Social network analysis resolves temporal dynamics of male dominance relationships

Social organization is often studied through point estimates of individual association or interaction patterns, which does not account for temporal changes in the course of familiarization processes and the establishment of social dominance. Here, we present new insights on short-term temporal dynamics in social organization of mixed-sex groups that have the potential to affect sexual selection patterns. Using the live-bearing Atlantic molly (Poecilia mexicana), a species with pronounced male size polymorphism, we investigated social network dynamics of mixed sex experimental groups consisting of eight females and three different-sized males over a period of 5 days. Analyzing association-based social networks as well as direct measures of spatial proximity, we found that large males tended to monopolize most females, while excluding small- and medium-bodied males from access to females. This effect, however, emerged only gradually over time, and different-sized males had equal access to females on day 1 as well as day 2, though to a lesser extent. In this highly aggressive species with strong social dominance stratifications, the observed temporal dynamics in male-female association patterns may balance the presumed reproductive skew among differentially competitive male phenotypes when social structures are unstable (i.e., when individual turnover rates are moderate to high). Ultimately, our results point toward context-dependent sexual selection arising from temporal shifts in social organization.     

The influence of refuge sharing on social behaviour in the lizard Tiliqua rugosa

Refuge sharing by otherwise solitary individuals during periods of inactivity is an integral part of social behaviour and has been suggested to be the precursor to more complex social behaviour. We compared social association patterns of active versus inactive sheltering individuals in the social Australian sleepy lizard, Tiliqua rugosa, to empirically test the hypothesis that refuge sharing facilitates social associations while individuals are active. We fitted 18 neighbouring lizards with Global Positioning System (GPS) recorders to continuously monitor social associations among all individuals, based on location records taken every 10 min for 3 months. Based on these spatial data, we constructed three weighted, undirected social networks. Two networks were based on empirical association data (one for active and one for inactive lizards in their refuges), and a third null model network was based on hypothetical random refuge sharing. We found patterns opposite to the predictions of our hypothesis. Most importantly, association strength was higher in active than in inactive sheltering lizards. That is, individual lizards were more likely to associate with other lizards while active than while inactive and in shelters. Thus, refuge sharing did not lead to increased frequencies of social associations while lizards were active, and we did not find any evidence that refuge sharing was a precursor to sleepy lizard social behaviour. Our study of an unusually social reptile provides both quantitative data on the relationship between refuge sharing and social associations during periods of activity and further insights into the evolution of social behaviour in vertebrates.     

Unraveling fission-fusion dynamics: how subgroup properties and dyadic interactions influence individual decisions

Many species show fission-fusion group dynamics because it has clear advantages for flexibly exploiting heterogeneous environments. However, the mechanisms by which these dynamics arise are not well known. We used a hierarchical Bayesian model to disentangle the different influences on spider monkey (Ateles geoffroyi) individual fissions and fusions, including the three dimensions of fission-fusion dynamics (subgroup size, dispersion, and composition). Furthermore, we considered the influences of other individuals also leaving or joining a subgroup at the same time. We found that the most important influence on individual fissions and fusions is whether other individuals are also doing the same. Subgroup size and dispersion did not have clear effects on the probability that an individual fissioned or fusioned, while individuals tended to leave subgroups that were biased toward the opposite sex and to join subgroups that were biased toward their own sex. The networks constructed by the inter-individual influences during fissions and fusions were cohesive and did not show assortativity by sex or by degree. Individuals had a similar degree in both networks and each was influenced by a different set of individuals, suggesting a high fluidity in the social networks. We suggest that these networks reflect the way in which information about the environment flows as individuals follow one another during fissions and fusions.     

