and memory than there dominant conspecifics. Similar results have been described for other rodents (Kavaliers, Colwell, and Choleris 2005) but also for crabs (Kaczer, Pedetta, and Maldonado 2007). On contrary, dominant mice (M. musculus) showed a better spatial learning ability (Fitchett et al. 2005; Barnard and Luo 2002). Obviously, the effect of the social rank on memory performance depends on the learning paradigm used. According to Barnard and Luo (2002), learning is costly, and thus the investment in learning will vary with its value for animals´ fitness. Dominant animals are more explorative than subordinates and thus acquire more information from their environment. This difference may be the reason for the better performance in spatial learning (Francia et al. 2006). In contrast, subordinate animals might focus on social events, because they should avoid aggressive conflicts with dominant individuals (Drews 1993). Hence, subordinate animals are highly motivated to identify and to memorize the social status of a conspecific, and this was shown in the present paper. The investigation time was not different between dominant and subordinate hamsters but, subordinate animals did perform better in the social recognition test, certainly due to their higher motivation. Important modulators of social behavior are the two neuropeptides oxytocin and vasopressin (Bielsky and Young 2004; Ross and Young 2009; Carter et al. 2008; Lukas and Neumann 2013; Oettl et al. 2016). Infusion of vasopressin and oxytocin into brain areas which are important for olfactory sense facilitates the recognition of conspecifics and increased the duration of social memory in rats (Dluzen et al. 1998; Dantzer et al. 1988). Studies on oxytocin knockout mice revealed an impaired social memory which can be improved by oxytocin administration (Ferguson et al. 2001; Ferguson et al. 2000). According to Crawley and co-authors, oxytocin deficiency has no effect on the sociability of mice measured as time spent with a novel unfamiliar conspecific, though diminishes their social memory. Subordinate mice (M. musculus) show high oxytocin and vasopressin levels due to chronic stress (Litvin, Murakami, and Pfaff 2011), and this may be the reason for their better social memory. Similar mechanisms can be assumed for Djungarian hamsters. In the present study, no differences depending on social rank in sociability but in cognitive performance were obtained. Also the neurotransmitter dopamine is involved in the control of social behavior 17

and memory (Aragona et al. 2003). High dopamine activity leads to an impairment of social memory (Millan et al. 2007). Dominant crab-eating macaques (Macaca fascicularis) have a significantly higher distribution of dopamine receptors than their subordinate conspecifics (Morgan et al. 2002). A similar mechanism may have caused the impaired memory performance of dominant hamsters in our study. It should be noted, however, that the memory process in subordinate and dominant individuals is not regulated by a single neurotransmitter or neuropeptide, but it is an interaction of several intermeshing neurobiological mechanisms (Bielsky and Young 2004; Lukas and Neumann 2013). In intact animals, neuropeptides and neurotransmitters show circadian rhythms, which are disturbed in AR hamsters, and this might be the reason for their diminished cognitive performance. As shown in the present paper, disrupted circadian rhythms do reduce social memory. According to other studies, AR hamsters have deficits in object recognition and spatial working memory (Müller, Fritzsche, and Weinert 2015; Ruby et al. 2013; Ruby et al. 2008). That the impaired cognitive performance is really due to the loss of circadian rhythmicity was confirmed by the results of a recently published study (Weinert et al. 2016). There, we have shown that, the impaired cognitive performance of AR hamsters did improve if the circadian activity rhythm had re-established as a consequence of intensive voluntary wheel running. In contrast to the present results, Reijmers and co-authors did not find an effect of disrupted circadian rhythms on social memory in male rats (Reijmers et al. 2001). In this study however, single 6-h or 12-h phase shifts were applied to disturb circadian rhythms, and this may have been one reason for the lack of an effect on cognitive abilities. Besides that, species differences must be taken into account. Noack and co-workers investigated the social memory of laboratory mice and rats (Noack et al. 2010). Whereas mice were able to distinguish between familiar and unfamiliar conspecifics after more than 24 h, rats could this only after 45 min. The hormone melatonin (Simonneaux and Ribelayga 2003) and the neurotransmitter γ-aminobutyric acid (GABA) (Belenky, Yarom, and Pickard 2008; Moore and Speh 1993) are important output signals of the SCN, i.e. of the main circadian pacemaker. Also, both are linked to memory performance (Argyriou, Prast, and Philippu 1998; Katz and Liebler 1978; Brioni 1993). Melatonin is realized by the pineal rhythmically with a high level during the dark period (Pevet 2003). In contrast to WT hamsters, which show a typical circadian rhythm, the melatonin level is low 18

throughout the whole day in our AR hamsters (Schöttner et al. 2011). This low level may be the reason for the diminished social memory performance since melatonin has been shown to facilitate memory formation in nocturnal rodents (Argyriou, Prast, and Philippu 1998). Also normally, the GABA release by the SCN is cyclic with higher values during the subjective day (Kalsbeek et al. 2000; Moore and Speh 1993). Accordingly, the inhibitory effect of this transmitter is low during the subjective night and as a consequence the cognitive performance is better. In AR hamsters, GABA might be released not cyclically but at a constant medium level which is above the night time level in WT animals. Ruby and co-workers assumed that the resulting overinhibition of synaptic circuits in brain regions involved in learning and memory processes may cause memory deficits {Ruby, 2013 #1869}{Ruby, 2008 #1437} . By administration of pentylenetetrazol , a noncompetitive GABAa receptor antagonist, it was possible to restore memory performance. In the present paper, we have shown that the circadian system affects individual recognition of social rank and social memory performance. WT animals, which are characterized by stable circadian rhythms, show a better ability to recognize the social rank and to remember familiar conspecifics than AR hamsters. The lack of circadian rhythmicity obviously also concerns neurotransmitters and neuropeptides that are involved in cognition and memory formation. The resulting impairment of individual recognition and social memory performance might have far-reaching consequences for the survival of AR animals.

DECLARATION OF INTEREST The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

19

References Akkerman, S., Blokland, A., Reneerkens, O., van Goethem, N.P., Bollen, E., Gijselaers, H.J., Lieben, C.K., Steinbusch, H.W. and Prickaerts, J. 2012. Object recognition testing: methodological considerations on exploration and discrimination measures. Behav Brain Res, 232: 335-47. Alcock, J. 2005. Animal Behavior - An Evolutionary Approach. Sinauer Associates, Sunderland. Antoniadis, E.A., Ko, C.H., Ralph, M.R. and McDonald, R.J. 2000. Circadian rhythms, aging and memory.[republished from Behav Brain Res. 2000 Jun 15;111(12):25-37; PMID: 10840129]. Behavioural Brain Research, 114: 221-233. Apps, P.J., Rasa, A. and Viljoen, H.W. 1988. Quantitative Chromatographic Profiling of Odors Associated with Dominance in Male Laboratory Mice. Aggressive Behavior, 14: 451-461. Aragona, B.J., Liu, Y., Curtis, J.T., Stephan, F.K. and Wang, Z. 2003. A critical role for nucleus accumbens dopamine in partner-preference formation in male prairie voles. J Neurosci, 23: 3483-90. Argyriou, A., Prast, H. and Philippu, A. 1998. Melatonin facilitates short-term memory. Eur J Pharmacol, 349: 159-62. Barnard, C.J. and Burk, T. 1979. Dominance hierarchies and the evolution of "individual recognition". J Theor Biol, 81: 65-73. Barnard, C.J. and Luo, N. 2002. Acquisition of dominance status affects maze learning in mice. Behav Processes, 60: 53-59. Belenky, M.A., Yarom, Y. and Pickard, G.E. 2008. Heterogeneous expression of gamma-aminobutyric acid and gamma-aminobutyric acid-associated receptors and transporters in the rat suprachiasmatic nucleus. J Comp Neurol, 506: 70832. Bielsky, I.F. and Young, L.J. 2004. Oxytocin, vasopressin, and social recognition in mammals. Peptides, 25: 1565-74. Brioni, J.D. 1993. Role of Gaba during the Multiple Consolidation of Memory. Drug Development Research, 28: 3-27. Carter, C.S., Grippo, A.J., Pournajafi-Nazarloo, H., Ruscio, M.G. and Porges, S.W. 2008. Oxytocin, vasopressin and sociality. Prog Brain Res, 170: 331-6. Chase, I.D., Bartolomeo, C. and Dugatkin, L.A. 1994. Aggressive Interactions and 20

Inter-Contest Interval - How Long Do Winners Keep Winning. Animal Behaviour, 48: 393-400. Choleris, E., Clipperton-Allen, A.E., Phan, A. and Kavaliers, M. 2009. Neuroendocrinology of social information processing in rats and mice. Front Neuroendocrinol, 30: 442-59. Crawley, J.N. 1984. Evaluation of a proposed hamster separation model of depression. Psychiatry Res, 11: 35-47. Dantzer, R., Koob, G.F., Bluthe, R.M. and Le Moal, M. 1988. Septal vasopressin modulates social memory in male rats. Brain Res, 457: 143-7. Devoino, L., Alperina, E., Kudryavtseva, N. and Popova, N. 1993. Immune responses in male mice with aggressive and submissive behavior patterns: strain differences. Brain Behav Immun, 7: 91-6. Dluzen, D.E., Muraoka, S., Engelmann, M. and Landgraf, R. 1998. The effects of infusion of arginine vasopressin, oxytocin, or their antagonists into the olfactory bulb upon social recognition responses in male rats. Peptides, 19: 999-1005. Drews, C. 1993. The Concept and Definition of Dominance in Animal Behavior. Behaviour, 125: 283-313. Engelmann, M., Wotjak, C.T. and Landgraf, R. 1995. Social discrimination procedure: an alternative method to investigate juvenile recognition abilities in rats. Physiol Behav, 58: 315-21. Fekete, M., van Ree, J.M., Niesink, R.J. and de Wied, D. 1985. Disrupting circadian rhythms in rats induces retrograde amnesia. Physiol Behav, 34: 883-7. Feoktistova, N.Y. 2008. Dwarf hamsters (Phodopus:Cricetinae): systematics, phylogeography, ecology, physiology, behaviour, chemical communication. [In Russian]. KMK Scientific Press. Ltd., Moscow. Ferguson, J.N., Aldag, J.M., Insel, T.R. and Young, L.J. 2001. Oxytocin in the medial amygdala is essential for social recognition in the mouse. J Neurosci, 21: 8278-85. Ferguson, J.N., Young, L.J., Hearn, E.F., Matzuk, M.M., Insel, T.R. and Winslow, J.T. 2000. Social amnesia in mice lacking the oxytocin gene. Nat Genet, 25: 284-8. Fitchett, A.E., Collins, S.A., Barnard, C.J. and Cassaday, H.J. 2005. Subordinate male mice show long-lasting differences in spatial learning that persist when housed alone. Neurobiol Learn Mem, 84: 247-51. 21

Flint, W.E. 1966. Die Zwerghamster der paläarktischen Fauna. A. Ziemsen Verlag, Wittenberg 1-97 pp. Floody, O.R. and Pfaff, D.W. 1977. Aggressive behavior in female hamsters: the hormonal basis for fluctuations in female aggressiveness correlated with estrous state. J Comp Physiol Psychol, 91: 443-64. Francia, N., Cirulli, F., Chiarotti, F., Antonelli, A., Aloe, L. and Alleva, E. 2006. Spatial memory deficits in middle-aged mice correlate with lower exploratory activity and a subordinate status: role of hippocampal neurotrophins. Eur J Neurosci, 23: 711-28. Grone, B.P., Chang, D., Bourgin, P., Cao, V., Fernald, R.D., Heller, H.C. and Ruby, N.F. 2011. Acute light exposure suppresses circadian rhythms in clock gene expression. J Biol Rhythms, 26: 78-81. Hsu, Y. and Wolf, L.L. 1999. The winner and loser effect: integrating multiple experiences. Anim Behav, 57: 903-910. Huck, U.W., Lisk, R.D., Allison, J.C. and Vandongen, C.G. 1986. Determinants of Mating Success in the Golden-Hamster (Mesocricetus-Auratus) - SocialDominance and Mating Tactics under Seminatural Conditions. Animal Behaviour, 34: 971-989. Johnston, R.E., Derzie, A., Chiang, G., Jernigan, P. and Lee, H.C. 1993. Individual Scent Signatures in Golden-Hamsters - Evidence for Specialization of Function. Animal Behaviour, 45: 1061-1070. Johnston, R.E. and Peng, M. 2000. The vomeronasal organ is involved in discrimination of individual odors by males but not by females in golden hamsters. Physiol Behav, 70: 537-49. Kaczer, L., Pedetta, S. and Maldonado, H. 2007. Aggressiveness and memory: subordinate crabs present higher memory ability than dominants after an agonistic experience. Neurobiol Learn Mem, 87: 140-8. Kalsbeek, A., Garidou, M.-L., Palm, I.F., Van Der Vliet, J., Simonneaux, V., Pévet, P. and Buijs, R.M. 2000. Melatonin sees the light: blocking GABA-ergic transmission in the paraventricular nucleus induces daytime secretion of melatonin. European Journal of Neuroscience, 12: 3146-3154. Katz, R.J. and Liebler, L. 1978. GABA involvement in memory consolidation: evidence from posttrial amino-oxyacetic acid. Psychopharmacology (Berl), 56: 191-3. 22

