Recent history improves prediction of collective rotation in beetle and fish groups

Macroscopic descriptions of collective motion require observables that retain information about future dynamics. We tested the predictive value of recent rotation beyond instantaneous group structure using three whirligig-beetle recordings for model development and 12 for evaluation, with nine golden-shiner recordings assessing transfer of fixed categorical settings. In beetles, adding rotation five frames earlier to current rotation, polarization, group radius and shape reduced mean-squared error by 13.73% and 13.20% at three- and ten-frame horizons. Estimating current rotation and its trend from six observations over the same interval further improved prediction in every recording at horizons of 3, 10 and 30 frames. In stochastic simulations, history reduced direction switching from 17.124 to 5.671 switches per minute, closer to the observed rate of 2.808, and improved waiting-time curves. Residual switching and correlation errors remained in smaller groups, and stationary distributions showed no consistent improvement. Time spent in the current rotational category accounted for 59.20% of the four-observation prediction-score gain, while longer contexts retained additional information. In fish, categorical history improved prediction under both body-heading and movement-based definitions of rotation; continuous gains depended on the definition. Short observation windows therefore provide effective coordinates for reduced prediction, with clearer benefits for switching dynamics than for stationary statistics.

Authors

Institutions

Publication Details

Journal
Physica Scripta
Published
2026-11-18
DOI
https://doi.org/10.1088/1402-4896/aec3dd
Primary Topic
Distributed Control Multi-Agent Systems
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Recent history improves prediction of collective rotation in beetle and fish groups

Mingxu Yin, Peng Qin, Runlin Li, Rui Liu
Physica Scripta
Distributed Control Multi-Agent Systems
article

Recent history improves prediction of collective rotation in beetle and fish groups

Mingxu Yin, Peng Qin, Runlin Li, Rui Liu
article en

Abstract

Macroscopic descriptions of collective motion require observables that retain information about future dynamics. We tested the predictive value of recent rotation beyond instantaneous group structure using three whirligig-beetle recordings for model development and 12 for evaluation, with nine golden-shiner recordings assessing transfer of fixed categorical settings. In beetles, adding rotation five frames earlier to current rotation, polarization, group radius and shape reduced mean-squared error by 13.73% and 13.20% at three- and ten-frame horizons. Estimating current rotation and its trend from six observations over the same interval further improved prediction in every recording at horizons of 3, 10 and 30 frames. In stochastic simulations, history reduced direction switching from 17.124 to 5.671 switches per minute, closer to the observed rate of 2.808, and improved waiting-time curves. Residual switching and correlation errors remained in smaller groups, and stationary distributions showed no consistent improvement. Time spent in the current rotational category accounted for 59.20% of the four-observation prediction-score gain, while longer contexts retained additional information. In fish, categorical history improved prediction under both body-heading and movement-based definitions of rotation; continuous gains depended on the definition. Short observation windows therefore provide effective coordinates for reduced prediction, with clearer benefits for switching dynamics than for stationary statistics.

Physica Scripta
Geely University of China (CN)
Distributed Control Multi-Agent Systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon