Advocacy
The Role of Animal Research in Advancing Brain Health & Research
Research involving animals underpins much of what we know about the brain. Explore the evidence and the facts behind the debate.
Few areas of medicine carry a greater burden of unmet need than brain conditions. In Europe, at least 1 in 3 people live with a brain condition, neurological or mental, at an annual cost exceeding 800 billion euros (Gustavsson et al, Olesen et al., 2012). For most of these conditions there is no cure, and in many cases no treatment that reaches the underlying disease. Progress depends on understanding how the brain functions, which remains one of the greatest challenges in science.
Much of that understanding has come, and continues to come, from research involving animals. In recent years this research has been challenged by a line of argument that has become familiar in public debate. It runs as a connected chain of claims: that animal models poorly predict human outcomes and so drive the high failure rate of new medicines, that the research is therefore scientifically invalid and its funding largely wasted, and that non-animal methods are now advanced enough to take its place. The conclusion drawn is that animal research should be phased out across the EU on a fixed timetable.
This page addresses the main questions at the center of this debate: why animals are used in neuroscience, whether animal models are scientifically valid, whether their funding is well spent, what non-animal methods can currently achieve, and how this research is regulated. Each section draws on evidence to explain why the responsible use of animal research remains necessary in neuroscience today, and how European science is working to reduce that reliance over time.
Despite the growing prevalence and burden of brain disorders, their underlying mechanisms and aetiologies remain poorly understood (Homberg et al., 2021). Neuroscience research plays a key role in addressing this issue by investigating the fundamental principles of brain function, elucidating disease mechanisms and developing novel treatments (Homberg et al., 2021). Animal models are fundamental to neuroscience research, providing critical insights into both normal brain function and the mechanisms underlying brain disorders. Research using animal models has substantially improved our understanding of disease pathophysiology and remains important for the preclinical evaluation of potential therapies (Chesselet and Carmichael, 2012; Romanova and Sweedler, 2018).
The brain is the most complex organ in the body, integrating signals from peripheral organs and sensory systems while interacting extensively with the immune and endocrine systems (Homberg et al., 2021). Therefore, understanding its function requires studying intact neural circuits within the natural context of a whole organism (Homberg et al. 2021, Genzel et al., 2024). Brain function is also intrinsically linked to behaviour and cognition, which cannot be captured in vitro or in silico (Homberg et al. 2021). Given this complexity, animal models remain essential in neuroscience as they provide whole-organism insights that non-animal approaches cannot replicate (Couch, 2026).
A variety of animal models are used in neuroscience research, with the choice of species carefully determined by the specific research question and the biological processes or brain disorders under investigation. Examples of animal species used to investigate brain function across health and disease include:
• Non-human primates (NHPs) are valuable models for studying complex processes, including higher cognitive functions, and their close anatomical and physiological similarity to the human brain makes them highly valid translational models (Homberg et al., 2021). In Europe, the use of NHPs in basic and translational neuroscience research is limited, accounting for less than 0.05% of laboratory animals used for these purposes (ALURES – Animal Use Reporting EU System, 2023).
• Rodents, including mice and rats, are the most commonly used animal species, accounting for 74.70% and 12.74% of laboratory animals used in basic and translational neuroscience research (ALURES – Animal Use Reporting EU System, 2023). They are employed in a wide range of studies, including modelling neurological and psychiatric disorders, mapping neural circuits and developing novel therapeutics. Their widespread use is largely driven by the availability of well-established transgenic models, cost-effectiveness, manageable husbandry requirements and rapid gestation.
• Zebrafish are a significant model in neuroscience research, accounting for 10.85% of laboratory animals used (ALURES – Animal Use Reporting EU System, 2023). Their transparent embryos, genetic manipulation, and suitability for live imaging and high-throughput studies enable real-time analysis of brain development and neuronal activity (Doszyn et al., 2024).
