1
BSc Psychobiology
Bachelor project
22 / 01 / 2021
The role of kinship on emotional contagion in rats
Zayel Smit
11669748
Supervisor : Erica Berretta
Second assessor: Simone Mesman
Word count: 5132
2
Abstract
Emotional contagion, the ability to feel what others feel without the compound of knowing what the source is, is an import compound of empathy. In rats, it is known that emotional contagion is
influenced by sex and familiarity. In response to pain to conspecifics, rats can show vicarious freezing. We used the emotional contagion test where the demonstrator is receiving footshocks and behavioural changes were measured. Previous research has shown that during emotional contagion, there is a bidirectional exchange of information between demonstrator and observer. Besides that, is shown that mothers can inhibit freezing in threatful situations when in presence of their pup. Presence of a
conspecific can also induce social buffering for the demonstrator which makes the demonstrator freeze less. In this research, we investigated the role of kinship on emotional contagion in Sprague Dawley rats. In the emotional contagion test, the observer dams were paired with either their own juvenile or the juvenile of another dam, the alien group. We were able to induce emotional contagion in the observer animals. However, the behavioural results show no significance. The maternal care data showed promising results in correlating with the freezing levels of the own observer group during the post shock period of the emotional contagion test. Bayesian statistics shows that the data is still inconclusive and a bigger sample size is needed.
3
Introduction
Empathy, the capacity to share the feelings of others, is a key factor for social interaction. In the process of empathy, there are multiple underlying processes involved. Empathy can be seen as a concept that contains three elements: motor mimicry, emotional contagion and affective resonance (de Waal, 2007). Emotional contagion is a process where an individual (the observer) catches the emotion of another individual (the demonstrator) and automatically activates the same emotional state. With emotional contagion, the observer is not aware of the source of his/its emotions. However, when talking about emotional empathy, the individual is aware that the emotional reaction is caused by the emotion of the demonstrator (Atsak et al., 2011).
The most studied emotions in rodents are pain and fear. Already in 1959, Church showed that rats stop pressing a lever to obtain a food reward when they observed another rat receiving aversive foot shocks caused by the lever pressing (Church, 1959). Also, in another set-up where an observer rat observes a demonstrator rat receiving aversive foot shocks (Emotional contagion test), behavioural changes can be seen. In response to pain to conspecifics, rats can show vicarious freezing (Atsak et al., 2011; Carrillo et al., 2019; Han et al., 2020).
Familiarity between observer and demonstrator can play a role in emotional contagion. A study in mice has shown that the observer freezes more when the demonstrator is a cagemate compared to a stranger (Gonzalez-Liencres et al., 2014). On the other hand, the demonstrator freezes less when the observer is a cagemate compared to a stranger. With other similar findings, it is indicated that in mice the presence of an unfamiliar conspecific has a soothing effect, however, a familiar conspecific can increase the level of experienced pain (Langford et al., 2006).
Indeed, in rats, it has been demonstrated that during the emotional contagion, there is a bidirectional change of information between observer and demonstrator (Han et al., 2020). This bidirectional change takes place in a way that the fear response of the demonstrator (freezing) influences the response of the observer and vice versa. The phenomenon by which the behavioural response to pain is decreased in presence of conspecifics is social buffering. This phenomenon states that being in the company of a
4 conspecific can reduce the negative experience of stress when a subject is exposed to distressing stimuli. The conspecific can have different kinds of relationships with the subject. For instance, mother-pup, mates, same-sex or opposite sex. Previous research showed that in rats, both males and females showed effects of social buffering (Kiyokawa et al., 2007; Ishii et al., 2016). Ishii et al. (2016) showed that female rats that were in presence of a conspecific, reduced the fear responses including freezing and increase in corticosterone levels caused by an auditory conditioned stimulus. However, they also showed that familiarity can influence this effect. When the rats were in presence of an unfamiliar conspecific, there was no effect of social buffering. Thus, the level of social buffering can change based on the degree of affiliation (Kikusui et al., 2006) between individuals. This degree of affiliation is very strong between mother and infant and thereby can be a great influence on the emotional contagion in this research.
Thus, emotional contagion can be influenced by multiple factors as familiarity. However, it is still unknown what the influence of kinship (in term of mother-infant relationship) is on the emotional contagion in rats. This leads to the main question: “What is the effect of kinship on the emotional contagion of pain in female rats?”. To be able to answer this question, we first tested mother rats at the emotional contagion either with their own pup or with an alien pup, focusing on the behavioural responses of the dams to the distress of the demonstrator.
