The review, authored by Camille A. Troisi of the Centre d’Étude en Éthologie et Cognition at the Université de Rennes in France, takes a deliberately analytical approach to the problem. Rather than treating social learning as a single, indivisible behaviour, Troisi separates it into two conceptual components: the social component, which concerns the availability and use of information produced by other individuals, and the learning component, which concerns the cognitive machinery that converts observed or transmitted information into lasting behavioural change. This distinction matters because urban factors may act on each component in different, and sometimes opposing, ways.

Social learning is widely predicted to be most useful in novel or variable environments. When an animal faces a new food source, a new predator or a new hazard, copying an experienced individual can be far cheaper and safer than trial-and-error exploration. Urban environments, with their rapid change and abundance of unfamiliar challenges, appear tailor-made for this strategy. Yet, as the review emphasises, whether animals actually rely on social information in cities depends on a chain of conditions: the information must be available, it must be genuinely useful, individuals must be inclined to use it, and they must be capable of learning from it. Disruption at any link in this chain can weaken the whole process.

Pollution emerges as one of the most pervasive urban factors with the potential to interfere with social learning, and Troisi examines it in three distinct forms. Chemical pollution, including heavy metals and endocrine-disrupting compounds, can impair cognition directly, degrading the attention, memory and sensory processing needed to detect and interpret social cues. Noise pollution poses a different problem: acoustic signals are among the most important channels of social information for birds, mammals and other vocal species, and chronic urban noise can mask calls and songs, effectively cutting the bandwidth through which social information flows. Light pollution adds a further layer, altering activity patterns and potentially desynchronising the timing of social interactions between individuals that would otherwise learn from one another.

Food resources represent another pathway with ambiguous consequences. Cities often provide abundant, predictable and clumped food sources, from rubbish bins to bird feeders. On the one hand, such predictability might reduce the need for social learning, since individuals can locate resources through simple routines rather than by following others. On the other hand, concentrated food can increase the frequency of social interactions and produce local traditions, as has been observed in urban birds and mammals that learn novel foraging techniques from conspecifics. The review highlights that the direction of the effect likely depends on how resources are distributed in space and time, and on how competition shapes the willingness of individuals to allow others close enough to be observed.

Temperature is a factor that is easy to overlook but potentially significant. Urban heat islands raise ambient temperatures relative to surrounding rural areas, and temperature influences metabolic rate, activity levels and the timing of behaviour. Because social learning depends on temporal overlap between demonstrators and observers, thermal shifts that alter daily activity patterns could change who meets whom, and when. Warmer nights, for example, may extend or shift activity windows, potentially increasing opportunities for social contact for some species while reducing them for others.

Habitat structure plays an equally complex role. The built environment alters sightlines, creates vertical structures and fragments vegetation, all of which influence how easily animals can observe one another. Dense buildings may block visual transmission of social information, while at the same time creating new vantage points such as ledges, wires and rooftops where animals congregate visibly. Furthermore, structural features of cities can change the value of what is learned: navigating a maze of glass and asphalt may demand entirely new route knowledge, and experienced residents may hold information about safe corridors and hazards that naive individuals cannot easily acquire alone.

Perhaps the most distinctive feature of urban environments is the sheer pace of environmental change. Social learning carries an inherent risk: information can become outdated. In rapidly changing environments, information copied from others may be obsolete by the time it is used, favouring asocial, individual learning instead. Troisi points out that this dynamic could cut against the intuitive prediction that cities should promote social learning. If the urban landscape, traffic patterns or human behaviours that generate rewards and risks change faster than social information can circulate, animals may do better by sampling the environment themselves. The usefulness of social information, in other words, is not a fixed property but a moving target shaped by the rate of change.

Across all of these factors, the review reaches a sobering conclusion. Although there are substantial bodies of research on each individual component — on how pollution affects cognition, on how noise alters communication, on how animals learn socially in laboratory and wild settings — very little empirical work directly examines the relationship between urbanisation and social learning as an integrated whole. Most conclusions about urban social learning are therefore extrapolations rather than demonstrations. The review serves as both a synthesis of what is plausibly known and a roadmap for what remains to be tested, identifying where the logical pathways from urban factors to social information use are strongest and where empirical data are thinnest.

