Effects of Mother’s Dominance Hierarchy on the Development of Social Relationships among Immature Tibetan Macaques

Simple Summary In this study, we explored the role of females’ social ranks on social behaviors among immature Tibetan macaques (Macaca thibetana). The results suggest that females’ social ranks affected their offspring’s social play and grooming during the juvenile and adolescent periods, but not the infancy period. The present study provides new insight into understanding the effects of the female dominance hierarchy on the development of social relationships among immature offspring in nonhuman primates. Abstract During a relatively long period of growth, immature individuals rely on their mothers to obtain nutrition, and a good environment for learning social skills needed to cope with complex environments in adulthood. In this study, we collected the behavioral data of Tibetan macaques (Macaca thibetana) to investigate the effects of females’ social rank on the development of social relationships among their immature offspring from November to June 2021. The results show that there was no difference in the rate/type of social play and grooming among infants. However, among juveniles and adolescents, the higher their mother’s social rank, the higher the rate of social play they participated in, and the more aggressive play they engaged in. Immatures with high-ranking mothers initiated more social play among each other. A similar pattern of playmates was found among juveniles/adolescents with middle-ranking and low-ranking mothers. We also found that immatures preferred immatures with higher-ranking mothers as grooming mates and initiated more grooming with immatures with higher-ranking mothers than with those with lower-ranking mothers. Our study suggests that females’ social ranks affect the development of social relationships among their immature offspring. In despotic nonhuman primates, this indicates that the mother’s dominance hierarchy would directly or indirectly influence the processes of participating in social interactions and choosing partnerships among immature individuals with age (i.e., infancy, juvenile, and adolescent periods).


Introduction
In nonhuman primates, immature individuals experience a relatively long period of growth and development [1]. During this period, immatures are vulnerable and completely incapable of independent survival [2]. Thus, they must rely on their mothers to provide nutrition and a social environment to participate in social interactions for promoting individual development and form social relationships to cope with a complex society in adulthood [3,4].
Social play and social grooming have been documented as the main and typical social interactive behaviors during the immature periods in nonhuman primates [5]. The main in social play was 12.5% in the daily activity budgets during the immature periods [28], which increased during the infant period (peaked at approximately 12 months of age) and decreased during the juvenile period [28]. The rate of intrasexual grooming increased with age [28].
In this study, we divided the immatures into three age classes (i.e., infants, juveniles, and adolescents) to investigate the effects of the mother's dominance hierarchy on the processes of participating in social interactions and choosing partnerships by comparing the social play and social grooming in Tibetan macaques. The following predictions were tested. If the mothers' dominance hierarchy positively affects the social interactions of their immature offspring, we predicted that immatures with higher-ranking mothers would engage in more social play and grooming than those with lower-ranking mothers. If the mothers' dominance hierarchy influences the partnership of their immatures, we predicted that the more similar the immatures' mothers' social ranks, the more social play they initiated with each other. However, the immatures with lower-ranking mothers were more likely to groom the immatures with higher-ranking mothers than vice versa.

Study Site and Subjects
This study was conducted at Mt. Huangshan, China. Several groups of Tibetan macaques are found throughout the reserve [29]. Beginning in 1987, we initiated observations of a single macaque study group (Yulinkeng A1, YA1). The study group was provisioned four times per day (a total of ca. 6 kg of dried corn per day) at the tourist viewing area during 2004 and 2014. Since 2014, the viewing site has been closed to tourists, and thus the subjects are wildlife.

Data Collection
Using a digital video camera (model: SONY HDR-CX680) and a digital voice recorder (model: Lenovo B618), we collected behavioral data during November 2020 and June 2021, according to Xia et al. [30]. We collected all the data when the monkeys were in the natural forest to reduce the potential influence of human activity. We used the focal animal sampling method and continuous recording method to collect behavioral data of social play and social grooming [31]. Each focal sampling period was set as 10 min. When the focal monkey could not be followed or disappeared from view during the sampling period, we randomly selected another individual [18,32]. We recorded the behavior of the disappeared individual in the next 10 min of sampling time [33][34][35]. A total of 123 h of data (mean ± SE = 3.70 ± 0.21 h; n = 33; range: 1.00 ± 4.50 total h per monkey) was collected during the focal sampling period. In addition, we used an ad libitum approach to record aggressive and submissive interactions of adult females for determining dominant relationships.