Then we investigated the effect of kinship on the freezing levels of the juvenile demonstrators in the presence of either their own dam or another dam. Finally, we asked if the level of maternal care predict the level of emotional contagion. It has been shown that proestrus females exhibit less anxious and depressive behaviour (Frye et al., 2000) and thereby we also have taken the estrus phase into account. It is known that there is a sex difference on emotional contagion (Han et al., 2020) and therefore only females are used in this experiment.
The hypothesis to the main question is that kinship will change the behavioural response to the emotional contagion of pain. According to previous findings, it is expected that when mothers are with their own pups they freeze less and exhibit more defensive behaviours towards the pups compared to mothers who are paired with an alien pup. Besides that, it is expected that pups also freeze less when paired with their
5 own mother. Furthermore, it is expected that increased maternal care will strengthen the effect of kinship on emotional contagion.
Methods and materials
Subjects
12 time pregnant Sprague Dawley (gestational day 15 at arrival) rats were obtained from the Janvier Laboratories in two separate batches. They were individually housed in the NIN SPF animal facility and were provided ad libitum with water and food. The temperature of the room was 22–24 °C, 50-55 % relative humidity with reversed 12:12 light cycles (light on at 7 pm). After arrival, the animals were left undisturbed for 1 week for acclimation and dams were checked twice a day on the days
surrounding the expected delivery (considered as P0). The dams were randomly divided into two experimental groups: Own and Alien. This determines if, during the emotional contagion, the dam will be paired with an alien pup demonstrator or a pup of her own. All experimental procedures were preapproved by the Centrale Commissie Dierproeven of the Netherlands (AVD801002015105; AVD8010020209945) and by the welfare body of the NIN (NIN201102;NIN203802).
Figure 1. Timeline of the experiment from arrival till perfusion. The scale pictures represent a weighting point
6 Culling
To control for potentially confounding factors like size and composition of the litter, all litters were culled to 6 pups; 3 males and 3 females at P2. The procedure lasted less than 5 minutes per litter. The dams were placed in a separate cage and the sex of the pups was determined by anogenital distance. The pups were placed on a heat pad to maintain their body temperature during the procedure. The dam and selected pup (3 males and 3 females) were weighted and placed back together in the cage. To maximize mothers’ acceptance of the pups after culling and decrease the potential distress of the procedure, pups were covered with home-cage scient (dirty bedding) before returning to their mother. Surplus pups were anaesthetized by cold ice and quickly euthanized by decapitation.
Handling
To allow the animal getting used with the researchers, the dams and the female pups were handled for 5 consecutive days prior to the emotional contagion test. The female pups were later used for the demonstrator role in the emotional contagion test. Handling of the animals was performed each day from P11 till P15. The weight of the dams and the female pups was measured on P11 and P15. Over the days of handling, pups were handled in a pseudorandom order but the dam was always handled first. Starting the procedure of handling, the dam was put in a separate cage with bedding but no nesting material for 5 minutes. During this time apart, the male pups were separated in another cage with bedding material and a bit of nesting material from the original cage. After 5 minutes, the dam was handled for 5 minutes and was put in the cage together with the male pups. After the handling of the dam, the female pups were individually handled for 5 minutes each and put back together with the dam when finished. Since it is known that handling in the postnatal period could affect the maternal care behaviour post-reunion, video and ultrasonic vocalizations (USV) recordings were obtained during the whole procedure. After handling the dam and the female pups, there was a 10 minute maternal care video observation. Video tapes were made from the lateral side using the AXIS P1364 Network Camera. USVs were recorded by using the Avisoft UltraSoungGate 416H microphone and recordings were recorded using Avisoft-Recorder USGH.
7 Behaviour was scored frame by frame by using Behavioural Observation Research Interactive Software, BORIS (Friard et al., 2016). The maternal care recordings during the handling were made with 30 frames per minute and expressed as percentage of duration over the total duration of the observation. The maternal care was calculated by combining the pup directed behaviours licking and grooming, general nursing and arched back. Also rearing, self-grooming, eating, general moving, carrying, tail chasing and freezing were scored
Figure 2. The emotional contagion test apparatus. Observers are in the left chamber and demonstrators on the right
with the shock bars. Partition consists small holes to provide sound and smell exchange between chambers. Shocks were visually represented by the light on the top of the right chamber.
