No abstract

Vries, 1995; also see Methods). Although they are not identical, the two indices generally yield the same values (see Methods; de Vries 1995) describing the degree to which an animal group adheres to a strict linear hierarchy. Thus, these indices are commonly referred to as metrics of 'linearity'. A major limitation of linearity indices is that they become biased when some pairs of individuals fail to interact (de Vries, 1995; Klass and Cords, 2011). This bias arises in part because strict linear hierarchies can occur only in groups where dominant-subordinate relations exist for every dyad in the group. Networks in which all dyadic relations are asymmetrical (e.g. dominant-subordinate) are called 'tournaments' (Harary & Moser 1966). However, tournaments rarely arise naturally in animal groups, and as we will show, unknown dyadic relationships are common because certain pairs of individuals fail to interact (hereafter 'null dyads'). Currently popular

"Across-year social stability shapes network structure in wintering migrant sparrows" (2014 IntroductionThe social structure of animal populations-e.g. the size, composition and stability of social groups-is a fundamental aspect of social evolution (Alexander 1974). In birds, studies of breeding systems have shown that ecological conditions can favor different social structures ranging from simple pairs to cooperative breeding groups (Emlen 1982). The winter social structure of year-round resident birds has also been investigated, but to a lesser degree than for the breeding season (Ekman 1989;Kraaijeveld & Dickinson 2001;Aplin et al. 2012). In migratory birds, the most basic aspect of winter social structure is known for many species-e.g. territoriality versus flocking in social groups. However, in species that form flocks (defined here as temporary aggregations of individuals in the same place at the same time), we know almost nothing about dynamics of flock membership over space and time (see Myers 1983;Piper & Wiley 1990;Conklin & Colwell 2008 for notable exceptions).Our lack of understanding of the winter societies of smallbodied birds is particularly surprising because these taxa were so crucial to the development of important theories in ecology. A large body of influential research on small-bodied birds in winter explored how food, predation and sociality interact to affect the evolution of optimal foraging (Stephens & Krebs 1986), sociality and optimal group size (Pulliam & Caraco 1984), energy management (Cuthill & Houston 1997), predator-prey interactions (Bertram 1978) and status signals (Rohwer 1975;Rohwer & Ewald 1981). For many of these topics, the pattern of group stability and the specific identities of group members matter. For example, the degree to which individuals form long-term associations could alter the dynamics of anti-predator behaviors and the form of cooperation involved Micheletta et al. 2012). In addition, the pattern of social structure also has critical implications for the mechanisms by which intragroup competition is mediated by signals (Rohwer 1975).In theory, the social structure of wintering birds could range from the small, highly stable groups observed in a variety of year-round resident birds (e.g. Ekman 1989) to short-term random associations with little or no structure (Myers 1983;Conklin & Colwell 2008). Between these two extremes, winter bird societies could also involve a complex mix of social stability and change in both space and timeoften termed fission-fusion dynamics (Aureli et al. 2008). Migration poses an added challenge to across-year stability because individuals that winter together do not necessarily breed together (Ryder et al. 2011;Seavy et al. 2012), and thus long-term social bonds must bridge a break in contact between winter seasons. However, high levels of site fidel- Published in Ecology Letters AbstractMigratory birds often form flocks on their wintering grounds, but important details of social structure such as the patterns of association betwe...

Demographic processes play a key role in shaping the patterns of social relations among individuals in a population. Social network analysis is a powerful quantitative tool for assessing the social structure formed by associations between individuals. However, demographic processes are rarely accounted for in such analyses. Here, we summarize how the structure of animal social networks is shaped by the joint effects of social behavior and turnover of individuals and suggest how a deeper understanding of these processes can open new, exciting avenues for research. Death or dispersal can have the direct effect of removing an individual and all its social connections, and can also have indirect effects, spurring changes in the distribution of social connections between remaining individuals. Recruitment and integration of juveniles and immigrant into existing social networks are critical to the emergence and persistence of social network structure. Together, these behavioral responses to loss and gain of social partners may impact how societies respond to seasonal or catastrophic turnover events. The fitness consequences of social position (e.g., survival and reproductive rates) may also create feedback between the social network structure and demography. Understanding how social structure changes in response to turnover of individuals requires further integration between long-term field studies and network modeling methods. These efforts will likely yield new insights into the connections between social networks and life history, ecological change, and evolutionary dynamics.