Agent-based model using adjacent possible space for generating waves of novelty

The continuous emergence of novelty is key to the creation of open-endedness. The phenomenon of the waves of popularity of novelties which generates novelties, and leads to their attracting attention and evolving, has been observed in groups of people. The exploration of this mechanism can provide clues to understanding the nature of open-ended evolution. It has been demonstrated that the behavior of people within this context can be successfully captured by an agent-based model that incorporates the adjacent possible space. In this study, we propose an extended model on this basis to create waves of novelty. The proposed model introduces a bias in that it increases the selection probability of agents who interact with influential agents. The introduction of this bias leads to the novelty gaining attention over preferential attachment. The proposed model fits better to real data than existing models, and captures the dynamics and structure of real social networks more effectively. The construction of speciations and niches is necessary for natural biological species to evolve. Similarly, the results of this study suggest that the provision of an advantageous structure in a particular local environment is important for the growth of novelty.