Comparative social network analysis in a leaf-roosting bat

Even though social network analysis provides an important tool to characterize and compare societies, no studies have used its analytical applications to characterize patterns of sociality in bats. Here I use social network analysis to characterize and compare patterns of sociality between three populations of the leaf-roosting bat Thyroptera tricolor. Sites differed in the density of furled leaves used by T. tricolor for roosting. Finca had more leaves per hectare (77), followed by Ureña (58), and Esquinas (7). The time period over which the probability of association is halved based on fitted models was 1,086 days for Finca, 714 days for Ureña, and 303 days for Esquinas. Finca and Ureña had very similar network topologies, with several small clusters, high-clustering coefficients, short path lengths, low node betweenness, and high network robustness. Social networks at Esquinas were composed of one large cluster and several small isolated ones. Esquinas also had high-clustering coefficients, but path length and node betweenness were high. Network resilience was lower in Esquinas compared to Finca and Ureña. These results show that, unlike many other forest-dwelling bats that switch roosts regularly, T. tricolor does not exhibit a typical fission–fusion social system, and that resource availability seems to affect social networks in this bat. In addition, this study highlights the importance of emigrating individuals in maintaining social cohesion, establishing network connectedness, and determining network robustness.     

Testing the generalization of artificial neural networks with cross-validation and independent-validation in modelling rice tillering dynamics

Neural networks (NN) rely on the inner structure of available data sets rather than on comprehension of the modeled processes between inputs and outputs. Therefore, neural networks have been regarded as highly empirical models with limited extrapolation capability to situations outside the range of the training and validation data sets. In this study, the generalization ability of neural networks in predicting rice tillering dynamics was tested and several techniques inducing the generalization ability of neural networks were compared. We compared the performance of cross-validated neural networks with independent-validated neural networks and found that neural networks were able to extrapolate and predict tillering dynamics if the data were within the range of inputs of the training set. An inadequate training set resulted in overfitting of available data and neural networks that were not generalizable. The training set size required to enable a neural network to generalize and predict rice tillering dynamics was found to be at least 9 times as many training patterns for each weight. When a large number of variables are included in the input vector of a neural network with inadequate amounts of training data, we strongly recommend that the dimension of the input vector is reduced using principle component analysis (PCA), correspondence analysis (CA) or similar techniques to decrease the number of weights before the training procedure to improve the generalization ability of the NN. If the amount of training data still is not sufficient after the dimension of the input vector is reduced, regularization techniques, such as early stopping, jittering, and especially the embedment of estimated results by a theoretical model into the training set, should be used to improve the generalization ability of the neural network. The generalization of neural networks presents a wide spectrum of problems, and the proposed approaches are not confined strictly to modelling rice tillering dynamics but can be applied to other agricultural and ecological systems.     

Male–female socio-spatial relationships and reproduction in wild chimpanzees

Recent research on primates and other taxa has shown that the relationships individuals form with members of the same sex affect their reproductive success. Evidence showing that intersexual relationships also influence reproduction, however, is more equivocal. Here, we show that male chimpanzees living in an exceptionally large community display long-term tendencies to associate with particular females. These association patterns are likely to arise because individuals of both sexes selectively range in specific areas of the communal territory, with males inheriting the ranging patterns of their mothers. These differentiated male–female socio-spatial relationships involved males of widely varying ranks, and their effect on reproduction is as strong as that of male dominance rank, which in turn is as strong a predictor of reproductive success at Ngogo as in other smaller chimpanzee communities. These results show that male–female socio-spatial relationships can play a large role in chimpanzee male reproductive strategies, although they probably neither weaken nor strengthen the relationship between male dominance rank and reproductive success. Our results linking male–female socio-spatial relationships to reproduction in chimpanzees suggest that the gap between the social and mating systems of humans and their closest living relatives may not be as large as previously thought.     

The social network structure of a wild meerkat population: 1. Inter-group interactions

Groups of individuals frequently interact with each other, but typically analysis of such interactions is restricted to isolated dyads. Social network analysis (SNA) provides a method of analysing polyadic interactions and is used to analyse interactions between individuals. We use a population of 12 groups (ca. 250 animals) of wild meerkats (Suricata suricatta) to test whether SNA can also be used to describe and elucidate patterns of inter-group interactions. Using data collected over 24 months, we constructed two sets of networks, based on direct encounters between groups and instances of roving males visiting other groups. We analysed replicated networks of each type of interaction to investigate similarities between networks of different social interactions as well as testing their stability over time. The two network types were similar to each other when derived from long-term data, but showed significant differences in structure over shorter timescales where they varied according to seasonal and ecological conditions. Networks for both types of inter-group interaction constructed from data collected over 3 months reliably described long-term (12- and 24-month) patterns of interactions between groups, indicating a stable social structure despite variation in group sizes and sex ratios over time. The centrality of each meerkat group in roving interactions networks was unaffected by the sex ratio of its members, indicating that male meerkats preferentially visit geographically close groups rather than those containing most females. Indeed, the strongest predictors of network structure were spatial factors, suggesting that, in contrast to analyses of intra-group interactions, analyses of inter-group interactions using SNA must take spatial factors into account.     