The implications extend beyond academic curiosity. Understanding how animals adapt cognitively to cities is increasingly relevant to conservation, urban planning and the management of human-wildlife conflict. Species that exploit social information effectively may be better equipped to colonise and thrive in urban areas, potentially explaining why some species flourish alongside humans while others retreat. If pollution, noise or the pace of change erode social learning in vulnerable species, cities could be imposing hidden cognitive costs that compound more visible threats. Filling the empirical gap that this review exposes will require targeted experiments comparing social learning performance across urban and rural populations, across gradients of urbanisation, and under controlled manipulations of the specific factors the review identifies. Until such studies accumulate, the question of whether city life makes animals more — or less — reliant on the wisdom of others remains one of behavioural ecology’s compelling open problems.

Subject of Research: The factors in urban environments that influence animals' reliance on and the usefulness of social learning.

Article Title: Factors impacting reliance on, and usefulness of, social learning in urban environments

Article References: Troisi, C. A. (2026). Factors impacting reliance on, and usefulness of, social learning in urban environments. Animal Cognition. https://doi.org/10.1007/s10071-026-02100-1

Image Credits: AI Generated

DOI: 10.1007/s10071-026-02100-1

Keywords: social learning, urbanisation, animal cognition, urban ecology, noise pollution, chemical pollution, light pollution, food resources, habitat structure, environmental change, behavioural ecology, animal behaviour

The research, led by Peiling Han, Jingwen Chen, Yongle Zhu, Jingyuan Yang, and Xuehua Li of Dalian University of Technology in China, together with Willie J. G. M. Peijnenburg of the Dutch National Institute for Public Health and the Environment and Leiden University, introduces the intelligent high-throughput multi-species physiologically based toxicokinetic model, abbreviated HM-PBTK. At its core, the model addresses a stubborn technical problem. Physiologically based toxicokinetic models have long been a cornerstone of toxicology because they simulate how a chemical enters an organism, distributes through its tissues, is metabolized, and is ultimately eliminated—the so-called ADME processes. But these models depend on dozens of species-specific parameters, such as blood flow rates, tissue volumes, tissue composition, and metabolic clearance rates, that are known for only a handful of laboratory species like zebrafish and rainbow trout. For the vast majority of the more than 30,000 fish species on Earth, such data simply do not exist.

The Chinese-led team’s solution was to build machine-learning models capable of predicting these physiological and biochemical parameters across species that have never been tested. Drawing on a multimodal dataset that integrates biological traits, phylogenetic information, environmental context, and chemical properties, the researchers trained AI systems to estimate the parameters that a toxicokinetic model needs, from cardiac output and oxygen consumption to in vitro intrinsic clearance rates. The resulting framework covers 151 freshwater and marine fish species, spanning a phylogenetic and ecological range that conventional modeling approaches could never approach. In effect, the team taught an algorithm to infer the internal plumbing and biochemistry of fish it has never seen, using patterns extracted from species that have been studied.

Once the AI-predicted parameters are plugged into the toxicokinetic model, the system can quantify how much of a given chemical accumulates in specific tissues—the liver, the gills, the blood, the gonads—under realistic exposure scenarios. This tissue-specific internal dose is the quantity that matters for toxicology, because the concentration of a chemical at its site of action, not merely its concentration in the surrounding water, determines whether harm occurs. The researchers validated the model against an extensive literature-derived dataset of 703 internal exposure measurements spanning multiple species and chemicals, providing an unusually rigorous test of the framework’s predictive power.

The performance results are striking. In a case study focused on oestrogenic effects—the induction of vitellogenin, an egg-yolk precursor protein that serves as a classic biomarker of endocrine disruption in fish—the model’s quantitative in vitro to in vivo extrapolation, or QIVIVE, placed 85 percent of toxicity predictions within fivefold of the corresponding experimental observations. In a field where predictions spanning orders of magnitude are common, and where animal testing for every species-chemical combination is impossible, a fivefold window across such a diverse species set represents a substantial gain in reliability. The case study is also ecologically pointed: synthetic oestrogens from wastewater treatment effluent have been shown in earlier work, including a landmark 2007 study in the Proceedings of the National Academy of Sciences, to collapse entire fish populations in experimental lakes.