Habituation
Habituation of the animals to the testing apparatus happened from P17 till P19. The weight of the dams and paired female pups was measured on P17 and P19. During habituation, the dams were paired with the same juvenile as the emotional contagion. For the Own group, the dam was paired with a juvenile from their own litter and for the Alien group, the dam was paired with a juvenile from an alien litter. The dams will be the observers in the emotional contagion test and the juveniles the demonstrators. Habituation took place in the apparatus of the emotional contagion test for 10 minutes (figure 2). This apparatus consists of 2 chambers (each chamber L: 24 cm x W: 25 m x H: 34 cm, Med Associates) with transparent Plexiglas walls and stainless-steel grid rods connected to a stimulus scrambler (ENV-414S, Med Associates). The compartment of the observer had a plastic grid to exclude the shocks. In between the chambers was a transparent Plexiglas partition which was perforated with small holes to allow
8 animals in both chambers to see, smell and hear each other. The order for habituation was random between Own dam and Alien dam pairs. During these 10 minutes, behaviour and audible and ultrasonic vocalizations were reordered. Videos were made from both a lateral and top view with Basler GigE camera (acA1300–60gm). To make the recordings, the software of Media Recorder 25 (Noldus), was used. USVs were also recorded using the same microphone and program as during handling. After each round of habituation, the apparatus was cleaned with 70% alcohol.
Emotional contagion
On P20 the emotional contagion test was performed (figure 2). The emotional contagion test consists of a 12 minute baseline and a 12 minutes test period where 5 shocks (1 sec, 1.5 mA) are randomly given with 2 or 3 minutes interval. After the emotional contagion, the dams were individually put in a clean cage and 90 minutes after they were euthanized by sodium pentobarbital and perfused for ex-vivo analysis of c-Fos brain activation (data not shown). Shocked juveniles were weaned, put together in a new clean cage and put in surplus. The remaining juveniles were divided by sex, weaned and put in a clean cage and later reused for other approved experiments The same camera and microphone were used as for the habituation and the apparatus was cleaned with 70% alcohol between animals.
Behaviour was scored frame by frame by using Behavioural Observation Research Interactive Software, BORIS (Friard et al., 2016). The videos were taken with 25 frames per minute. It is unknown how dams react to observing a juvenile rat receiving foot shocks and it is expected that the own dam will exhibit other behaviours compared to the alien dam. Therefore not only freezing was scored frame by frame but also, exploring, rearing, partition rearing, self-grooming, digging, climbing, tail chasing, jumping/running, laying down and laying down backwards. The same ethogram was used to score both the observer dams and the demonstrator juvenile rats during the emotional contagion.
Estrous phase determination
One factor that is known that it could influence behaviour are hormones and in particular hormones that are involved in the estrous cycle. After giving birth, the estrus cycle of the dams does not go back to
9 normal immediately. Within 24 hours after parturition, the postpartum estrus usually occurs. When the dam is not conceived during this period, the lactational diestrus starts. She will stay in this phase until the pups reach a weaning age which is usually around 25 to 30 days postpartum (Carrillo-Martínez et al., 2011). After this period the normal estrus cycle starts again and this can influence behaviour. This happens especially during the proestrus phase where the progesterone levels are higher compared to the estrus and diestrus phase. To investigate whether the estrus phase of the dams could have impacted their behavioural response at the emotional contagion test, before killing the animals and harvesting the brains vaginal smears were collected by vaginal washes and placed on a glass. The estrous phase was determined by visual inspection of unstained samples under the microscope (10-20x) and after Giemsa staining.
Statistical analysis
All statistical calculations were made using JASP (JASP Team, 2020). Normality was tested for all the data with the Shapiro-Wilk test for normality. To compare the freezing data of the emotional
contagion (baseline vs shock period), the repeated measures ANOVA was used for both observer and demonstrator roles. Also, bodyweight comparisons were calculated using the repeated measures ANOVA. Other behaviours were tested with the independent sample t-test. Bayesian statistics were used to complement the classical null-hypothesis testing approach. Bayes Factor shows if there is evidence that including an effect worsens or improves a model. It allows to calculate whether there is evidence for or against an effect and also whether the data is inconclusive and thus indicate that sample size should be increased to enhance statistical power. BFincl values above 3 are considered moderate evidence that including the effect improves the model. On the other hand, BFincl below 1/3 is considered moderate evidence that including the effect worsens the model. Every BFincl value between 1/3 and 3 indicates that the data is still inconclusive . The Bayesian rmANOVA and Bayesian independent sample t-test were used to calculate the Bayesian statistics. For the analysis of the
10
Results
Normality
Figure 3 shows the freezing data during the emotional contagion test for both the observers and demonstrators of both groups. The results of the normality tests are shown in table 1-4. The test of normality for the freezing shows that the freezing data for the post shock period is for both groups (Alien/Own) and roles (Demo/Observer) normally distributed. For the baseline data, while the data are normally distributed for the own group and both groups of the observer role, the freezing data of the demonstrator alien group are not normally distributed (Shapiro-Wilk, W = 0.764, p < 0.05). Due to the non-normally distributed data in the baseline, parametric testing with the baseline included should be interpreted with caution and where possible supplemented with non-parametric tests.
