I. D. Couzin

Honey bee drones are synchronously hyperactive inside the nest

This study tracks honey bee drones inside the nest and finds they exhibit synchronized hyperactivity in the afternoon, influenced by weather and social cues.

Quantifying the movement, behavior, and environmental context of group-living animals using drones and computer vision

This paper introduces a system using drones and computer vision to track animal movement, posture, and environmental context in natural settings with high spatiotemporal resolution.

Both Prey and Predator Features Determine Predation Risk and Survival of Schooling Prey

This study quantifies predation risk and survival in schooling fish by analyzing predator and prey spatial and kinematic features during predator attacks.

The geometry of decision-making

This study reveals that animals use geometric principles to break multi-choice decisions into binary decisions during spatial navigation, supported by experiments with fruit flies, locusts, and zebrafish.

The dominant axes of lifetime behavioral variation in honey bees

This study quantifies behavioral variation in honey bees over their lifetimes and under environmental changes, identifying two dominant behavioral axes and showing individual consistency and colony-level adaptability.

Collective detection based on visual information in animal groups

This study investigates how visual detection in animal groups depends on group size, position, and stimulus location, using golden shiner fish as a model system.

Social networks predict the life and death of honey bees

The paper develops a method to predict honey bee task allocation, survival, and behavior using social network descriptors without interfering with colonies.

TRex, a fast multi-animal tracking system with markerless identification, and 2D estimation of posture and visual fields

TRex tracks multiple animals in real-time with markerless identification and estimates 2D posture and visual fields.

VAE-SNE: a deep generative model for simultaneous dimensionality reduction and clustering

VAE-SNE is a deep generative model that simultaneously reduces dimensionality and clusters data, performing well on real-world datasets and applicable to various life science data types.

Schistocephalus parasite infection alters sticklebacks' movement ability and thereby shapes social interactions

This study shows that Schistocephalus infection impairs stickleback movement, leading to altered social interactions in groups.

DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning

DeepPoseKit provides a fast and robust software toolkit for animal pose estimation using deep learning, improving speed without sacrificing accuracy.

Counteracting estimation bias and social influence to improve the wisdom of crowds

The paper improves collective estimation accuracy by addressing individual biases and social influence through new aggregation methods.

Mobility can promote the evolution of cooperation via emergent self-assortment dynamics

This paper shows mobility can promote cooperation by enabling self-assortment of cooperators in dynamic populations without requiring information about others' traits or strategies.

Consistent Individual Differences Drive Collective Behaviour And Group Functioning Of Schooling Fish

The paper shows consistent individual differences in fish behavior affect group structure, movement, and foraging, with effects emerging from a generic model of self-organizing groups.