Female gregariousness in Western Chimpanzees (Pan troglodytes verus) is influenced by resource aggregation and the number of females in estrus

Among social animals, group size is constrained by competition over resources. Because female reproductive success is limited by access to food resources, and that of males by access to fertile females, chimpanzee females are proposed to be less social than males and to maintain weak intrasexual relations. Findings from Taï National Park, Côte d’Ivoire, challenged this view, as chimpanzee females were described as generally gregarious, and close intrasexual bonds were common. Here, in a new analysis that focuses on the South Group of chimpanzees in Taï forest, we reevaluate the proposed differences in female association patterns between the Taï and East African populations. We find that mean party size and dyadic association index between females has decreased in Taï, although the level of dyadic associations remains high compared with East African chimpanzees. We attribute the decrease in female gregariousness to the decline in community size over the last 10 years. In addition, we use a multivariate approach to analyze social and ecological factors influencing party size in females. We show that female gregariousness increased when the fruit resources were more clumped and with increased number of females in estrus present. Party size of mothers with sons, however, was smaller with increasing number of sexually receptive females. The results of our model and the reviewed findings of other studies support the socioecological model because food distribution affects female gregariousness, but social and demographic aspects are equally influencing female grouping tendencies.     

The social network structure of a wild meerkat population: 2. Intragroup interactions

Knowledge of the structure of networks of social interactions is important for understanding the evolution of cooperation, transmission of disease, and patterns of social learning, yet little is known of how environmental, ecological, or behavioural factors relate to such structures within groups. We observed grooming, dominance, and foraging competition interactions in eight groups of wild meerkats (Suricata suricatta) and constructed interaction networks for each behaviour. We investigated relationships between networks for different social interactions and explored how group attributes (size and sex ratio), individual attributes (tenure of dominants), and ecological factors (ectoparasite load) are related to variation in network structure. Network structures varied within a group according to interaction type. Further, network structure varied predictably with group attributes, individual attributes, and ecological factors. Networks became less dense as group size increased suggesting that individuals were limited in their number of partners. Groups with more established dominant females were more egalitarian in their grooming and foraging competition interactions, but more despotic in their dominance interactions. The distribution of individuals receiving grooming became more skewed at higher parasite loads, but more equitable at low parasite loads. We conclude that the pattern of interactions between members of meerkat groups is not consistent between groups but instead depends on general attributes of the group, the influence of specific individuals within the group, and ecological factors acting on group members. We suggest that the variation observed in interaction patterns between members of meerkat groups may have fitness consequences both for individual group members and the group itself.     

Simulation of information propagation in real-life primate networks: longevity, fecundity, fidelity

In many vertebrate species, we find temporally stable traditions of socially learned behaviors. The social structure of animal populations is highly diverse and it has been proposed that differences in the social organization influence the patterns of information propagation. Here, we provide results of a simulation study of information propagation on real-life social networks of 70 primate groups comprising 30 different species. We found that models that include the social structure of a group differ significantly from those that assume random associations of individuals. Information spreads slower in the structured groups than in the well-mixed groups. While we found only a minor effect on the path lengths of the transmission chains, robustness against information extinction was strongly influenced by the group structure. Interestingly, robustness against information loss was not correlated with propagation speed but could be predicted reasonably well by relative strength assortativity—a structural network metric. In those groups where highly pro-social individuals preferentially interact with other pro-social individuals, information was more likely to be lost. Our results show that incorporating group structure in any social propagation model significantly alters predictions for spreading patterns, speed, and robustness of information.     