Body weight
During the whole experiment (arrival to emotional contagion) there were no significant differences in the body weights of own and alien dams (rmANOVA, F = 1.191, p = 0.327)(figure 4). Body weight of the pups included in the emotional contagion was measured on habituation day 1 and 3. These pups were all from the own group and dividing them to the two groups for the emotional contagion did not create a group difference (rmANOVA, F = 1.135, p = 0.312).
Figure 3 Body weight of the dam during the whole experiment. P days referring to the post-natal days with P0 at
the day of delivery 7 days after arrival. Bodyweight in presented in grams. Error bars represent the standard deviation.
11
Emotional contagion test
Average freezing levels are lower in observers Own dams
Observers
The freezing levels during both the baseline and post period for both observers and demonstrators are shown in figure 3. For observers a 2 Group (own vs alien) X 2 Time (baseline vs shock) ANOVA revealed a significant main effect of Time (F = 45.328, p < 0.001, BFincl = 7928.622) and a trend for Group (F = 1.236, p = 0.292, BFincl = 2.848) and a trend towards an interaction effect (F = 4.168, p = 0.068, BFincl = 2.848). According to Jeffreys (1961) a Bayes Factor of 7928.622 indicates that there is extreme evidence for an effect of time. In this case the BFincl value of 2.848 indicates that there is a trend towards the effect where the observers of the own group freeze less compared to the observers of the alien group.
Demonstrators
For demonstrators a 2 Group (own vs alien) X 2 Time (baseline vs shock) ANOVA also reveals a significant main effect of Time (F = 62.207, p < 0.001, BFincl = 725723.473), a non significant trend for Group (F = 0.375, p = 0.554, BFincl = 0.817) and for interaction (F = 1.425, p = 0.260, BFincl = 1.960). This effect is similar to the observers where the own group freezes less than the alien group but the effect is less strong.
Figure 3. Freezing levels during the baseline and post shock period of the emotional contagion. Demo represents
12 Figure 4. Percentages of observer exhibited behaviours during the 12 minute baseline of the emotional
contagion. Error bars present the standard error of the mean.
No significant differences in other behaviours between Own and Alien observers
To create a complete picture of the displayed behaviours of Own and Alien mother during the emotional contagion test and investigate which behaviours were exhibited by the Own mother instead of freezing, we analysed ten other behaviours. The exhibition level of these behaviours during the baseline are shown in figure 4. The results of the normality tests, statistical tests and other descriptive results are shown in supplementary table 1 for the normally distributed data and in supplementary table 2 for the non-normally distributed data.
Figure 5. Percentages of observer exhibited behaviours during the 12 minute post shock period of the emotional
13 The data of the baseline are normally distributed for both groups for the following behaviours: rearing, partition rearing, self-grooming and exploring. For the climbing, tail chasing, laying down, laying down back and digging behaviours the data is for both groups not normally distributed As expected, there is no jump/run behaviour in both groups during the baseline (“free shock period”). During the baseline the alien dams show more partition rearing, self-grooming and exploring compared to the own group (Student, t(10) < 0.872, p > 0.404). The alien group also displayed more tail chasing compared to the own group (Mann-Whitney U = 23.5, p1 = p2 = 6, p = 0.34). On the other hand performs the own group more rearing (Student, t(10) = -1.180, p = 0.265), climbing, digging and laying down (Mann-Whitney U > 14, p1 = p2 = 6, p > 0.532).
The results for the post shock period are shown in figure 5, supplementary table 3 (normally distributed) and supplementary table 4 (non-normally distributed). The post shock data are normally distributed for the rearing, partition rearing, self-grooming and exploring behaviours for both groups However, the data for the partition rearing did not meet the assumption of the equal variance and thus the Mann-Whitney U test was used (Levene, p <0.05). The data for the climbing and the laying down is in both groups not normally distributed. To note, the jump/run, tail chasing and laying down back only occurred in the own group and this data was also not normally distributed (Shapiro-Wilk, W < 0.576, p = 0.001). The alien dams performed more self-grooming (Student, t(10) = 0.791, p = 0.447) compared to the own dams. On the other hand, the own dams showed more rearing and exploring even as more climbing, partition rearing, laying down and digging. However, these findings are not significant.
Time course of Oservers’ behavioral response during the post-shock session
To better understand the behavioral response of observer mothers in response to the distress of an own or alien juvenile demonstrator, we performed a qualitatively analysis of the behaviors over the time of the post-shock session. Given the random 2/3 minutes interval between the shocks delivered to the demonstrator, to compare results across observers, we restricted the analyses to the first 2 minutes after each shock.
14 Figure 6 Freezing, climbing, rearing and laying down divided in Alien (pink) and Own (blue). Percentage of
behaviours during the 2 minutes after each shock (S). Errorbars represent the standard error of the mean.