Kavaliers, M., Colwell, D.D. and Choleris, E. 2005. Kinship, familiarity and social status modulate social learning about ''micropredators'' (biting flies) in deer mice. Behavioral Ecology and Sociobiology, 58: 60-71. Koyama, S. and Kamimura, S. 2000. Influence of social dominance and female odor on the sperm activity of male mice. Physiology & Behavior, 71: 415-422. Krames, L., Carr, W.J. and Bergman, B. 1969. A Pheromone Associated with Social Dominance among Male Rats. Psychonomic Science, 16: 11-&. Lai, W.S., Chen, A.Y. and Johnston, R.E. 2004. Patterns of neural activation associated with exposure to odors from a familiar winner in male golden hamsters. Hormones and Behavior, 46: 319-329. Lai, W.S. and Johnston, R.E. 2002. Individual recognition after fighting by golden hamsters: A new method. Physiology & Behavior, 76: 225-239. Lai, W.S., Ramiro, L.L., Yu, H.A. and Johnston, R.E. 2005. Recognition of familiar individuals in golden hamsters: a new method and functional neuroanatomy. J Neurosci, 25: 11239-47. Lamont, E.W., Robinson, B., Stewart, J. and Amir, S. 2005. The central and basolateral nuclei of the amygdala exhibit opposite diurnal rhythms of expression of the clock protein Period2. Proc Natl Acad Sci U S A, 102: 41804. Liebenauer, L.L. and Slotnick, B.M. 1996. Social organization and aggression in a group of olfactory bulbectomized male mice. Physiol Behav, 60: 403-9. Lisk, R.D. and Baron, G. 1983. Conditions necessary to the establishment of mating dominance by the male hamster. Behav Neural Biol, 39: 105-15. Litvin, Y., Murakami, G. and Pfaff, D.W. 2011. Effects of chronic social defeat on behavioral and neural correlates of sociality: Vasopressin, oxytocin and the vasopressinergic V1b receptor. Physiol Behav, 103: 393-403. Litvin, Y., Phan, A., Hill, M.N., Pfaff, D.W. and McEwen, B.S. 2013. CB1 receptor signaling regulates social anxiety and memory. Genes Brain Behav, 12: 47989. Lukas, M. and Neumann, I.D. 2013. Oxytocin and vasopressin in rodent behaviors related to social dysfunctions in autism spectrum disorders. Behav Brain Res, 251: 85-94. Mateo, J.M. 2003. Kin recognition in ground squirrels and other rodents. Journal of Mammalogy, 84: 1163-1181. 23

Matochik, J.A. 1988. Role of the main olfactory system in recognition between individual spiny mice. Physiol Behav, 42: 217-22. Millan, M.J., Di Cara, B., Dekeyne, A., Panayi, F., De Groote, L., Sicard, D., Cistarelli, L., Billiras, R. and Gobert, A. 2007. Selective blockade of dopamine D(3) versus D(2) receptors enhances frontocortical cholinergic transmission and social memory in rats: a parallel neurochemical and behavioural analysis. J Neurochem, 100: 1047-61. Moore, R.Y. and Speh, J.C. 1993. GABA is the principal neurotransmitter of the circadian system. Neuroscience Letters, 150: 112-116. Morgan, D., Grant, K.A., Gage, H.D., Mach, R.H., Kaplan, J.R., Prioleau, O., Nader, S.H., Buchheimer, N., Ehrenkaufer, R.L. and Nader, M.A. 2002. Social dominance in monkeys: dopamine D-2 receptors and cocaine selfadministration. Nature Neuroscience, 5: 169-174. Müller, D., Hauer, J., Schöttner, K., Fritzsche, P. and Weinert, D. 2015. Seasonal adaptation of dwarf hamsters (Genus Phodopus): differences between species and their geographic origin. Journal of Comparative Physiology B, 185: 917930. Müller, L., Fritzsche, P. and Weinert, D. 2015. Novel object recognition of Djungarian hamsters depends on circadian time and rhythmic phenotype. Chronobiology International, 32: 458-467. Noack, J., Richter, K., Laube, G., Haghgoo, H.A., Veh, R.W. and Engelmann, M. 2010. Different importance of the volatile and non-volatile fractions of an olfactory signature for individual social recognition in rats versus mice and short-term versus long-term memory. Neurobiol Learn Mem, 94: 568-75. Oettl, L.L., Ravi, N., Schneider, M., Scheller, M.F., Schneider, P., Mitre, M., da Silva Gouveia, M., Froemke, R.C., Chao, M.V., Young, W.S., Meyer-Lindenberg, A., Grinevich, V., Shusterman, R. and Kelsch, W. 2016. Oxytocin Enhances Social Recognition by Modulating Cortical Control of Early Olfactory Processing. Neuron. Petrulis, A., Weidner, M. and Johnston, R.E. 2004. Recognition of competitors by male golden hamsters. Physiol Behav, 81: 629-38. Pevet, P. 2003. Melatonin: from seasonal to circadian signal. J Neuroendocrinol., 15: 422-426. Phan, T.X., Chan, G.C., Sindreu, C.B., Eckel-Mahan, K.L. and Storm, D.R. 2011. The 24

