CCSS Meeting #83: The Illusion of Control: Intervention in Complex Adaptive Systems

This lecture will be held in physical format at the CCSS Living Room (Min. 4.16) with lunch and refreshments provided. The overarching topic of the CCSS Lunch Meetings of the academic year 2026/2027 is Interventions in Complex Systems.

Speaker Overview
Peter Sloot is (emeritus) professor Complex Adaptive Systems at the University of Amsterdam and at the Nanyang University in Singapore and ITMO in St. Petersburg. His interest is in understanding the way nature processes information by means of computational modeling of complex systems at large, with applications in biomedicine, complex networks and the dynamics of social interactions.

Lecture Overview
Advances in experimental technologies—including interactive detectors, sensors and scanners—together with increasingly powerful methods for integrating heterogeneous data, have opened new vistas onto physical, biological and social processes across multiple levels of organisation. The resulting systems span many orders of magnitude in spatial and temporal scale and display a rich variety of phenomena, including order and disorder, self-organisation, adaptation, collapse and self-destruction.

The interacting components of these systems form intricate, multi-layered networks in which individual elements are coupled nonlinearly to selected interaction partners. Such systems are not merely complicated; they are complex. Understanding, quantifying and intervening in this complexity is one of the major scientific challenges of our time.

Complex systems are notoriously difficult to control. Interventions frequently produce unexpected or unintended consequences because the systems themselves respond, adapt and reorganise. Their behaviour is shaped by nonlinear interactions, feedback loops, path dependence, partial observability and processes operating across multiple scales. In many cases, computational irreducibility and intractability further limit our ability to predict the consequences of an intervention in advance.

In this talk, I will present a personal perspective on complex systems and discuss how computational modelling can help identify causal mechanisms, explore intervention strategies and assess whether a system can be nudged towards preferred regions of its state space. Examples will be drawn from biomedicine, ecosystems and criminal networks. Together, they illustrate both the potential and the limitations of attempting to control complex adaptive systems. 

There will be 45-min lecture from the speaker, followed by a 15-min discussion session.

To attend the lecture, please signup below before 15:00 on Wednesday 16 September.