The results showed in figure 6 show us that the freezing behaviour changes between the groups after the third shock. Where the alien observers continue to increase the freezing, the freezing levels of the own observers stay around the same. On the other hand, we see that after the fourth shock, the own observers show more climbing and after the fifth shock also more rearing compared to the alien observers. Figure 6 also showed that there is a clear difference in the laying down behaviour. Where the alien observers don’t show any laying down behaviour until after the fifth shock, the own observers only do not express this behaviour after the second shock.
Influence of individual maternal care on emotional contagion
Maternal care behaviours were scored on the first day of handling during the handling procedure with an addition 10minute observations after replacement of the third pup. The pup directed behaviours like licking and grooming and nursing were taken together as a score of maternal care. The independent sample non parametric Mann-Whitney test showed that there was no significant difference between
15 the alien and the own group in terms of pup directed behaviour (Mann-Whitney, U = 29, p = 0.093). To test whether the level of maternal care can predict the level of emotional contagion, we correlated the level of pup directed behaviour with the freezing levels during the post shock period of the emotional contagion test. In particular, these pup-directed behaviour levels were mainly driven by licking and grooming. The results in figure 8 show that there is no correlation between licking and grooming levels and the freezing levels in the observers (t(10) = 1.6073, p = 0.14, cor = 0.45). However if we divide the analysis between the groups we can see that there is a positive correlation between the amount of licking and grooming and the level of freezing in the own group (t(4) = 6.9973, p = 0.002, cor = 0.96)(figure 9). This correlation is absent in the alien group (t(4) = -0.043345, p = 0.96, cor = -0.021).
Figure 7 Percentage pup direct behaviour during the maternal care observations on the first day of handling.
Figure 8. Correlation between licking and grooming behaviour during maternal care observations and post shock
16 Figure 9. Correlation between licking and grooming behaviour during maternal care observations and post shock
period freezing during the emotional contagion in observers. Alien observers and Own observers graphs divided. Alien pearsons p = -0.021, p =0.96. Own pearsons p = 0.96, p = 0.002. Alpha < 0.05
Estrous
The independent sample t-test showed that there is no difference in estrus phase between the dams of the own group and the dams of the alien group (Student, t(10)=-0.191, p = 0.852). If we add the estrus phase as a covariate to the ANCOVA, it showed that the estrus phase does not predict the freezing level (F = 0,010, p = 0.924). Also, the Bayesian ANCOVA shows that estrus does not benefits the model (BF10 = 0.332).
Discussion
The main goal of this research was investigating the role of kinship on the emotional contagion in rats. Since the used emotional contagion test was never used on juvenile animals before, it was needed to prove that shocks given to the juvenile demonstrators induced emotional contagion in the dam observers. Previous published data (Carillo et al., 2011) on Long Evans rats shows that vicarious freezing exhibited by observers while witnessing the demonstrator in distress are higher in pre-exposed observers compared to pre-pre-exposed. Although in this experiment, the animals were non-pre-exposed and it is also known that Sprague Dawley rats overall freeze less compared to Long Evans
17 (Carrillo-Martínez et al., 2011) we detected significant increases in freezing levels of the observers between baseline and post-shock periods. The significant increase in freezing between the baseline period and post shock period in the demonstrator animals, it showed that the shocks induced expressions of pain in juvenile animals. Besides that, the increased freezing levels in the post shock period compared to the baseline period in the observers showed that there was emotional contagion coming from the demonstrator animals. These results were supported by the Bayesian statistics, which shows extreme evidence for this effect. This proves that the emotional contagion test was successful in this experiment.
We did not show any significant differences in behaviour of the observers between groups. However, with the Bayesian statistics we were able to show that the data is still inconclusive and needs a bigger sample size. This was expected since power analysis showed earlier that at least 20 animals per group were needed to show an effect. Taken this into account, the results that show less freezing in the own group compared to the alien group have already a quite strong tendency towards an effect. We see similar results in the demonstrator groups but the effect is smaller compared to the observer results. This was expected since the reduction in freezing of the demonstrators was hypothesised to be caused by the higher social buffering from the own mother compared to the alien mother. Reduced freezing in the demonstrator is already showed to decrease freezing in the observer too (Han et al., 2020). Besides that, Rickenbacher and colleagues showed that mothers switch from self-defense behaviour to offspring protection in case of a threat. At a young age of the pups (post natal day (P)4-6), dams will suppress their freezing reactions to protect the offspring by displaying defensive behaviour towards the threat. On the other hand, when pups reach an age where they are mobile (P19-21), the dams will change the behaviour by increasing the display of contact with the pups. The increased contacted should prevent the pups to go towards the threat (Rickenbacher et al., 2017). If we combine the results of Han et al. (2020) and Rickenbacher et al. (2017), we would expect more reduced freezing behaviour in the own mother observers and more active behaviour to get to the juvenile demonstrator. These behaviours contain digging, rearing and climbing. Especially the results of the time course of the shocks show this difference. However, these results are still inconclusive and not significant. Besides that, there is also
18 an increase in laying down in the own observer group that is not seen in the alien observer group. This differs from the results from Rickenbacher, where they showed an inhibition of maternal freezing, where in presence of the pups, in favour to more active defensive behaviours. This can be due to the limited moving space that the observers had in our experiment and the fact that they couldn’t reach the juvenile demonstrator on the other side. It can therefore be hypothesised that when the observers of the own group after time realise that they cannot go to the pup, they switch to a more helpless behaviour. Further analysis of the behaviour after each shock needs to show this. Furthermore, more research needs to be done to clarify the meaning of the laying down behaviour.