diurnal oscillation of MAP (mitogen-activated protein) kinase and adenylyl cyclase activities in the hippocampus depends on the suprachiasmatic nucleus. J Neurosci, 31: 10640-7. Prendergast, B.J., Onishi, K.G., Patel, P.N. and Stevenson, T.J. 2014. Circadian arrhythmia dysregulates emotional behaviors in aged Siberian hamsters. Behavioural Brain Research, 261: 146-157. Pugh, P.L., Ahmed, S.F., Smith, M.I., Upton, N. and Hunter, A.J. 2004. A behavioural characterisation of the FVB/N mouse strain. Behav Brain Res, 155: 283-9. Raab, A., Dantzer, R., Michaud, B., Mormede, P., Taghzouti, K., Simon, H. and Le Moal, M. 1986. Behavioural, physiological and immunological consequences of social status and aggression in chronically coexisting resident-intruder dyads of male rats. Physiol Behav, 36: 223-8. Reijmers, L.G., Leus, I.E., Burbach, J.P., Spruijt, B.M. and van Ree, J.M. 2001. Social memory in the rat: circadian variation and effect of circadian rhythm disruption. Physiol Behav, 72: 305-9. Reppert, S.M. and Weaver, D.R. 2002. Coordination of circadian timing in mammals. Nature, 418: 935-941. Ross, H.E. and Young, L.J. 2009. Oxytocin and the neural mechanisms regulating social cognition and affiliative behavior. Front Neuroendocrinol, 30: 534-47. Ruby, N.F., Fernandez, F., Garrett, A., Klima, J., Zhang, P., Sapolsky, R. and Heller, H.C. 2013. Spatial memory and long-term object recognition are impaired by circadian arrhythmia and restored by the GABAA antagonist pentylenetetrazole. PLoS One, 8: e72433. Ruby, N.F., Hwang, C.E., Wessells, C., Fernandez, F., Zhang, P., Sapolsky, R. and Heller, H.C. 2008. Hippocampal-dependent learning requires a functional circadian system. Proc Natl Acad Sci U S A, 105: 15593-8. Schöttner, K., Hauer, J. and Weinert, D. 2016. Non-parametric photic entrainment of Djungarian hamsters with different rhythmic phenotypes. Chronobiol Int, 1-14. Schöttner, K., Limbach, A. and Weinert, D. 2011. Re-entrainment behavior of djungarian hamsters (phodopus sungorus) with different rhythmic phenotype following light-dark shifts. Chronobiol Int, 28: 58-69. Schöttner, K., Simonneaux, V., Vuillez, P., Steinlechner, S., Pevet, P. and Weinert, D. 2011. The daily melatonin pattern in Djungarian hamsters depends on the circadian phenotype. Chronobiol Int, 28: 873-82. 25

