A study recently published in Science reveals: Classical models of collective behaviour fail to explain the mechanisms driving desert locust swarms – an ecological phenomenon that affects millions of lives worldwide. This research, conducted by scientists from the Cluster of Excellence "Collective Behaviour" at the University of Konstanz and Max Planck Institute of Animal Behavior, offers a new perspective on the cognitive and sensory mechanisms that underpin collective motion, challenging long-held beliefs in the field of animal behaviour.
Desert locusts, a notorious Biblical pest, form some of the largest insect groups in nature and are estimated to threaten the livelihood of one in ten people due to their impact on food security. Swarms begin when flightless juveniles aggregate and start marching in unison. Understanding how these plague insects coordinate their motion is crucial for developing evidence-based control, such as forecasting swarm movements. In addition, revealing the nature of inter-individual interactions is key to understanding how collective motion emerges among social animal species more broadly.
For decades, a principle borrowed from theoretical physics – treating individuals as "self-propelled particles" – has been used to model collective motion in animals. Similar to particles in physical systems like magnets, this hypothesis assumes that animals actively align with one another. However, unlike in magnets, these "particles" are constantly in motion. Such models have shown that even when individuals align only with their local neighbors, large-scale coherent movement can emerge, with vast numbers of individuals moving in the same direction.
The longstanding hypothesis also states that the density between the animals is a decisive factor for the change from non-coherent motion – where individuals move in random directions – to coherent large scale collective motion. When enough animals come together in a space, they are predicted to spontaneously transition from disordered to ordered swarm motion. This prediction was later seemingly corroborated by laboratory experiments with large locust groups, thereby strengthening the claims of these classical models.
Testing long-held hypotheses
Through a combination of fieldwork during East Africa’s locust outbreak of 2020, laboratory studies, virtual reality experiments, and a reevaluation of past data, researchers from the Cluster of Excellence "Collective Behaviour" at the University of Konstanz have concluded that the behavioural mechanisms governing collective motion in locust swarms cannot be explained by these classical models. Their findings challenge the traditional view by which collective motion is thought to emerge in animal groups.
"Inferring the mechanism of interaction in mobile animal groups is notoriously difficult", says Professor Iain Couzin, the study’s senior author, noting that "individuals both influence, and are influenced, by the behaviour of others in a complex interplay." To overcome this challenge, the Konstanz team leveraged immersive 3D virtual reality, enabling them to study how freely moving locusts interact with a computer-generated "holographic" virtual swarm. "This approach allowed us to rigorously test hypotheses about what drives their behaviour in ways that would be impossible in natural swarms", adds first author Dr Sercan Sayin.
The precise control of visual information afforded by virtual reality meant that the researchers could establish how sensory input is translated into movement decisions by locusts. Contrary to previous assumptions, the team observed that the "optomotor response" – an innate reflex in which locusts (and many other species) follow motion cues – is not responsible for coordinating collective motion. Indeed, they found no evidence that locusts explicitly align with the direction of motion of others at all.
In one virtual reality experiment, for example, focal locusts were placed in between two virtual swarms, one to their left and one to their right, both moving in the same direction. Classical models predict that under such circumstances, locusts should "go with the flow". However, the Konstanz team saw that locusts would turn to face one swarm, or the other, and move towards it.
Furthermore, the researchers found that group order is not simply a product of increasing density, as was previously thought. Alignment occurred in response to coherent visual cues, almost entirely independent of density. "It’s really about the quality of information, not the quantity", says Sercan Sayin. A reanalysis of a large number of previous laboratory experiments, which had argued for density-dependent transition to coherent motion, confirmed the Konstanz team’s findings, challenging previous assumptions about the behavioural mechanisms underlying swarming in locusts.
A new cognitive framework for collectives
In order to explain their results, it was necessary for the Konstanz team to rethink the approach of modeling collectives from the bottom up. "Locusts are not behaving like simple particles that align with one another", says Iain Couzin. "We realized that we need to model them as cognitive agents – processing their surroundings and making decisions about where to move next."
The research team developed a simple cognitive model, informed by the neurobiology of the neural circuits used by animals for spatial navigation, termed a "ring attractor" neural network. In this model, individuals have a simple neural representation of the bearing towards, but not the body orientation or direction of motion, of neighbours. Movement decisions emerge through a dynamic process in which neural representations compete or converge based on relative positioning, ultimately reaching a consensus that determines the direction of motion. "Our model is based on known neurobiological principles", explains Dr. Sayin, "and we found it can account for all of our key experimental findings".
The study, published in
Science, represents nothing less than a paradigm shift in swarm research.
By providing fundamental new insights into how locust behaviour results in devastating swarms, the Konstanz research may provide critical knowledge for improved locust control strategies, such as for effective modeling of swarm movement.
Moreover, the consequences of these findings will likely extend beyond locusts to broader applications in understanding the coordination of motion in other species, as well as robotics, artificial intelligence and the study of collective intelligence. Swarm robotics and autonomous vehicle coordination, for example, may benefit from algorithms inspired by locusts’ highly effective cognitive strategies for collective motion.
Key facts:
- Original publication: Sercan Sayin, Einat Couzin-Fuchs, Inga Petelski, Yannick Günzel, Mohammad Salahshour, Chi-Yu Lee, Jacob M. Graving, Liang Li, Oliver Deussen, Gregory A. Sword & Iain D. Couzin, The behavioral mechanisms governing collective motion in swarming locusts, Science387,995-1000(2025).
DOI:10.1126/science.adq7832
Link: https://www.science.org/doi/10.1126/science.adq7832
- This study was conducted by researchers in the Cluster of Excellence "Collective Behaviour" and the Max Planck Institute of Animal Behavior (MPI-AB).
- Iain D. Couzin is speaker of the Cluster of Excellence "Collective Behaviour", professor of biodiversity and collective behaviour at the University of Konstanz and director of the Max Planck Institute of Animal Behavior (MPI-AB).
- Sercan Sayin is a postdoctoral researcher in the Cluster of Excellence "Collective Behaviour" at the University of Konstanz.
- For more information, contact Professor Iain D. Couzin at icouzin@about.mpg.de or Dr Sercan Sayin at sercan.sayin@uni-konstanz.de.
|