• Invertebrates have simpler nervous systems that support conserved fundamental neural processes, including synaptic transmission and neural circuit function. Models like nematodes (Caenorhabditis elegans) and the fruit fly (Drosophila melanogaster) have well characterised nervous systems and extensive genetic toolkits, making them valuable to study the molecular basis of neurotransmission, learning, and memory (Chesselet and Carmichael, 2012; Romanova and Sweedler, 2018).
Animal models are suitable for studying human brain function because they share conserved genetic, molecular and neuroanatomical features with humans (Homberg et al. 2021; Suleiman, 2025). They also enable investigation of brain function within the context of a whole organism, recapitulating the complexity of brain interactions with sensory, immune, and endocrine systems, as well as behaviour and cognition.
Different species provide complementary insights across levels of biological complexity. Rodents and non-human primates are particularly valuable due to their close functional and anatomical similarities to the human brain (Suleiman, 2025). Simpler invertebrate organisms also possess conserved molecular pathways and well-defined neural circuits, supporting the study of fundamental biological processes and high-throughput drug screening (Homberg et al., 2021; Chesselet and Carmichael, 2012; Romanova and Sweedler, 2018).
A common objection holds that the funds directed to animal research are largely wasted, particularly when studies do not lead directly to a new treatment. This view rests on a narrow definition of return, in which only an immediate therapeutic success counts as value and everything else is loss.
Research rarely progresses in this way. A study that does not produce a treatment still generates knowledge that narrows the field of inquiry, refines subsequent questions, and improves the design of later work. Negative and null results carry particular value: when shared, they prevent the costly duplication of unproductive avenues and allow resources to be redirected toward more promising ones (Yu, 2020). Failing to make use of such findings is itself a recognised source of waste in biomedical research (Yu, 2021).
Preclinical animal research has also substantially broadened the understanding of the mechanisms underlying brain disorders (Chesselet & Carmichael, 2012), providing the foundation on which later therapeutic work is built. Its value lies not only in individual successes but in the cumulative knowledge that makes future research, including many non-animal approaches, more efficient.
Where animal research does yield limited returns, the cause is typically not the funding itself but shortcomings in methodology. Translational failures are mainly due to limitations in experimental design, data analysis, reporting, or inappropriate model selection, rather than to an inherent lack of relevance of the models themselves (Chesselet and Carmichael, 2012; Couch, 2026). Efforts to strengthen rigour, reproducibility and transparency are therefore central to ensuring that investment translates into scientific and clinical benefit (Spanagel, R.; Radabaugh et al., 2023). Substantial progress has already been made in this direction, improving the quality and reproducibility of preclinical neuroscience through measures such as preregistration, randomisation, blinding and the routine consideration of sex, age and comorbidities (Couch, 2026).
In recent years, the availability and use of non-animal approaches have increased significantly. For instance, the European Commission has published a study identifying 568 non-animal models for neurodegenerative disorders, including in vitro models based on cells and tissues cultured in the laboratory, ex vivo models using cells and tissues explanted from patients, and in silico models based on computer modelling and simulation (Gribaldo et al. 2021). Examples of key new approach methodologies (NAMs) currently in use are presented below, together with an overview of their respective strengths and limitations:
• In silico approaches use computer models which allow us to model molecular dynamics at a speed and scale that is not feasible in vivo (Couch 2026). As such, they help to prioritise and refine candidates, optimise study design and focus resources on drug development pipelines (Couch 2026). However, computer models are limited by current knowledge and can only simulate known biological processes (Homberg et al., 2021). Therefore, they are the most effective when used alongside experimental approaches, complementing animal studies and helping to reduce animal use, while remaining unable to replace animal research to investigate complex unknown biological processes (Homberg et al., 2021).
