The interplay between social networks and culture: theoretically and among whales and dolphins

Abstract

Culture is increasingly being understood as a driver of mammalian phenotypes. Defined as group-specific behaviour transmitted by social learning, culture is shaped by social structure. However, culture can itself affect social structure if individuals preferentially interact with others whose behaviour is similar, or cultural symbols are used to mark groups. Using network formalism, this interplay can be depicted by the coevolution of nodes and edges together with the coevolution of network topology and transmission patterns. We review attempts to model the links between the spread, persistence and diversity of culture and the network topology of non-human societies. We illustrate these processes using cetaceans. The spread of socially learned begging behaviour within a population of bottlenose dolphins followed the topology of the social network, as did the evolution of the song of the humpback whale between breeding areas. In three bottlenose dolphin populations, individuals preferentially associated with animals using the same socially learned foraging behaviour. Homogeneous behaviour within the tight, nearly permanent social structures of the large matrilineal whales seems to result from transmission bias, with cultural symbols marking social structures. We recommend the integration of studies of culture and society in species for which social learning is an important determinant of behaviour.

Figure 1.

Two representations of the dynamic relationship between social structure and culture. (a) Individual characteristics ultimately influence both social structure, through their effects on social relationships (i), and the cultural context, through variation in both behaviour and the individuals’ partialities for social learning (ii). Overall, the interplay between individuals and social relationships influences—and is influenced by—the interplay between social structure and information transmission. (b) This is represented by a coevolutionary social network in which the coevolution of nodes (circles, with different shading representing individuals with different behavioural repertoires) and edges (links, with thickness being proportional to the rate of social interaction) (vii,viii) shapes and is shaped by (xi,xii) the coevolution of network topology and transmission mechanisms (ix,x). Thick dashed arrows illustrate the Hinde's [14] conceptual framework for social structure (i). Thick black arrows illustrate the elements of the concept of culture from Laland & Hoppitt [12] (ii). Thin arrows (iii,vi) represent additional effects hypothesized in this review. For further details on network terminology, see the electronic supplementary material, table S1.

Figure 2.

How social networks affect information transmission at two structural scales: the large-scale structure of the population and the structure within social modules. In both, the lower the connectance of the network, the longer the path length; thus, more time is required for the information flow, which makes the information more susceptible to loss and transcription errors but more prone to generate diversity. Arrows represent the overall directions of effects of network topology on network properties and on the transmission of information (described by their respective metrics) as indicated by the theoretical literature [–18,46,49]. Arrows in parentheses represent our own speculations. Up arrows indicate a positive relationship and down arrows a negative relationship. In the hypothetical networks, nodes representing individuals are connected by weighted edges whose thickness is proportional to the rate of social interaction, assumed to be proportional to probability of social learning. Efficiency was measured by the number of steps until all individuals acquired the new information (speed) [16]; consistency was measured by the average path length (minimum number of steps along a chain of relationships from one individual to another), reasonably assuming that longer paths are more likely to be subjected to transcription errors [16]; persistence over time was assessed by simulating the forgetting of acquired information and estimating its extinction risk [16]; and diversity was measured by the standard deviation of continuous behavioural measures or the Shannon diversity index for categorical behaviour [18]. See the electronic supplementary material, table S1 for definitions and interpretation of the network terminology; network metrics formulae can be found elsewhere [15,19,50,51].

Figure 3.

Humpback whale song in the Pacific. Principal breeding grounds are shown by star symbols in the North Pacific (dark grey), South Pacific (light grey) and Indian Ocean (black). Seasonal migration routes are indicated by dashed lines, and routes of information flow by thick arrows. The evolution of the South Pacific song between 1998 and 2008 is shown by the block diagram (adapted from [61]). The different song types are indicated by different colours and missing data by white boxes. The vertical columns of the block diagram are aligned approximately above the study areas where the songs were recorded in the South Pacific map.

Figure 4.

(a) Hypothetical effects of behaviour matching (i.e. when individuals tend to associate with those with similar behaviour) on social structure as influenced by patterns of behavioural specialization. The columns indicate different distributions of individual behaviour within the populations: unspecialized niches with similar widths; individual variation in niche width and location; individual variation in niche width around a common central value; and specializations around several modal values with similar niche widths. The first row represents the distribution of continuous behaviour (thicker lines, population behavioural repertoires; thinner lines, individual repertoires); the second row uses binary two-mode networks to represent equivalent distributions of categorical behaviour types (squares) used by individuals (white circles, ‘generalists’; grey circles, ‘specialists’) and the third row represents the weighted social network that behaviour matching might induce in each case (individuals connect by edges whose thicknesses are proportional to the rate of social interaction). (b) Adding conformism, behavioural repertoires become narrower. With low and moderate individual specialization, the social networks tend to random topologies, but when modules of individuals with specialized behaviour are present, conformism increases their isolation. Definitions and interpretation of network terms are available in the electronic supplementary material, table S1.