Dominance Hierarchy
We calculated individual David's scores for dominance hierarchies of adult females based on a dyadic aggressive/submission matrix, which showed the direction of agonistic interactions given and received. We used k-means cluster analysis (100 iterations) to classify each adult female into the following rank classes: high-ranking (YXX, YH, YCY, YXY), middle-ranking (YCH, TXH, YCL, TH), and low-ranking (TXX, TQL, HH, THY, THX, HXW) (see Table 1). In addition, we defined nine categories of mother-mother dyads based on rank classes (H-H: dyads consisting of two high-ranking mothers; H-M and M-H: dyads consisting of one high-and one middle-ranking mother; H-L and L-H: dyads consisting of one high-and one low-ranking mother; M-M: dyads consisting of two middle-ranking mothers; M-L and L-M: dyads consisting of one middle-and one low-ranking mother; and L-L: dyads consisting of two low-ranking mothers, with the former as initiator and the latter as receiver).
Aggressive behavior was defined as one individual threatening, chasing, slapping, grabbing, or biting another individual [36]. Submissive behavior was defined as an individual showing fearful behaviors, such as a fear grin, cower, mock leave, avoid, flee, or scream, as defined by Berman et al. [27].

Behavioral Definition
According to Burghardt's identification criteria of social play, it was feasible to distinguish the difference between social play and non-play behaviors. Social play was defined as behaviors functioned to develop, practice, or maintain physical or cognitive abilities and social relationships, including tactics by varying, repeating, or recombining already functional sub-sequences of behavior outside their primary context [28]. We categorized them into aggressive play (rough pattern) and affiliative play (soft pattern) [28]. The aggressive play is usually in the form of individual slapping, wrestling, gently biting and chasing. The affiliative play is usually in the form of individual mounting, cuddling, and sucking a penis. The detailed definition of social play can be found in Wang et al. [28]. For data collection, a play episode began when two or more individuals came into direct contact and engaged in slapping or other play behaviors (see [28]). A play episode ended when all players started to initiate non-play activities (e.g., resting, social grooming, and travel) or withdrew from the episodes, or interference from adults resulted in all players stopping their play activities [37].
Social grooming was defined as any act in which an individual (groomer) used their hand or mouth to touch, clean, or manipulate the fur of another individual (groomee) for a continuous period lasting at least 5 s [30,38]. The detail of the recording rules can be found in Xia et al. [30,39].

Data Analysis
We report all data as the mean ± SE for the rate of social play (episodes/h), the percentage of aggressive and affiliative play, the rate of social grooming (bouts/h), the percentage of grooming initiated and grooming received, and the percentage of initiated play and grooming subject selections. They were calculated as individual behavioral data for each immature per month that a focal subject participated in social play or social grooming. We used a one-sample Kolmogorov-Smirnov test to examine the normality of the data (p > 0.05).
We used a Kruskal-Wallis H test to analyze differences in the rate of social play and social grooming engaged in by infant, juvenile, and adolescent offspring of high-, middle-, and low-ranking females. We also used a Kruskal-Wallis H test to determine differences in social play type and social grooming engaged in by infant, juvenile, and adolescent offspring of high-, middle-, and low-ranking females. We used a Mann-Whitney U test to compare them in pairs. In addition, we used a Kruskal-Wallis H test to determine the difference in initiated social play/grooming among infant, juvenile, and adolescent offspring with high-, middle-, and low-ranking mothers. We used a two-tailed test with the alpha set at 0.05. All the analyses were performed using the SPSS 26.0 software (SPSS Inc., Chicago, IL, USA).