• In vitro approaches, ranging from immortalised cell lines to patient-derived induced pluripotent stem cell (iPSC) cultures, are also widely used. Human iPSC-derived models can recapitulate key aspects of central nervous system physiology and enable disease-relevant assays, demonstrating promising predictive value for drug efficacy and toxicology studies (Couch 2026). However, these models often rely on a limited number of cell lines and small sample sizes, which may not adequately capture inter-line variability (Couch 2026). In addition, they do not fully reproduce the cellular complexity, tissue architecture, and systemic interactions of the human brain (Couch 2026).
• Organ-on-chip technologies are advanced in vitro systems designed to better mimic tissue and organ function using human cells (often iPSC-derived) within microfluidic devices. They can generate human-relevant data and can be linked together as “body-on-chip” systems to model interactions between organs (Couch 2026). However, these systems still have important limitations. They often require the use of cells specifically selected for their ability to survive in these conditions, thus reducing physiological accuracy, and they cannot fully reproduce whole-body processes such as immune activity, hormonal regulation, or natural biological variability (Couch 2026). Technical constraints, including materials used and limited scalability, also restrict wider application. In addition, results can vary between labs due to differences in chip design, flow rates, and cell sources (Couch 2026).
• Brain organoids are 3D structures derived from pluripotent stem cells or organ-restricted progenitors that recapitulate several aspects of human development, cellular diversity and spatial organisation that are absent in regular cultures (Couch 2026). They provide a human-relevant system that can model specific cellular processes in vitro, supporting the study of disease processes and drug responses in a patient-specific context (Couch 2026). However, they remain incomplete models of the brain: they lack vasculature, immune and endocrine components, as well as interactions with other organs in the body (Homberg et al. 2021; Couch 2026; Suleiman). They cannot reproduce higher-order functions such as behaviour or cognition, nor do they capture the full range of neural interactions or physiological responses observed in animal models (Homberg et al., 2021; Couch 2026; Suleiman 2025). Additional limitations include variability between protocols, limited standardisation, long culture times, high cost, and low scalability, which restrict reproducibility and high-throughput use (Couch 2026).
Even the most advanced in vitro and in silico models cannot fully recapitulate the complexity and integrated physiology of a living organism (Couch 2026). Moreover, many non-animal approaches have been developed or validated using animal-based research and often still rely on animal-derived materials, such as serum or antibodies.
Rather than replacing animal models outright, a complementary approach integrating animal and non-animal models should be adopted. This strategy could enable researchers to refine and reduce animal use while improving scientific understanding, with each approach informing and strengthening the other.
Animal research in the European Union operates within one of the most developed regulatory systems governing any scientific activity. Its foundation is Directive 2010/63/EU, binding in every Member State and built on the principle of the 3Rs; to Replace animals wherever possible, Reduce the numbers used, and Refine procedures to minimise pain, suffering and distress. More than a decade after it took effect, the Directive is recognised as having introduced significant welfare improvements and novel requirements while safeguarding research integrity, setting clear provisions across scope, authorisation, ethical review, severity classification and reporting (Marinou and Dontas, 2023).
The Directive is implemented through a dense institutional framework. Across the EU and Norway, more than 1,700 designated competent authorities authorise establishments, evaluate and authorise projects, and carry out inspections and retrospective assessment (European Commission, SWD (2024) 183). Each Member State maintains a National Committee for the protection of animals used in science, and every establishment must operate an Animal Welfare Body overseeing the animals in its care.
No research may proceed until it passes a formal project evaluation: authorisation is granted only where animal use is justified, the 3Rs have been applied, and the expected benefits are judged to outweigh the anticipated harms. Everyone who carries out procedures, cares for animals, or designs studies must be trained across defined functions and supervised until their competence is assessed, guided by a common Commission Education and Training Framework.
Competent authorities must inspect at least one-third of user establishments each year, a proportion without prior warning, with provision for retrospective assessment, withdrawal of authorisation, and penalties for non-compliance (European Commission, SWD (2024) 183).
Together, these elements form a multi-layered system through which the EU regulates the use of animals in science: binding law, prior authorisation, institutional oversight, trained personnel and mandatory inspection.
