We’re constantly looking to expand the boundaries of knowledge about collective behaviour. We aim to solve specific problems while uncovering fundamental principles that connect different domains.
We answer ecological questions from a new perspective, pioneering the shift to computational models and virtual experiments to overcome time, experimental and cost constraints.
Our work in computational ecology revolves around the behaviour of social insect colonies. This includes the foraging and pollination patterns of bees, the colony organisation of ants and the impact of environmental change on their societies.
Beyond scientific interest, our work can improve management and farming methods – and enhance conservation. For example, good models for bee foraging patterns are crucial for better pollination methods. Deciphering how environmental stressors change insect colony behaviour is also key to understanding colony collapse disorder.
We are working closely with experimental biologists, applied ecologists and the agricultural industry to expand our understanding of insect behaviour and improve agricultural and environmental management.
Social insects are the most widely used model system to study how coordinated group-level action can emerge from individual actions without central control. One core stream of our research targets the mystery of task allocation: How does a society adequately allocate its workforce resources to all the different tasks it faces without central coordination? Insights drawn from this can also help elucidate human behaviour.
To understand and increase cooperation in the world, we analyse social phenomena through simulation and mathematical theory. We collaborate with social scientists and psychologists to create and investigate realistic models based on cutting-edge computational modelling techniques.
We are particularly interested in social dilemmas with real-world relevance, such as embracing sustainable behaviour to slow climate change.
Individuals find sustainable behaviour costly. If they choose not to contribute, they can often still benefit from the investment of others – a logic that makes it impossible to facilitate change if individuals are driven by primarily selfish motivations. We see similar social dilemmas in technological systems such as Bitcoin, Open Source Software and Crowdsourcing.
We study the dynamics of these dilemmas to identify mechanisms that can align individuals with collective goals so better societal outcomes can be achieved.
One of the fastest-growing areas of AI, Multi-agent systems (MAS) are a very general modelling and problem solving concept that captures situations in which a number of autonomously acting agents interact with each other and their environment. It’s a paradigm to understand many complex natural phenomena, such as the behaviour of ant colonies, fish swarms and human groups.
Conversely, MAS approaches are crucial in designing some of the most cutting-edge AI and cyber-physical systems, such as swarm robots. Hybrid cyber-physical systems, in which natural and artificial agents interact, represent the third important form of MAS. The internet, where millions of humans and computational agents interact in intricate and complex ways, is the paradigmatic example of such a hybrid.
A variety of rich mathematical theories, such as game theory and reinforcement learning, can be used to describe and analyse MASs, but extensive agent-based simulation studies are often the method of choice when complex interactions have to be captured.