Neural dynamics of social behavior

Abstract

<p>Social behavior can be found on almost every level of life, ranging from microorganisms to human societies. However, explaining the evolutionary emergence of cooperation, communication, or competition still challenges modern biology. The most common approaches to this problem are based on game-theoretic models. The problem is that these models often assume fixed and limited rules and actions that individual agents can choose from, which excludes the dynamical nature of the mechanisms that underlie the behavior of living systems. So far, there exists a lack of convincing modeling approaches to investigate the emergence of social behavior from a mechanistic and evolutionary perspective.<br /> Instead of studying animals, the methodology employed in this thesis combines several aspects from alternative approaches to study behavior in a rather novel way. Robotic models are considered as individual agents which are controlled by recurrent neural networks representing non-linear dynamical system. The topology and parameters of these networks are evolved following an open-ended evolution approach, that is, individuals are not evaluated on high-level goals or optimized for specific functions. Instead, agents compete for limited resources to enhance their chance of survival. Further, there is no restriction with respect to how individuals interact with their environment or with each other.<br /> As its main objective, this thesis aims at a complementary approach for studying not only the evolution, but also the mechanisms of basic forms of communication. For this purpose it can be shown that a robot does not necessarily have to be as complex as a human, not even as complex as a bacterium. The strength of this approach is that it deals with rather simple, yet complete and situated systems, facing similar real world problems as animals do, such as sensory noise or dynamically changing environments.<br /> The experimental part of this thesis is substantiated in a five-part examination. First, self-organized aggregation patterns are discussed. Second, the advantages of evolving decentralized control with respect to behavioral robustness and flexibility is demonstrated. Third, it is shown that only minimalistic local acoustic communication is required to coordinate the behavior of large groups. This is followed by investigations of the evolutionary emergence of communication. Finally, it is shown how already evolved communicative behavior changes during further evolution when a population is confronted with competition about limited environmental resources. All presented experiments entail thorough analysis of the dynamical mechanisms that underlie evolved communication systems, which has not been done so far in the context of cooperative behavior. This framework leads to a better understanding of the relation between intrinsic neurodynamics and observable agent-environment interactions.<br /> The results discussed here provide a new perspective on the evolution of cooperation because they deal with aspects largely neglected in traditional approaches, aspects such as embodiment, situatedness, and the dynamical nature of the mechanisms that underlie behavior. For the first time, it can be demonstrated how noise influences specific signaling strategies and that versatile dynamics of very small-scale neural networks embedded in sensory-motor feedback loops give rise to sophisticated forms of communication such as signal coordination, cooperative intraspecific communication, and, most intriguingly, aggressive interspecific signaling. Further, the results demonstrate the development of counteractive niche construction based on a modification of communication strategies which generates an evolutionary feedback resulting in an active reduction of selection pressure, which has not been shown so far. Thus, the novel findings presented here strongly support the complementary nature of robotic experiments to study the evolution and mechanisms of communication and cooperation.</p>