We observe a trend for group differences in freezing during the emotional contagion. Kinship however involves both sharing genes and sustained postnatal mother-pup interactions. With the present setup of the experiments, we are not able to distinguish these components and therefore are not able to determine which component predicts the results on freezing better. To determine this, further research with cross-fostering procedure is needed.
Nevertheless, behaving in a different way in the presence of an own or other pup requires pup recognition. With this trend, we can suggest that the mothers are able to recognize there own pup. This is in line with literature of Beach and Jaynes (1956) where pup discrimination has been found in a retrieval task in rats. Only in mice it has been shown that this recognition is based on both odor and USVs (Bowers & Alexander, 1967; Mogi et al., 2017). In the end, it is very clear that auditory cues are very important in creating and having a mother-infant bond (Nagasawa et al., 2012). It is also known that the transmission of emotions in rats finds place via the expression of sound. These ultrasonic vocalizations (USVs) can have either a positive as a negative emotional meaning. Research showed that USVs of 22 kHz are often associated with negative contexts such as fear and thus can be seen as a distress call (Litvin et al., 2007). It has shown that the distress calls only occur when there is an audience (Blumstein et al., 1997). It is known that pups express different frequencies of USVs as alarm calls. A typical USV at 40 kHz is often measured when the pup is removed from the dam (Brudzynski et al., 1999). On the other hand, the pups will inhibit this USV when there is a potential threat. They will then
19 shift to an USV of 27 kHz which is thought to be related to the 22 kHz vocalizations of adult rats (Takahashi, 1992). Further analysis of the recruited USV data is needed to see what role the USVs play in this setup. To find out what causes the effects between the own and alien observer freezing levels more research has to be done. Not only a control with a virgin naïve observer and a juvenile demonstrator is needed, but also mother observer with an adult demonstrator. Besides that, to prove that the dams only switch in behaviour in presence of their own pup, there needs to be a setup where there is a mother observer with a potential threat but without the presence of a demonstrator.
The results on the maternal care showed that, only in the own group, the maternal care was positively correlated with the freezing levels in the observer. However, these results need to be interpreted with caution. The maternal care score was only taken from the first day of handling. Furthermore, the maternal care was interrupted because of the replacement of the female pups after handling. This could have been a big influence on the outcome of the behavioural data. Thereby, the data of the other days of handling need to be taken and analysed all together. Besides that, the data of the observations used in this research was obtained during the complete handling procedure. In the future, only the data from the last 10 minutes should be taken into account since there is no interruption of the maternal care in this period.
Results of the estrus analysis show that the estrus does not have any influence on the freezing behaviour outcome. However, these results are questionable. In this experiment the estrus was taken using plain water and no staining. In another experiment, the estrus phase was first determined with the plain water but also with the Giemsa staining. This showed that the determination of the estrus phase using plain water was often different. Besides that, Carrillo-Martinez et al. (2011) showed that the lactational diestrus phase usually ends around 25 to 30 days postpartum. In this experiment, the dams were only 21 days postpartum on the day that the estrus was obtained. This indicates that the results are not reliable and therefore these results cannot be concluded and further research is needed.
20 In conclusion, the results are still inconclusive and thereby can be concluded that there is emotional contagion between juvenile animals and adult dams but that the effect of kinship is still unknown. Further analysis and research are needed to determine the effect of kinship and the influence of other factors as motherhood and estrus phase.
21 References
Beach, F. A., & Jaynes, J. (1956). Studies of maternal retrieving in rats I: Recognition of young. Journal of
Mammalogy, 37(2), 177-180.
Blumstein, D. T., & Armitage, K. B. (1997). Alarm calling in yellow-bellied marmots: I. The meaning of situationally variable alarm calls. Animal Behaviour, 53(1), 143-171.