Chesselet, M.-F., & Carmichael, S. T. (2012). Animal models of neurological disorders. Neurotherapeutics, 9(2), 241–244. https://link.springer.com/content/pdf/10.1007/s13311-012-0118-9.pdf
European Commission. (2024, July 19). Union overview on the implementation of Directive 2010/63/EU on the protection of animals used for scientific purposes in the Member States of the European Union (SWD(2024) 183 final). https://d8aaf127-0203-427a-b8b6-1f1b942cd1af.usrfiles.com/ugd/d8aaf1_3673afd6252b4ebe824da451eb1f8c53.pdf
Marinou, K. A., & Dontas, I. A. (2023). European Union legislation for the welfare of animals used for scientific purposes: Areas identified for further discussion. Animals, 13(14), Article 2367. https://pmc.ncbi.nlm.nih.gov/articles/PMC10376073/
Radabaugh, H. L., et al. (2023). Increasing rigor of preclinical research to maximize opportunities for translation. Neurotherapeutics, 20, 1433–1445. https://link.springer.com/article/10.1007/s13311-023-01400-5
Spanagel, R. (2022). Ten points to improve reproducibility and translation of animal research. Frontiers in Behavioral Neuroscience, 16, Article 869511. https://pmc.ncbi.nlm.nih.gov/articles/PMC9070052/
Yu, H. (2020). Leveraging research failures to accelerate drug discovery and development. Therapeutic Innovation & Regulatory Science, 54, 788–792. https://link.springer.com/article/10.1007/s43441-019-00005-5
Yu, H. (2021). Responsible use of negative research outcomes—Accelerating the discovery and development of new antibiotics. The Journal of Antibiotics, 74(8), 543–546. https://doi.org/10.1038/s41429-021-00439-w
Homberg, J. R., Adan, R. A. H., Alenina, N., et al., & Genzel, L. (2021). The continued need for animals to advance brain research. Neuron, 109, 2374–2379.
Romanova, E. V., & Sweedler, J. V. (2018). Animal model systems in neuroscience. ACS Chemical Neuroscience, 9, 1869–1870.
Genzel, L., Froudarakis, E., & Rapti, G. (2024). Openness and transparency in animal research: Why and how. European Journal of Neuroscience, 60, 6866–6873.
Couch, Y. (2026). The use of animals in neuroscience research. Brain, 1–3.
Gribaldo, L., Dura, A., & Whelan, M. (2021). Advanced non-animal models in biomedical research – Neurodegenerative diseases – Executive summary (EUR 30334/2 EN, JRC124723). Publications Office of the European Union. https://doi.org/10.2760/005446
Suleiman, R. (2025). Beyond animal testing: Bridging neuroscience innovation with ethical responsibility. European Journal of Neuroscience, 62, e70222.
Doszyn, O., Dulski, T., & Zmorzynska, J. (2024). Diving into the zebrafish brain: Exploring neuroscience frontiers with genetic tools, imaging techniques, and behavioral insights. Frontiers in Molecular Neuroscience, 17, Article 1358844.
Animal Use Reporting – EU System (ALURES). (2023). Section 2 – Details of all uses of animals for research, testing, routine production and education and training purposes in the EU. Reporting year: 2023 – EU27 & Norway; Purpose: Basic research – Nervous system; Translational and applied research – Human nervous and mental disorders.
As Europe continues to navigate the future of neuroscience research, EBC believes that progress depends on decisions grounded in robust scientific evidence, ethical responsibility, and a shared commitment to improving the lives of those affected by brain disorders. We are committed to promoting informed, transparent dialogue on the role of animal research, while advancing scientific innovation that serves both patient outcomes and animal welfare. These principles are reflected in our Pledge for Science and our Position Statement on Animal Research, which together set out the foundation of our approach to this complex and evolving field. We invite readers to consider these documents alongside the evidence presented here, as part of a balanced and informed discussion about the future of brain research in Europe.