Nature-inspired model optimizes efficiency in collective tasks
December 4, 2025 | Bharti Dharapuram
In many natural systems, such as foraging ants or schools of fish, collective behaviour is driven by individual events triggering group-wide synchronization. Inspired by this, scientists from The Institute of Mathematical Sciences, Chennai have developed a theoretical model that enables a group of search agents to find their target efficiently. This framework is useful in understanding the foraging strategies of animals and collective behaviour in swarms of search robots.
“In social groups, reaching a shared decision is often essential for survival and success. In many cases, the fate of an entire group can be steered by just a few individuals,” explains Arnab Pal, IMSc faculty, who is one of the authors of the study.
Previous work has shown that this coordinated response occurs when an individual switches its state on reaching a threshold and interacts with its neighbours to propagate the change through the population. Ecological studies have shown that only a few individuals are sufficient to trigger a change that can cascade across a large social group.
The research team, including Pal and Arup Biswas from IMSc, and Satya N Majumdar from Laboratoire de Physique Théorique et Modèles Statistiques, Paris, incorporated these ideas in their study. According to their ‘threshold resetting’ model, whenever any individual in a group of random search agents reaches a threshold condition, it triggers the entire group to collectively reset to a new state.
The team first derived the mean search time of the agents as a function of their survival probability – the probability that a search agent has not reached the target with or without first hitting the threshold. They then evaluated the model’s efficiency in terms of the time to find the target and the number of resets experienced during the search. The researchers tested this model for two types of search behaviour commonly seen in nature and technology: ballistic search, where individuals move in a directed, straight-line manner, and diffusive search, where individuals wander randomly.
Using analytical methods and simulations, the researchers found that agents behave optimally under threshold resetting by reducing the costs associated with the search process. This happens as synchronous resetting allows agents to explore diverse paths during the search process.
“We were able to use simple models inspired by statistical physics to show how individual behaviours can spontaneously trigger collective motion that improves the group’s overall performance,” says Pal. “This can occur when individuals act according to 'safety rules' or 'threshold principles', which encourage them to share cues with others, thereby triggering a global system-wide reorganization.”
The study suggests potential avenues of future work, by incorporating the cost of information transfer and spatial dynamics into the resetting process.
Reference: Biswas, A., Majumdar, S. N., & Pal, A. (2025). Target search optimization by threshold resetting. Physical Review Letters, 135(22), 227101. Biswas, A., Majumdar, S. N., & Pal, A. (2025). Target search optimization by threshold resetting. Physical Review Letters, 135(22), 227101. https://doi.org/10.1103/752c-wqly
Back Subscribe
