Self-Organisation Lab

Stuart P. Wilson | University of Sheffield UK

The aim of our research is to understand how self-organisation and natural selection interact to shape complex systems, such as brains. To this end, we construct mathematical and computational models of adaptive self-organising networks. Our research has focused on two model systems; i) self-organising neural network models of the development of topological maps in sensory and motor cortex, and ii) self-organising models of the evolution and development of collective behaviour in animal groups. Our recent research focuses on the interplay between these two systems, exploring how self-organising interactions between developing animals shape, and in turn are shaped by, self-organising networks in developing brains. We are particularly interested in the interplay between brain and behaviour in the evolution of social and cognitive systems.

What is self-organisation?

''Self-organisation describes a dynamic in a system whereby local interactions between individuals collectively yield global order, i.e., spatial patterns unobservable in their entirety to the individuals. By this working definition, self-organisation is intimately related to chaos, i.e., global order in the dynamics of deterministic systems that are locally unpredictable. A useful distinction is that a small perturbation to a chaotic system causes a large deviation in its trajectory, i.e., the butterfly effect, whereas self-organising patterns are robust to noise and perturbation. For many, self-organisation is as important to the understanding of biological processes as natural selection. For some, self-organisation explains where the complex forms that compete for survival in the natural world originate from.''
Wilson (2017), Living Machines Handbook.

Self-organisation as a theory of cortical development

''Understanding how external stimuli are represented in the brain is one of the central questions of neuroscience. Researchers have attacked this issue on many fronts, but perhaps most directly by studying cortical maps: the response properties and organisation of the entire set of neurons comprising a cortical area. Cortical maps provide important clues about how brains form and maintain representations of the external world. [Computational models] demonstrate how maps of the world can emerge from self-organising principles, that is, how maps emerge from individually simple interactions between neurons, without plan or instruction. The success of these models in accounting for a wealth of experimental data on cortical maps motivates self-organisation as an important theory of cortical maps, with no other type of model currently able to account for this range of observations.''
Bednar & Wilson (2015), The Neuroscientist.

Self-organising thermoregulatory huddling

''Huddling is an adaptive behavior that emerges from simple interactions between animals. Huddling is a particularly important self-organising system because the behavior that emerges at the level of the group directly impacts the fitness of the individual. The huddle insulates the group, allowing pups to thermoregulate at a reduced metabolic cost, however a huddle can only self-organise if pups in turn contribute heat. Contributing too much heat is costly but contributing too little compromises the ability of the huddle to self-organise. [Evolutionary simulations of] the resulting tension between co-operation and competition in the huddle [...result...] in the emergence of a phenomenon called self-organised criticality. Criticality is a hallmark of complex systems [...] The model therefore reveals how complexity can emerge in a well-defined biological system (thermoregulation), where experiments can be designed to investigate the interaction between self-organisation and natural selection.''
Wilson (2017), PLoS Computational Biology.

How self-organisation can guide evolution

''Self-organisation and natural selection are fundamental forces that shape the natural world. Substantial progress in understanding how these forces interact has been made through the study of abstract models. Further progress may be made by identifying a model system in which the interaction between self-organisation and selection can be investigated empirically. To this end, we investigate how the self-organising thermoregulatory huddling behaviours displayed by many species of mammals might influence natural selection of the genetic components of metabolism. By applying [simple evolutionary algorithms] to a well-established model of the interactions between environmental, morphological, physiological and behavioural components of thermoregulation, we arrive at [clear, sometimes counterintuitive, predictions]. Confirmation of these predictions in future experiments with rodents would constitute strong evidence of a mechanism by which self-organisation can guide natural selection.''
Glancy, Stone & Wilson (2016), Royal Society Open Science.