Université de Montréal, Montréal, Canada.
BMC Ecol Evol. 2024 Aug 30;24(1):116. doi: 10.1186/s12862-024-02302-6.
While most game theoretical models assume that individuals randomly interact with all other group members, strong evidence indicates that individuals tend to preferentially interact with some of them. The position of an individual in a network affects, among other factors related to survival, its predation risk and competitive success. Here I then modified the Hawk-Dove game to explore the effect of social network structure on competitive strategy of individuals that differ in their fighting ability and may adjust their use of the Hawk, Dove and Assessor tactics to maximize their foraging success when they meet opponents they are connected with.
From randomly generated networks, I demonstrate that phenotypic assortment by fighting ability reduces individuals' aggressiveness and, as such, favours cooperative interactions. Furthermore, the success of individuals with the weakest fighting ability is usually highest within networks where they most frequently meet opponents with the same fighting ability as their own, suggesting they might benefit from breaking connections with strong contestants. This might be the case when strong contestants systematically rely on the aggressive Hawk tactic or the risk of being predated is low and independent of the number of neighbours. Thus, I extended the model and built a dynamic model to allow individuals not only to adjust their behaviour to local conditions but also to modify the structure of the social network. The number of connections and degree of phenotypic assortment are then affected by ecological factors (e.g. resources value and predation risk), but above all by whether individuals can reliably assess the competitive ability of their opponents and adjust their behaviour accordingly.
These findings provide strong evidence that behaviour can play a key role in shaping network structure and highlight the importance of considering the coevolution of network and behaviour to apprehend its consequences on population dynamics.
虽然大多数博弈论模型假设个体随机与所有其他群体成员相互作用，但强有力的证据表明，个体倾向于优先与其中一些个体相互作用。个体在网络中的位置会影响其捕食风险和竞争成功等与生存相关的因素。在这里，我修改了鹰鸽博弈模型，以探讨社交网络结构对个体竞争策略的影响，这些个体在战斗能力上存在差异，并且可能会根据遇到的对手调整鹰、鸽和评估者策略的使用，以最大限度地提高其觅食成功率。
从随机生成的网络中，我证明了战斗能力的表型聚类会降低个体的攻击性，从而有利于合作性的相互作用。此外，在与自己战斗能力相同的对手最频繁相遇的网络中，个体中战斗能力最弱的个体的成功率通常最高，这表明他们可能受益于与强大的竞争者断开联系。如果强大的竞争者系统地依赖于攻击性的鹰策略，或者被捕食的风险低且不依赖于邻居的数量，那么这种情况就可能发生。因此，我扩展了模型，并构建了一个动态模型，允许个体不仅根据局部条件调整行为，还可以修改社交网络的结构。连接的数量和表型聚类的程度受到生态因素（例如资源价值和捕食风险）的影响，但最重要的是，个体是否能够可靠地评估对手的竞争能力并相应地调整行为。
这些发现有力地证明了行为在塑造网络结构方面起着关键作用，并强调了考虑网络和行为的共同进化以理解其对种群动态的影响的重要性。