Schöttner, K., Vuillez, P., Challet, E., Pévet, P. and Weinert, D. 2015. Light-induced cFos expression in the SCN and behavioural phase shifts of Djungarian hamsters with a delayed activity onset. Chronobiol Int, 32: 596-607. Schöttner, K., Waterhouse, J. and Weinert, D. 2011. The circadian body temperature rhythm of Djungarian Hamsters (Phodopus sungorus) revealing different circadian phenotypes. Physiol Behav, 103: 352-58. Schöttner, K. and Weinert, D. 2010. Effects of light on the circadian activity rhythm of Djungarian hamsters (phodopus sungorus) with delayed activity onset. Chronobiol Int, 27: 95-110. Silver, R., LeSauter, J., Tresco, P.A. and Lehman, M.N. 1996. A diffusible coupling signal from the transplanted suprachiasmatic nucleus controlling circadian locomotor rhythms. Nature, 382: 810-813. Simonneaux, V. and Ribelayga, C. 2003. Generation of the melatonin endocrine message in mammals: a review of the complex regulation of melatonin synthesis by norepinephrine, peptides, and other pineal transmitters. Pharmacol Rev, 55: 325-95. Steinlechner, S., Stieglitz, A. and Ruf, T. 2002. Djungarian hamsters: a species with a labile circadian pacemaker? Arrhythmicity under a light-dark cycle induced by short light pulses. Journal of Biological Rhythms, 17: 248-258. Taglialatela, G., Hogan, D., Zhang, W.R. and Dineley, K.T. 2009. Intermediate- and long-term recognition memory deficits in Tg2576 mice are reversed with acute calcineurin inhibition. Behav Brain Res, 200: 95-9. Thor, D.H. and Holloway, W.R. 1982. Social Memory of the Male Laboratory Rat. Journal of Comparative and Physiological Psychology, 96: 1000-1006. Vasilieva, N.Y., Lai, S.C., Petrova, E.V. and Johnston, R.E. 2001. Development of species preferences in two hamsters, Phodopus campbelli and Phodopus sungorus: Effects of cross-fostering. Ethology, 107: 217-236. Weinert, D. and Schöttner, K. 2007. An inbred lineage of Djungarian hamsters with a strongly attenuated ability to synchronize. Chronobiol Int, 24 1065-1079. Weinert, D., Schöttner, K., Müller, L. and Wienke, A. 2016. Intensive voluntary wheel running may restore circadian activity rhythms and improves the impaired cognitive performance of arrhythmic Djungarian hamsters. Chronobiol Int, doi10.1080/07420528.2016.1205083. Wollnik, F. and Turek, F.W. 1989. SCN lesions abolish ultradian and circadian 26

components of activity rhythms in LEW/Ztm rats. Am J Physiol, 256: R1027R1039.

27

Figure 1a: Representative activity patterns of WT and AR hamsters (from top to bottom). Data of 10 days are double-plotted. The white and black bars of the top of actograms and vertical lines indicate the light-dark regimen (L:D = 14:10 h). Figure 1b: Chi² periodogram of the AR hamster shown in Fig. 1a. Straight line corresponds to the p ≤ 0.05 significance level (Chi²-test). Figure 2: Escape latency in agonistic encounters. Each pair of hamsters underwent agonistic encounters on three successive weeks (trial 1 – 3). The escape latency of AR hamsters was lower compared to WT animals, but only significantly in trial 2 (*, p ≤ 0.05, posthoc Tukey test). Data are expressed as mean values ± SEM. Figure 3: Individual recognition of the social rank by WT hamsters. The investigation time on stimulus animals was dependent on social rank and familiarity of the test hamster with the stimulus animal. Both subordinate (a) and dominant (b) WT hamsters investigated the unfamiliar subordinate stimulus animals more than dominant ones. Data are expressed as mean values ± SEM. **, p ≤ 0.01; ***, p ≤ 0.001 (posthoc Tukey test) Figure 4: Individual recognition of the social rank by AR hamsters. The hamsters investigated the stimulus boxes longer than the empty boxes (all differences except one are significant). However, neither subordinate (a) AR hamsters nor (b) dominant (b) AR hamsters preferred any of the stimulus animals (p > 0.05, RM-ANOVA). Data are expressed as mean values ± SEM. *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001 (posthoc Tukey test) Figure 5a: The discrimination index of subordinate and dominant WT and AR hamsters after a training-test interval of 2 min. With one exception (dominant AR animals), the values were significantly different from zero, i.e. these hamsters did recognize the novel stimulus animal. Data are expressed as mean values ± SEM. *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001 one-sample t-test 28

Figure 5b: The discrimination index of subordinate and dominant WT as well as AR hamsters after a training-test interval of 24 h. Only the subordinate WT hamsters could remember the stimulus animal (*, p ≤ 0.05 one-sample t-test). In all other cases the DI values were not significantly different from zero. Data are expressed as mean values ± SEM.

29

Figure 1a

30

Figure 1b

31

Figure 2

32

Figure 3

33

Figure 4

Figure 5a

34

Figure 5b

35