arXiv subjects

Andrew M. Hein

Publications and source records attributed to Andrew M. Hein.

5 recordsLinked to original sources

arXiv2026

WildFin: An In-the-Wild Dataset for Fish Behavioral Recognition

Recent advances in field technology have led to a massive influx of in-the-wild video data for ecological science. The primary bottleneck in leveraging this data is the high cost of expert annotation. While computer vision offers a potential solution, current models frequently fail when deployed in complex marine environments. To characterize these failures, we introduce WildFin, a novel benchmark for fish behavior recognition collected and annotated by ecologists. WildFin spans two critical real-world paradigms: stationary cameras monitoring groups of fish and dynamic divers following individual subjects. The dataset represents a massive curation effort, involving 1,350 hours of fieldwork and 600 hours of expert annotation to produce 9 hours of behavioral data with over 2 million frame-by-frame labels. We benchmark modern vision foundation models and quantify tradeoffs between static and spatiotemporal architectures, revealing the substantial gap that remains between current model capabilities and the demands of real-world underwater behavioral analysis. Project website: https://team-wildfin.github.io/.

cs.CV↗

arXiv2026

Messaging strategies and the emergence of echo chambers in collective decision-making

Collective decision-making arises from individual agents integrating their own personal observations with information obtained from social partners. In many biological systems that exhibit collective decision-making, the process by which social information is produced, transmitted, and used is subject to two key constraints. First, individuals often do not observe the internal states or personal observations of their neighbors; instead, they observe neighbors' discrete actions. Second, agents often have limited attention, such that, at any given moment, only a subset of social partners influences decisions. Using methods from nonlinear dynamics, we show that either of these constraints can cause collective accuracy to become extremely sensitive to the weight individuals place on the information they receive from others. This sensitivity arises from the spontaneous formation of echo chamber-like states in which individuals receive and transmit homogeneous social messages. Under such conditions, collectives become locked in self-reinforcing states that prevent them from tracking changes in the environment. We reveal the mathematical basis of this phenomenon, and show that it emerges not only in generic models of collective decision-making but also in models developed to describe specific biological systems, including neural circuits, eusocial insect colonies, and mobile animal groups. Finally, we identify biologically plausible mechanisms through which individuals may reduce the risk of echo chamber formation and achieve robust yet sensitive collective decisions without requiring fine-tuning parameters. Our results reveal how fundamental constraints on communication shape the dynamics and reliability of collective decisions across diverse biological systems.

q-bio.QM↗

arXiv2019

Bacteria push the limits of chemotactic precision to navigate dynamic chemical gradients

Ephemeral aggregations of bacteria are ubiquitous in the environment, where they serve as hotbeds of metabolic activity, nutrient cycling, and horizontal gene transfer. In many cases, these regions of high bacterial concentration are thought to form when motile cells use chemotaxis to navigate to chemical hotspots. However, what governs the dynamics of bacterial aggregations is unclear. Here, we use a novel experimental platform to create realistic sub-millimeter scale nutrient pulses with controlled nutrient concentrations. By combining experiments, mathematical theory and agent-based simulations, we show that individual \textit{Vibrio ordalii} bacteria begin chemotaxis toward hotspots of dissolved organic matter (DOM) when the magnitude of the chemical gradient rises sufficiently far above the sensory noise that is generated by stochastic encounters with chemoattractant molecules. Each DOM hotspot is surrounded by a dynamic ring of chemotaxing cells, which congregate in regions of high DOM concentration before dispersing as DOM diffuses and gradients become too noisy for cells to respond to. We demonstrate that \textit{V. ordalii} operates close to the theoretical limits on chemotactic precision. Numerical simulations of chemotactic bacteria, in which molecule counting noise is explicitly taken into account, point at a tradeoff between nutrient acquisition and the cost of chemotactic precision. More generally, our results illustrate how limits on sensory precision can be used to understand the location, spatial extent, and lifespan of bacterial behavioral responses in ecologically relevant environments.

physics.bio-ph↗

arXiv2015

Physical Limits on Bacterial Navigation in Dynamic Environments

Many chemotactic bacteria inhabit environments in which chemicals appear as localized pulses and evolve by processes such as diffusion and mixing. We show that, in such environments, physical limits on the accuracy of temporal gradient sensing govern when and where bacteria can accurately measure the cues they use to navigate. Chemical pulses are surrounded by a predictable dynamic region, outside which bacterial cells cannot resolve gradients above noise. The outer boundary of this region initially expands in proportion to $\sqrt{t}$, before rapidly contracting. Our analysis also reveals how chemokinesis - the increase in swimming speed many bacteria exhibit when absolute chemical concentration exceeds a threshold - may serve to enhance chemotactic accuracy and sensitivity when the chemical landscape is dynamic. More generally, our framework provides a rigorous method for partitioning bacteria into populations that are "near" and "far" from chemical hotspots in complex, rapidly evolving environments such as those that dominate aquatic ecosystems.

physics.bio-ph↗

arXiv2012

Sensing and decision-making in random search

While microscopic organisms can use gradient-based search to locate resources, this strategy can be poorly suited to the sensory signals available to macroscopic organisms. We propose a framework that models search-decision making in cases where sensory signals are infrequent, subject to large fluctuations, and contain little directional information. Our approach simultaneously models an organism's intrinsic movement behavior (e.g. Levy walk) while allowing this behavior to be adjusted based on sensory data. We find that including even a simple model for signal response can dominate other features of random search and greatly improve search performance. In particular, we show that a lack of signal is not a lack of information. Searchers that receive no signal can quickly abandon target-poor regions. Such phenomena naturally give rise to the area-restricted search behavior exhibited by many searching organisms.

q-bio.PE↗