Discussion
To our knowledge, this is the first study to investigate the effects of the mother's dominance hierarchy on the processes of participating in social interactions and choosing partnerships by comparing the social play and grooming among infant, juvenile, and adolescent Tibetan macaques. We did not find a difference in social play and grooming between infants with high-, middle-, and low-ranking mothers. Juveniles and adolescents with high-ranking mothers engaged in more social play than those with lower-ranking mothers. The greater the similarity of their mothers' social rank, the more social play they initiated with each other. Although there was no difference in grooming participated in by juveniles/adolescents with high-, middle-, and low-ranking mothers, juveniles and adolescents with high-ranking mothers initiated less or received more social grooming than those with middle-and low-ranking mothers. Furthermore, immatures preferred individuals with higher-ranking mothers as grooming mates. Our study suggests that the mother's dominance hierarchy affects social play and grooming during the juvenile and adolescent periods, but not the infancy period. This indicates that the mother's dominance hierarchy plays a varied role in the growth environment and the socialization processes among immature individuals in different growth stages (i.e., infancy, juvenile, and adolescent periods).
Our results show that, in the despotic Tibetan macaque society, the females' dominance hierarchy has no effect on the social play and grooming of their infant offspring. However, a previous study found that high-ranking females are more laissez-faire towards their infants and low-ranking females are more protective of their infants in more despotic rhesus monkeys (Macaca mulatta) [40]. Thus, infants with high-ranking mothers engage in more social interaction, while infants with low-ranking mothers engage in less social interaction [20,40]. However, we have no data to show whether high-ranking or low-ranking mothers are rejective or protective in Tibetan macaques. Wright et al. reported that female Tibetan macaques rarely interrupted their infant offspring's social interactions in the same group [41]. This result suggests that Tibetan macaques adopt a mothering style characterized by low restrictiveness contrary to the expectation for more despotic species (such as rhesus monkeys [40]). Therefore, the different ranks of mothers Figure 6. The relationships between grooming partners and the mothers' social rank categories among immature individuals (H: high-ranking mother, M: middle-ranking mother, L: low-ranking mother).