Bowers, J. M., & Alexander, B. K. (1967). Mice: individual recognition by olfactory cues. Science, 158(3805), 1208-1210.
Brudzynski, S. M., Kehoe, P., & Callahan, M. (1999). Sonographic structure of isolation‐induced ultrasonic calls of rat pups. Developmental Psychobiology: The Journal of the International Society for Developmental
Psychobiology, 34(3), 195-204.
Church, R. M. (1959). Emotional reactions of rats to the pain of others. Journal of comparative and
physiological psychology, 52(2), 132.
Carrillo-Martínez, G. E., Gómora-Arrati, P., González-Arenas, A., Morimoto, S., Camacho-Arroyo, I., & González-Flores, O. (2011). Role of progesterone receptors during postpartum estrus in rats. Hormones and
behavior, 59(1), 37-43.
Friard, O. Gamba M. BORIS: a free, versatile open-source event-logging software for video/audio coding and
live observations. Methods in Ecology and Evolution, 10.
Frye, C. A., Petralia, S. M., & Rhodes, M. E. (2000). Estrous cycle and sex differences in performance on anxiety tasks coincide with increases in hippocampal progesterone and 3α, 5α-THP. Pharmacology Biochemistry
and Behavior, 67(3), 587-596.
Gonzalez-Liencres, C., Juckel, G., Tas, C., Friebe, A., & Brüne, M. (2014). Emotional contagion in mice: the role of familiarity. Behavioural brain research, 263, 16-21.
Han, Y., Sichterman, B., Maria, C., Gazzola, V., & Keysers, C. (2020). Similar levels of emotional contagion in male and female rats. Scientific Reports, 10(1), 1-12.
22
Ishii, A., Kiyokawa, Y., Takeuchi, Y., & Mori, Y. (2016). Social buffering ameliorates conditioned fear responses in female rats. Hormones and Behavior, 81, 53-58.
JASP Team. (2020). JASP (Version 0.14.1) [Computer software]. https://jasp-stats.org/
Jeffreys, H. (1998). The theory of probability. OUP Oxford.
Kiyokawa, Y., Takeuchi, Y., & Mori, Y. (2007). Two types of social buffering differentially mitigate conditioned fear responses. European Journal of Neuroscience, 26(12), 3606-3613.
Kikusui, T., Winslow, J. T., & Mori, Y. (2006). Social buffering: relief from stress and anxiety. Philosophical
Transactions of the Royal Society B: Biological Sciences, 361(1476), 2215-2228.
Mogi, K., Takakuda, A., Tsukamoto, C., Ooyama, R., Okabe, S., Koshida, N., ... & Kikusui, T. (2017). Mutual mother-infant recognition in mice: The role of pup ultrasonic vocalizations. Behavioural brain research, 325, 138-146.
Nagasawa, M., Okabe, S., Mogi, K., & Kikusui, T. (2012). Oxytocin and mutual communication in mother-infant bonding. Frontiers in human neuroscience, 6, 31.
Litvin, Y., Blanchard, D. C., & Blanchard, R. J. (2007). Rat 22 kHz ultrasonic vocalizations as alarm cries. Behavioural brain research, 182(2), 166-172.
Panksepp, J. (2007). Neuroevolutionary sources of laughter and social joy: Modeling primal human laughter in laboratory rats. Behavioural brain research, 182(2), 231-244.
Rickenbacher, E., Perry, R. E., Sullivan, R. M., & Moita, M. A. (2017). Freezing suppression by oxytocin in central amygdala allows alternate defensive behaviours and mother-pup interactions. Elife, 6, e24080.
Takahashi, L. K. (1992). Developmental expression of defensive responses during exposure to conspecific adults in preweanling rats (Rattus norvegicus). Journal of Comparative Psychology, 106(1), 69.
de Waal, F. B. M. (2007). The 'Russian doll' model of empathy and imitation. In S. Bråten (Ed.), Advances
in consciousness research: Vol. 68. On being moved: From mirror neurons to empathy (p. 49–69). John
24 Supplementary materials
Supplementary Table 1. Descriptive statistical values of the normally distributed behavioural data of the observer dams during the baseline of the emotional contagion. The Shapiro-Wilk test was used for the normality statistics and the student version of the independent sample t-test was used for the student statistics. The alpha was put at 0.05.