Social networks and models for collective motion in animals

The theory of collective motion and the study of animal social networks have, each individually, received much attention. Currently, most models of collective motion do not consider social network structure. The implications for considering collective motion and social networks together are likely to be important. Social networks could determine how populations move in, split up into and form separate groups (social networks affecting collective motion). Conversely, collective movement could change the structure of social networks by creating social ties that did not exist previously and maintaining existing ties (collective motion affecting social networks). Thus, there is a need to combine the two areas of research and examine the relationship between network structure and collective motion. Here, we review different modelling approaches that combine social network structures and collective motion. Although many of these models have not been developed with ecology in mind, they present a current context in which a biologically relevant theory can be developed. We argue that future models in ecology should take inspiration from empirical observations and consider different mechanisms of how social preferences could be expressed in collectively moving animal groups.     

Association patterns and kinship in female Indo-Pacific bottlenose dolphins (Tursiops aduncus) of southeastern Australia

Kinship has been shown to be an important correlate of group membership and associations among many female mammals. In this study, we investigate association patterns in female Indo-Pacific bottlenose dolphins (Tursiops aduncus) inhabiting an embayment in southeastern Australia. We combine the behavioral data with microsatellite DNA and mitochondrial DNA data to test the hypotheses that genetic relatedness and maternal kinship correlate with associations and social clusters. Mean association between females was not significantly different from a random mean, but the standard deviation was significantly higher than a random standard deviation, indicating the presence of nonrandom associates in the dataset. A neighbor-joining tree, based on the distance of associations between females, identified four main social clusters in the area. Mean genetic relatedness between pairs of frequent female associates was significantly higher than that between pairs of infrequent associates. There was also a significant correlation between mtDNA haplotype sharing and the degree of female association. However, the mean genetic relatedness of female pairs within and between social clusters and the proportion of female pairs with the same and different mtDNA haplotypes within and between clusters were not significantly different. This study demonstrates that kinship correlates with associations among female bottlenose dolphins, but that kinship relations are not necessarily a prerequisite for membership in social clusters. We hypothesize that different forces acting on female bottlenose dolphin sociality appear to promote the formation of flexible groups which include both kin and nonkin.     

Determining association networks in social animals: choosing spatial–temporal criteria and sampling rates

Social Network Analysis has become an important methodological tool for advancing our understanding of human and animal group behaviour. However, researchers tend to rely on arbitrary distance and time measures when defining ‘contacts’ or ‘associations’ between individuals based on preliminary observation. Otherwise, criteria are chosen on the basis of the communication range of sensor devices (e.g. bluetooth communication ranges) or the sampling frequencies of collection devices (e.g. Global Positioning System devices). Thus, researchers lack an established protocol for determining both relevant association distances and minimum sampling rates required to accurately represent the network structure under investigation. In this paper, we demonstrate how researchers can use experimental and statistical methods to establish spatial and temporal association patterns and thus correctly characterise social networks in both time and space. To do this, we first perform a mixing experiment with Merino sheep (Ovis aries) and use a community detection algorithm that allows us to identify the spatial and temporal distance at which we can best identify clusters of previously familiar sheep. This turns out to be within 2–3 m of each other for at least 3 min. We then calculate the network graph entropy rate—a measure of ease of spreading of information (e.g. a disease) in a network—to determine the minimum sampling rate required to capture the variability observed in our sheep networks during distinct activity phases. Our results indicate the need for sampling intervals of less than a minute apart. The tools that we employ are versatile and could be applied to a wide range of species and social network datasets, thus allowing an increase in both the accuracy and efficiency of data collection when exploring spatial association patterns in gregarious species.     

Social relationships among adult female baboons (Papio cynocephalus) II. Variation in the quality and stability of social bonds

A growing body of evidence suggests that social bonds have adaptive value for animals that live in social groups. Although these findings suggest that natural selection may favor the ability to cultivate and sustain social bonds, we know very little about the factors that influence the quality or stability of social bonds. Here, we draw on data derived from a 16-year study of baboons living in seven different social groups in the Amboseli basin of Kenya to evaluate the quality and stability of social bonds among females. Our results extend previous analyses, which demonstrate that females form the strongest bonds with close maternal and paternal kin, age mates (who may be paternal kin), and females who occupy similar ranks but are not maternal relatives. Here we show that the same factors influence the quality and strength of social bonds. Moreover, the results demonstrate that the quality of social bonds directly affects their stability.     

