Collective Information Processing Romanczuk Lab

Research

The main focus of our group is the development and analysis of individual-based mathematical models of collective behaviour in biological systems, ranging from cellular ensembles to animal collectives. Our overarching goal is to understand how self-organised collective dynamics emerge from interactions among individuals, and how collective structure and behaviour relate to collective function, information processing, and evolutionary adaptation.

Our work on (bio)physical systems of active particles focuses on the emergence of complex, large-scale collective dynamics from idealised physical interactions. In our work on animal collectives, we explicitly incorporate sensory perception at the individual level as a key proximate mechanism underlying social interactions. We combine mathematical modelling and methods from statistical physics and stochastic processes with simplified models of sensory and behavioural processes and empirical data from experimental collaborators. This approach enables us to develop mechanistic models that connect individual perception and behaviour to the collective dynamics of the groups they form. Our theoretical work combines mathematical analysis and numerical simulations, while close interaction with experimental research provides empirical grounding and opportunities to test model predictions.

Building on this framework, we investigate how the constraints and capabilities of individual organisms shape collective behaviour. We are particularly interested in how groups acquire, process, and distribute information; how collective decisions and coordinated responses emerge from local interactions; and under which conditions collective organisation provides functional advantages. These questions are studied across different biological systems, including insects, fish, and birds, with a broader interest in the principles governing collective intelligence and the evolutionary dynamics of collective behaviour.

Our research is primarily theoretical, but we work closely with experimental groups to connect mathematical models with observations of real biological systems. These collaborations allow us to formulate models that are directly informed by empirical data and to confront theoretical predictions with experimental evidence. Among our collaborators is the Krause Lab in the Department of Fish Biology and Ecology at the Leibniz Institute for Freshwater Ecology and Inland Fisheries (IGB) in Berlin.

Our main research activity focuses on the following interconnected research questions:

  1. Do schools, flocks & crowds – from cellular aggregates to animal groups (incl. humans) – share common functional features as collective information processing system?
  2. What is the interplay between the self-organised structure and dynamics of collective systems and their collectively intelligent (or not so intelligent) behaviour?
  3. How does collective behaviour emerge from sensory inputs? Here, we focus on question on how individuals integrate visual social information and use it to coordinate their actions.
  4. How do animal groups perform collective decisions under sensory constraints? Here, the central question is how animal groups are able to perform near-optimal and robust collective decisions despite sensory constraints and highly variable environments.

Ongoing/Recent projects

citation image
[no title info]
[no publisher info]  ·  [no date info]  ·  [no id info]

–>