The Theory of Collective Decisions
In quantitative biology, a major challenge is understanding the core drivers of functional collective behavior. Examples include evasive movements of fish schools that foil predators, power structures supporting conflict management in monkey societies, gene interactions producing distinct cell types, and neurons combining information to make a decision. In each case, control lies with individual components, but adaptive consequences are a property of the aggregate.
There is an urgent need to find theories of collective behavior that can operate in complex, data-rich environments. To tackle this problem, my research begins with biological data, uses model inference to discover predictive mechanisms, and then abstracts away from them to analyze collective function.
This project focuses specifically on binary decision-making, arguably the simplest form of collective computation. What forms of information sharing produce reliable decisions, and what mechanisms allow for adjusting and controlling these collective computations?
We are starting to see commonalities in collective decision-making across multiple systems. For instance, fish schools collectively encode the risk of predators, dynamically changing to allow a startle decision to spread quickly during periods of high risk (by condensing the school into a smaller area), while avoiding false alarms during periods of low risk (by spreading out). Similarly, in neurons driving a decision in a macaque brain, we see evidence of regulation of the distance from a symmetry-breaking transition that creates the two decision states.
Intriguingly, these results suggest that collective decisions typically involve dynamically varying the distance from a point of aggregate instability. In this work, I seek to better understand how systems regulate their position relative to such instabilities. In this way, I aim to discover fundamental tradeoffs that underlie adaptation and the control of collective decisions.
Daniels, Bryan C., William S. Ryu, and Ilya Nemenman (2019). "Automated, predictive, and interpretable inference of Caenorhabditis elegans escape dynamics." PNAS 116, 15: 7226-7231.
Daniels, Bryan C., Jessica C. Flack, and David C. Krakauer (2017). "Dual coding theory explains biphasic collective computation in neural decision-making." Frontiers in Neuroscience 11: 313. doi:10.3389/fnins.2017.00313.
Daniels, Bryan C., David C. Krakauer, and Jessica C. Flack (2017). "Control of finite critical behaviour in a small-scale social system." Nature Communications 8, 14301. doi: 10.1038/ncomms14301.
Publikationen aus der Fellowbibliothek
Daniels, Bryan C. (
Dual coding theory explains biphasic collective computation in neural decision-making
Daniels, Bryan C. (
Control of finite critical behaviour in a small-scale social system