Determining the behavioral rules underlying collective behavior

Since last year's report, I have had one article published as a first author. Currently, the total cite count for my articles related to collective behavior is now 117, a number I overall am satisfied with. This has the effect that I have become a recogni zed name in the field, and am frequently asked to be a reviewer by different journals. As I have learned more about the field and its challenges throughout my postdoc, I realize that of our first attempts of reverse engineering the interaction rules i n collective biological systems lack in detail and explanatory power. Personally, I believe we need to complement the data analysis with a better understanding of how simple toy models of collective behavior work. Part of my research is therefore now focu sed on understanding from a theoretical and modeling perspective how different interactions mechanisms affect the global dynamics of a group. In parallel with this I work on experimental data. Specifically, new tracking software specifically designed for our fish species has made it possible to probe deeper into both the individual and group dynamics. We are now capable of sampling discrete decision making points in terms of individual tails beats, as well as what visual input each fish receives. Comb ining the data, we build networks of interactions and study how information flows in these networks. This is currently one of my main directions of research. At a completion stage is work I have done on scale free correlations that we have observed in fish. But we have now developed theory and models that help us understand the phenomenon, as well included data from other organism. A second project nearing completion is a detailed dynamical study of the fish. The new tracking software has allowed us to study the dynamics of individuals and group over long time scales with high accuracy, to produce an accurate picture of individual behavior and how individuals interact. Our next step is to see how the properties we find, e.g. diffusion, or correlations, relate to the assumptions used by physics based approaches, or simulations. Far on the side I have also written a paper in theoretical mathematics. I made the initial discovery almost accidentally a year ago, without knowing its significance. But la ter, I learned that my finding solves a very important problem, and I decided to write that up. The paper was very recently submitted, but from what I can understand after meeting with mathematicians, it could have a very high impact.

Swarming behavior, the mysterious and fascinating phenomenon of animals moving together in groups--often remarkably coordinated--is everywhere in nature: Flocks of birds dancing in the sky, dense schools of herring traversing the oceans, or penguins march ing to their breeding grounds. Swarms have marveled humans since dawn of history--or pestered, remember, of the ten Biblical plagues, swarms of frogs, lice, flies, and locusts made up four of them--but our knowledge is far from complete. A question recei ving considerable attention in biology is: What behavioral rules do individuals in a swarm follow? The ambition of this project is to provide a general framework for answering the question, specifically, for the strongly schooling fish species Golden shi ner. With detailed individual motion data of Golden shiners, and a reverse-engineering process that parallels finding the law of gravitation from the planetary motions, we will elucidate the individual rules used by the fish. Until recently, it was hard to gather data on individuals within swarms, resulting in a bias toward studying swarms through simulation models rather than experiments. This is changing. New technologies (cameras, sonar, GPS, software) allow for tracking individuals, and researchers a re now obtaining high-resolution motion data in both field and laboratory settings. The methods developed in this project will help closing the gap between models and real swarms. A successful outcome will provide deep insights into the nature of swarms, helping to explain why collective behavior repeatedly emerges from natural selection. Collective behavior is central to such diverse fields as anthropology, biology, economics, politics, psychology and engineering. A better appreciation of swarming ther efore has the potential to impact a range of areas, for example our understanding of migrating cancer cells, management of fish populations, or herd mentality among financial investors.