Spatial organization and behaviour of badgers <Emphasis Type="Italic">(Meles meles)</Emphasis> in a moderate-density population

Badgers are carnivores that show considerable variation in their social and spatial organization. At the westernmost part of their range, in Britain and Ireland, variation in spatial organization appears to be determined by the availability of resources. However, the majority of studies has focussed at one end of the social/spatial spectrum, where population densities are high and adjacent territories are contiguous and non-overlapping. To examine whether the same limiting factors appear to apply across a wider range of badger densities, we established a study site in a predominantly coniferous habitat within an upland area of northeast England, where population densities were predicted to be low. Seasonal home ranges of individual badgers were largest in autumn, followed by summer and spring, then winter. This pattern is reflective of the likely seasonal changes in food availability within the area, as opposed to being related to breeding patterns. There were also significant correlations between territory size and the number of grassland patches (positive) and the proportion of grassland (negative), which are consistent with predictions from the Resource Dispersion Hypothesis. Although badgers at the site were living at low to moderate densities relative to many other studied populations in Britain, they showed patterns of spatial organization that were close to those of high-density populations. The nature of the relationship between resource availability and abundance patterns is likely to have important consequences for the conservation and management of badgers and other species that show flexible spatial organization.     

The Role of Research in Evaluating Conservation Strategies in Tanzania: the Case of the Katavi-Rukwa Ecosystem

Abstract:  Strict protectionism, resource extraction, protected-area community outreach, ecotourism, an integrated conservation and development program, comanagement schemes, and citizen-science initiatives are all being used to help conserve the remote Katavi-Rukwa ecosystem in western Tanzania. Biological and social research show that protectionism is successful in the conservation of large mammals but fails to capture diverse species communities; extractivism is appropriate for some resources but not for others; protected-area outreach can be effective for some communities; and devolved control over wildlife, in conjunction with ecotourism and citizen science, has considerable potential in the area. The long-term nature of the research provides the necessary time frame to evaluate outcomes of different conservation strategies, uncovers dynamics within communities that affect attitudes and responses to conservation initiatives, provides impartial recommendations because changing research personnel offers different viewpoints, and, probably most importantly, enhances trust among stakeholders. Currently, there are limited institutional mechanisms for ensuring the input of biological and social science in shaping conservation practice in Tanzania, and long-term research can help informally bridge the gap     

Marine aquaculture and bottlenose dolphins’ (<Emphasis Type="Italic">Tursiops truncatus</Emphasis>) social structure

In this study, we investigate association patterns of 249 bottlenose dolphin feeding groups off Sardinia Island (Italy) from January 2000–May 2007 and describe how their association behaviour is related to their response to food patches created by a marine fin fish farm. We also tested the hypothesis that dolphins have different social structures with different feeding activities: Associations should decrease during opportunistic feeding behaviours as it is easier to capture prey, and cooperation is not as necessary. Sixteen individually identified bottlenose dolphins were observed participating in both opportunistic and not opportunistic feeding activities, with a mean of 30 ± 8 times and 9.6 ± 1 times, respectively. Bottlenose dolphins show non-random social behaviour during feeding and this behaviour differs depending on their specific foraging activity. Dolphin associations during feeding can be divided into three categories: acquaintances, affiliates, and feeding associates. Association behaviour during fish farm feeding is consistent with our hypothesis that during opportunistic behaviours, benefits from cooperation decrease, as it is easier to capture prey. Group size homogeneity in both feeding activities demonstrates that the number of dolphins engaging in foraging is not necessarily related with cooperation levels. Moreover, an adult dolphin may prefer to associate with a specific individual, independent of the sex, who shares the same foraging priorities. This study is the first to show how aquaculture is not only directly affecting marine predators but could also indirectly affect their social structure and behaviour.