Normality Mean
Std. Deviation
Minimum Maximum Student
W p t df p Baseline freezing Alien 0.890 0.318 7.883 5.975 0.900 15.000 0.463 10 0.653 Own 0.964 0.852 9.350 4.956 3.100 16.000 Baseline rearing Alien 0.960 0.819 17.567 5.090 10.900 24.500 -1.180 10 0.265 Own 0.902 0.385 22.183 8.123 13.400 35.600 Baseline partition rearing Alien 0.918 0.492 5.300 3.803 1.000 12.000 0.493 10 0.632 Own 0.941 0.668 4.467 1.629 2.600 7.100 Baseline self grooming Alien 0.876 0.253 28.300 10.906 11.000 39.000 0.493 10 0.633 Own 0.991 0.992 24.817 13.454 5.900 43.900 Baseline explore Alien 0.903 0.395 40.000 6.107 33.500 51.000 0.872 10 0.404 Own 0.896 0.352 36.433 7.948 23.000 44.000
25 Supplementary Table 2. Descriptive statistical values of the non-normally distributed behavioural data of the observer dams during the baseline of the emotional contagion. The Shapiro-Wilk test was used for the normality statistics and the student version of the independent sample t-test was used for thstudent statistics. The alpha was put at 0.05.
Normality Median IQR Minimum Maximum Mann-Whitney
W p U p Baseline climbing Alien 0.557 <0.001 0.000 0.075 0.000 1.000 14.00 0.532 Own 0.599 <0.001 0.350 0.775 0.000 7.200 Baseline tail chasing Alien 0.773 0.033 0.050 0.100 0.000 0.300 23.5 0.340 Own 0.496 <0.001 0.000 0.000 0.000 0.200 Baseline laying down Alien 0.496 <0.001 0.000 0.000 0.000 0.600 17.5 1.0 Own 0.496 <0.001 0.000 0.000 0.000 6.000 Baseline laying down back Alien 0.496 <0.001 0.000 0.000 0.000 3.000 18.5 1.0 Own 0.496 <0.001 0.000 0.000 0.000 2.400 Baseline digging Alien 0.496 <0.001 0.150 0.300 0.000 0.400 22.5 0.445 Own 0.781 0.039 0.000 0.000 0.000 1.000 Baseline jump/run Alien NaN NaN 0.000 0.000 0.000 0.000 NaN NaN Own NaN NaN 0.000 0.000 0.000 0.000
26 Supplementary Table 3. Descriptive statistical values of the normally distributed behavioural data of the observer dams during the post shock period of the emotional contagion. The Shapiro-Wilk test was used for the normality statistics and the student version of the independent sample t-test was used for the student statistics. The alpha was put at 0.05.
Normality Mean
Std. Deviation
Minimum Maximum Student
W p t df p Post shock freezing Alien 0.883 0.284 27.367 6.094 21.000 37.600 1.960 10 0.078 Own 0.800 0.058 19.767 7.287 7.200 25.000 Post shock rearing Alien 0.832 0.111 10.383 3.730 7.000 16.100 -0,074 10 0.942 Own 0.922 0.518 10.633 7.360 3.000 21.000 Post shock self grooming Alien 0.923 0.524 12.967 9.411 0.800 25.000 0.791 10 0.447 Own 0.961 0.831 9.700 3.703 4.300 15.000 Post shock explore Alien 0.875 0.245 44.183 6.154 36.300 50.800 -0,607 10 0.557 Own 0.911 0.443 47.267 10.811 31.000 58.000
27 Table 4. Descriptive statistical values of the non-normally distributed behavioural data of the observer dams
during the post shock period of the emotional contagion. The Shapiro-Wilk test was used for the normality statistics and the student version of the independent sample t-test was used for the student statistics. The alpha was put at 0.05
Normality Median IQR Minimum Maximum Mann Whitney
W p U p Post shock jump/run Alien NaN NaN 0.000 0.000 0.000 0.000 NaN NaN Own 0.496 <0.001 0.000 0.000 0.000 0.200 Post shock climbing Alien 0.496 <0.001 0.000 0.000 0.000 1.000 15.0 0.599 Own 0.496 <0.001 0.000 0.375 0.000 6.600 Post shock tail chasing Alien NaN NaN 0.000 0.000 0.000 0.000 NaN NaN Own 0.611 <0.001 0.000 0.150 0.000 1.000 Post shock partition rearing Alien 0.906 0.408 3.950 2.550 2.000 7.000 -0,139 0.892 Own 0.919 0.497 4.450 5.650 0.200 9.000 Post shock laying down Alien 0.496 <0.001 0.000 0.000 0.000 6.000 6.5 0.060 Own 0.739 0.016 1.900 10.525 0.000 14.000 Post shock laying down back Alien NaN NaN 0.000 0.000 0.000 0.000 NaN NaN Own 0.576 <0.001 0.000 0.300 0.000 3.000
28 Post shock digging Alien 0.666 0.003 0.000 0.225 0.000 0.900 17.5 1 Own 0.526 <0.001 0.000 0.225 0.000 6.300