Discussion
To our knowledge, this is the first study to investigate the effects of the mother's dominance hierarchy on the processes of participating in social interactions and choosing partnerships by comparing the social play and grooming among infant, juvenile, and adolescent Tibetan macaques. We did not find a difference in social play and grooming between infants with high-, middle-, and low-ranking mothers. Juveniles and adolescents with high-ranking mothers engaged in more social play than those with lower-ranking mothers. The greater the similarity of their mothers' social rank, the more social play they initiated with each other. Although there was no difference in grooming participated in by juveniles/adolescents with high-, middle-, and low-ranking mothers, juveniles and adolescents with high-ranking mothers initiated less or received more social grooming than those with middle-and low-ranking mothers. Furthermore, immatures preferred individuals with higher-ranking mothers as grooming mates. Our study suggests that the mother's dominance hierarchy affects social play and grooming during the juvenile and adolescent periods, but not the infancy period. This indicates that the mother's dominance hierarchy plays a varied role in the growth environment and the socialization processes among immature individuals in different growth stages (i.e., infancy, juvenile, and adolescent periods).
Our results show that, in the despotic Tibetan macaque society, the females' dominance hierarchy has no effect on the social play and grooming of their infant offspring. However, a previous study found that high-ranking females are more laissez-faire towards their infants and low-ranking females are more protective of their infants in more despotic rhesus monkeys (Macaca mulatta) [40]. Thus, infants with high-ranking mothers engage in more social interaction, while infants with low-ranking mothers engage in less social interaction [20,40]. However, we have no data to show whether high-ranking or lowranking mothers are rejective or protective in Tibetan macaques. Wright et al. reported that female Tibetan macaques rarely interrupted their infant offspring's social interactions in the same group [41]. This result suggests that Tibetan macaques adopt a mothering style characterized by low restrictiveness contrary to the expectation for more despotic species (such as rhesus monkeys [40]). Therefore, the different ranks of mothers may show the same parenting style for their offspring during the infancy periods. Similar results were also found in other primate species. For example, in a study of wild golden snubnosed monkeys (Rhinopithecus roxellana), Li et al. found that the mother's social rank did not influence the social play among infants [21]. Accordingly, our results provide more evidence that the difference in the effect of the mother's dominance hierarchy between species is related to the difference in the species-typical social structure and mother's parenting style.
Our results imply that the juvenile and adolescent periods might be critical stages for immature individuals to learn or exercise social skills needed in adulthood. Juveniles and adolescents need to do more than survive for nutrition [4]. Thus, they need to prepare for better integration into the social group or for forming social relationships as adults [4]. As a result, they become more independent, more likely to explore the environment, and participate in social interactions (social play and social grooming) [20,40]. Meanwhile, the tolerance of group members would decrease when the immature individuals are in the juvenile and adolescent periods [18]. As such, juveniles and adolescents must learn and understand the social structure of the group. Additionally, the characteristics of each partner interact to adjust their fitness to better engage in social interaction and choose social partners to cope with the complex environment in future adulthood.
Our results highlight that the effects of the mother's dominance hierarchy on immature individuals' socialization appeared in Tibetan macaques when their offspring were in the juvenile and adolescent periods. Previous studies have proposed that the higherranking mothers can provide a better environment (such as protection and aggressive support) for their offspring to grow up in, making their offspring bolder and allowing them to engage in more social interactions [42]. In this study, we found that juveniles and adolescents with higher-ranking mothers engaged in a higher rate of social play than those with lower-ranking mothers. Moreover, juveniles and adolescents with higher-ranking mothers engaged in more aggressive play than those with lower-ranking mothers. A similar pattern was also found in studies of juvenile lowland gorillas (Gorilla gorilla gorilla) [43] and juvenile vervet monkeys (Cereopitheeus aethiops) [44]. This indicates that the mother's dominance hierarchy affects the rate and the type of social play. Juvenile and adolescent Tibetan macaques of higher-ranking mothers would have more chances to participate in social play, and to establish and form social relationships in the future. They would be likely to participate in aggressive play to rehearse the ability to fight seriously later with conspecifics (the moto training hypothesis [45]). This social play pattern could be particularly fruitful in providing immediate or long-term feedback on the physical skills of developing individuals [46]. For example, for male immature Tibetan macaques, Wang et al. found that aggressive play facilitates the skills to integrate into a new social group when they mature [28]. Accordingly, our results suggest the mother's dominance hierarchy might affect future integration for dispersed male immatures and strategies to obtain a higher social rank in a natal group for philopatric female immatures.
In addition, since social play and aggressive play could be effective for individuals, the coming question is how to choose partnership. Our results show that the greater the similarity of the mothers' social rank among immature dyads, the more social play they initiated with each other. This suggests that social play is helpful for immatures to choose a well-matched partner according to their mothers' dominance hierarchy. In order to play without incurring severe aggression or injury, immatures must know all the characteristics of each immature individual and the social ranks of their mothers. This tendency is evident in many primate species (baboons, Papio cynocephalus [47,48]; rhesus macaques, Macaca mulatta [49]; chimpanzees, Pan troglodytes [50,51]; gorillas, Gorilla gorilla gorilla [52]). Previous studies proposed that symmetry characterizing play sessions among matched partners allows playmates to compete, practice, and strategize in a safer context [53]. Immatures of higher-ranking mothers grow faster and have greater physical strength, whereas immatures of lower-ranking mothers grow slower and have lower physical strength [24]. Accordingly, two playmates whose mothers are of a similar social rank might be fit enough to play together in order to test and improve their strength and motor ability due to the symmetry characterizing the two playmates [54].
Among juvenile and adolescent dyads, the present study also found that the effects of the dominance hierarchy on grooming patterns were similar to those among adult dyads. In nonhuman primates, previous studies demonstrated that lower-ranking adults initiated more social grooming with higher-ranking adults than vice versa (e.g., mandrills, Mandrillus sphinx [55]; Tibetan macaques, Macaca thibetana [30,39]). Social grooming is used to establish and maintain social relationships among immature individuals, which provides a better basis for group identification in adulthood [56]. However, less is known about the patterns of similarity between immature individuals and adults through social transmission whereby offspring learn the behavior from observing their mother or others [57,58]. Our results suggest that patterns of affiliative interactions (social grooming) between individuals begin to emerge in the juvenile and adolescent periods, not in adulthood.

Conclusions
The dominance hierarchy of adult females influences the rate of social interaction, the type or the number of social behaviors, and the social interaction partnerships among immature individuals. The present study provides new insight into understanding the effects of the mother's dominance hierarchy on the development of social relationships among immature nonhuman primates. Future studies would be necessary to further explore the influence of maternal styles (i.e., protectiveness or rejection), the sex-specific strategies of male or female offspring, and the long-term effects of the female dominance hierarchy on the social strategies of immature individuals (i.e., mother's social rank, social relationships, and reproductive success in the future).

Data Availability Statement:
The data that support the findings of this study are available on request from the corresponding author; the data are not publicly available due to privacy or ethical restrictions.