The implications extend well beyond endocrine disruption. The researchers demonstrated the model’s application to chemicals that are frequently detected in the environment, simulating absorption, distribution, metabolism, and excretion in both freshwater and marine species under real exposure conditions. The framework handles both neutral and ionizable chemicals, a critical distinction because many pharmaceuticals and emerging contaminants carry electrical charges that dramatically alter how they move through biological membranes and how they partition into tissues. Earlier multispecies toxicokinetic efforts, including those by Brinkmann and colleagues and Mangold-Döring and colleagues in Environmental Science & Technology, laid important groundwork but were limited in species coverage and chemical scope; the new AI-driven approach scales the concept by orders of magnitude.

Recognizing that a powerful model is only as useful as it is accessible, the team also built a user-friendly web platform designed to make the technology available to risk assessors, regulators, and researchers who are not modeling specialists. The platform allows users to conduct comprehensive exposure-toxicity predictions for chemicals simply by setting up an exposure scenario—specifying the chemical, the environmental concentrations, and the species or conditions of interest—without writing code or manually parameterizing differential equations. This kind of operational tooling matters because regulatory ecological risk assessment, governed by frameworks such as the European Union’s chemicals legislation and the United Nations’ post-2020 global biodiversity framework, is under mounting pressure to evaluate thousands of substances for which experimental data are sparse.

The timing of the work is significant for reasons that go beyond computational novelty. Chemical pollution is now recognized alongside climate change and habitat destruction as a primary driver of global biodiversity decline. The landmark tire-rubber-derived chemical 6PPD-quinone, which was shown in 2021 in Science to kill coho salmon within hours of stormwater runoff exposure, illustrated how a single ubiquitous contaminant can devastate a wild fish population before anyone knew it was toxic. Meanwhile, the demand for animal testing is under ethical and practical strain: European statistics show millions of fish used in regulatory toxicity testing, and the scientific community has embraced replacement, reduction, and refinement principles. A validated computational framework that predicts internal exposure and toxicity without live animals directly serves those goals, offering regulators a route to screening that is faster, cheaper, and humane.

The study’s technical architecture reflects broader trends in computational toxicology, where machine learning has begun to infuse every layer of physiologically based pharmacokinetic and toxicokinetic modeling. Prior work had demonstrated multimodal deep learning for predicting drug clearance in humans and machine-learning models for tissue-to-blood partition coefficients, but the translation of these techniques to ecological species—where data are scarcer, species diversity is vastly greater, and environmental variables such as temperature and salinity complicate parameterization—required the kind of systematic data assembly and model integration this team undertook. By combining AI-predicted physiology with established toxicokinetic equations, the approach retains the mechanistic interpretability that regulators demand while gaining the coverage that pure data-driven models lack.

The researchers have made their work openly available to accelerate adoption: the source code for the HM-PBTK model is hosted on GitHub, and the underlying data have been deposited on figshare, alongside extensive supplementary information detailing the species datasets, model construction, and validation results. The study was supported by the National Key Research and Development Program of China, the National Natural Science Foundation of China, and the Programme of Introducing Talents of Discipline to Universities. Whether the framework becomes a standard tool in regulatory risk assessment will depend on further independent validation and integration into formal assessment guidelines, but the direction is clear. As chemical inventories continue to expand and monitoring budgets remain constrained, the ability to predict which chemicals will reach which tissues of which fish—and at what internal concentrations—may prove one of the most consequential applications of artificial intelligence in the service of aquatic biodiversity protection.

Subject of Research: AI-driven prediction of internal chemical exposure and toxicity in freshwater and marine fish for aquatic ecological risk assessment.

Article Title: Advancing aquatic ecological risk assessment through AI-driven prediction of chemical exposure and toxicity in freshwater and marine fish

Article References: Han, P., Chen, J., Zhu, Y., Yang, J., Peijnenburg, W. J. G. M., & Li, X. (2026). Advancing aquatic ecological risk assessment through AI-driven prediction of chemical exposure and toxicity in freshwater and marine fish. Nature Water. https://doi.org/10.1038/s44221-026-00709-7

Image Credits: AI Generated

DOI: 10.1038/s44221-026-00709-7

Keywords: aquatic toxicology, ecological risk assessment, machine learning, PBTK model, chemical pollution, fish biodiversity, internal exposure, toxicity prediction, freshwater ecosystems, marine ecosystems, endocrine disruption, ADME simulation