Network reaction norms: taking into account network position and network plasticity in response to environmental change

Abstract

6Recent studies have highlighted the link between consistent inter-individual differences 1 7 in behaviour and consistency in social network position. There is also evidence that network 1 8 structures can show temporal dynamics, suggesting that consistency in social network position 1 9 3 1 behaviour. 3 2 3 3 Introduction: 3 4Consistent inter-individual differences in behaviour are common in animal groups [1, 2]. 5How such behavioural differences develop, are maintained, and selected form the focus of 3 6 current research efforts, as do questions relating to the ecological and evolutionary consequences 3 7of such inter-individual differences [3, 4]. There is also great interest in the extent to which 3 8 animals are able to vary their behaviour in response to environmental changes, and whether and 3 9 how this behavioural plasticity co-varies with inter-individual differences in mean behavioural 4 0rates. This is captured by the notion of a "behavioural reaction norm" (BRN)-the set of 4 1 behavioural phenotypes an individual produces in a given set of environments-a concept drawn 4 2 from life history theory [5, 6]. Analytically, a BRN can be identified using multi-level regression 4 3 models, where the fitting of random intercepts captures inter-individual differences in the 4 4 expression of a given behaviour, and random slopes capture individual differences in plasticity 4 5 across an environmental gradient. Thus, BRNs provide information on how animals differ both 4 6 in their mean level of a given behavioural trait, and in how strongly they respond to 4 7 environmental variation. 4 8 Within social groups, patterns of inter-individual consistency have also been observed in 4 9 social network position (e.g., central individuals remain central across time) [7-11]. Again, it is 5 0an open question as to how these differences arise and are maintained [12]. There is also 5 1 evidence to suggest that consistent inter-individual differences in behaviour and social network 5 2 position interact [13, 14], which implies either that, for social animals, network position reflects 5 3 certain behavioural predispositions, and/or that certain behavioural predispositions may arise in 5 4 7 1 whether some animals are more consistently central -measured as eigenvector centrality -over 7 2 time (i.e., inter-individual differences in intercepts); (ii) the extent to which individuals on 7 3 average respond to social and environmental changes (i.e., mean network plasticity in the 7 4 population/group); (iii) the extent to which individuals differ from this mean response (i.e., 7 5inter-individual differences in network plasticity or variation in slope); and finally, (iv) to 7 6 quantify the relationship between inter-individual differences in mean network position and 7 7 network plasticity (i.e., covariation between intercepts and slopes), e.g., do more (or less) central 7 8 individuals display a stronger (or weaker) response to environmental changes? 7 9 Methods: 8 0 Networks through time: 8 1 Social networks are often built on...

Within-individual phenotypic plasticity in flowers fosters pollination niche shift

“…As the dependent variable, we used a composite variable including all the five floral traits analysed independently in the previous analyses. We scaled and centred both the independent and dependent variables and used weakly informative prior with normal distribution and centred on zero 90 . We ran four chains with 2000 iteration each, burning 1000 samples per chain.…”

Constructing and analysing time‐aggregated networks: The role of bootstrapping, permutation and simulation

“…In some cases, these theoretical and empirical concerns are temporal in nature, requiring some measure of how social networks change in time, i.e. viewing networks as dynamic rather than static (Aplin et al, 2015; Bonnell, Vilette, Henzi, & Barrett, 2019; Formica, Wood, Cook, & Brodie III, 2017). For example, the magnitude of repeatability in social network position has direct implications for the selection of social phenotypes within a population (Aplin et al, 2015).…”