Hide table of contents

This research was conducted for my final-year  Module “Animal Justice” at University College London led by Professor Jane Holder, investigating how biomedical researchers make welfare and value comparisons between species.

Thanks to Professor Jane Holder, Bob Fischer, Alistair Stewart, Jack Hancock-Fairs and Karolina Sarek for their guidance, thoughts and feedback.

A PDF of the research poster can be accessed here.


Inter-species tensions are ubiquitous when it comes to human engagement with animals. In many cases, when seeking to reduce harm to animals, the interests of two or more species come into conflict with each other. In these cases, we decide, whether implicitly or explicitly, how we trade the well-being and moral worth of individuals from one species against another. Despite its complexity, inter-species comparisons are inescapable, making it critical to develop a coherent and robust framework to navigate these tensions, with steps towards this being made (Gaffney et al., 2022). To do this, we must first examine our priors with regards to comparing the moral worth of animal groups against each other, explore how these may vary in different contexts and determine whether they have any validity.

Animals in Science

The use of animals in research is a strong case study of a context where we extensively make direct comparisons between the moral worth of different species, manifesting in the choices we make for which animals we conduct experiments on. 

This project sought to investigate interspecies tensions, when considering the application of the 3Rs principles from the Animals (Scientific Procedures) Act 1986, which is the piece of legislation that outlines permissions and protections with regards to the use of animals in science in the UK. 

What animals are protected?

Section 1 of the Act specifies which species of animal are protected in the context of scientific experimentation, and the life stages at which they may be protected. Protected animals are:

Any living vertebrate

Mammals, birds and reptiles are protected from ⅔ through the gestation/incubation period

Any other vertebrate is protected from the point of independent feeding

Any living cephalopod

Protected from the point of independent feeding

What does it mean to be protected?

In section 2A of the 1986 Act the 3Rs are explained, which provide the guiding ethical principles for animal experimentation. ‘Replacement’ stipulates that methods that do not use protected animals should be favoured whenever possible. ‘Reduction’ stipulates that the number of protected animals used should be reduced to a minimum. ‘Refinement’ stipulates that the handling of animals must be refined to minimise pain and distress as much as possible. However, from an ethical standpoint these principles come into tension. 

A common dilemma

Replacement is often practically applied by substituting a “more complex” animal with a “less complex” animal. This often also leads to an increase in the number of animals used, demonstrating an ethical tension between Replacement and Reduction. Therefore, although this application of Replacement is often lauded and encouraged for its compliance with the principles laid out by the 1986 Act, it reveals an ethical dilemma: at what point do the aggregate impacts of using a higher number of “lower order” animals morally outweigh the aggregate impact of using a lower number of “higher order” animals?

Research Question

This project sought to investigate how researchers justify using a higher number of a “less complex” species over a lower number of a “more complex” species and how they navigate this tension through the application of the 3Rs, through a case study of use of mice and zebrafish in biomedical research.

Case Study: Zebrafish and Mice

Why use zebrafish at all?

Zebrafish have emerged as a promising alternative to traditional rodent models in various biomedical experiments. “Due to its fully sequenced genome, easy genetic manipulation, high fecundity, external fertilisation and rapid development, and nearly transparent embryo, zebrafish are a unique model animal for biomedical research, including studies of biological processes and human diseases” (Teame et al., 2019)

Under the Animals (Scientific Procedures) Act 1986, fish only become protected animals at the point when they are capable of independent feeding, which is considered as five days post fertilisation (5 dpf) for zebrafish. 

Can zebrafish suffer?

Whilst there is a strong consensus that adult zebrafish are sentient (Sneddon et al., 2018), less research has been conducted into sentience at earlier life stages. However, recent work has demonstrated that zebrafish at 5 dpf respond to noxious stimuli in the same way as adults, from gene expression to behavioural changes (Curtwright et al. 2015; Lopez-Luna et al. 2017 a, b, c, d). Despite the absence of legal protection for zebrafish larvae up to 5 dpf, the presence of a mature brain and nerve tracts underscores their likely underscores their capacity for pain perception (Rothenbücher et al., 2019). 

Despite this, in biomedical research, zebrafish are widely used at <= 5 dpf, as a replacement for rodent models.


I conducted a literature review on the research question “how do researchers justify using ‘less sophisticated’ animals over “more sophisticated” animals in biomedical research?”. I used Google Scholar to find papers with the following search terms: 3R OR 3Rs OR 3rs zebrafish reduction higher OR lower OR order OR sophisticated. I started with 50 papers which were screened for relevance, retaining 30. I ended my literature review at this point, as I had reached saturation. 

Findings: How do biomedical researchers justify using zebrafish over mice through the 3Rs?


Biomedical researchers commonly justify the use of zebrafish at <= 5 dpf over mice through full replacement - the replacement of a protected animal with an unprotected animal (Rupprechter, 2021; Vliegenthart et al., 2014). 

“the use of zebrafish is in line with the ‘3Rs’ (reduce, refine and replace) approach of animal use for scientific purposes by replacing higher-order animals with lower-order zebrafish (particularly, zebrafish embryos).” (Vliegenthart et al., 2014)

Some researchers refer to zebrafish at this stage in development as “in vitro” models, suggesting that these animals are considered as non-living objects.

“zebrafish larvae up to 5 days post fertilization (dpf) are considered in vitro models and are accepted as an alternative to animal testing” (Cornet et al. 2017)

Zebrafish <= 5 dpf are not legally protected, so their use at this life stage is deliberate, to avoid the restrictions that come into place for the use of protected animals in experimentation.

"As the EU directive 2010_63 explicitly states only “independently feeding larval forms” must be classified as animal experiments, therefore only zebrafish larvae past 120 h post fertilization should be subject to the regulations of European animal protection guidelines. For our experiments, we did not use larvae that have reached an “independent feeding” stage and therefore we did not have to submit an ethical approval to the competent local/national ethical/legal bodies." (Bercier et al. 2019)


Biomedical researchers do not seek to apply the reduction principle directly to zebrafish use at all. Given their legal status as non-protected animals at the age of use (<= 5 dpf), there is no discussion of the reduction of the numbers of animals due to ethical considerations. Rather, the high numbers of zebrafish that can be used as a result of their unprotected status is often lauded. 

The high n numbers available per study when using zebrafish is discussed as a comparative advantage of zebrafish over mice, allowing for improved statistical analysis (Scholz, 2013)

“What they lack in size they make up for in numbers” (Zon and Peterson, 2005)

[List of advantages of zebrafish over mice] “High n numbers available per study, allowing improved statistical analysis - Lower-order mammal (in line with ‘3Rs’ principles)” (Vliegenthart et al., 2014)

Indeed, the use of a high number of zebrafish for high throughput analysis is actually justified through the reduction principle, in terms of the numbers of mice, as it reduces the total number of higher order animals needed across phases of research (de Abreu et al., 2019; Dal et al., 2022)

“This, in turn, can contribute to the refinement of experiments in higher vertebrates, such as mice and rats, and reduce the number of animals needed at a later stage in the drug discovery process." (Dal et al., 2022)


The moral weight placed on animals in biomedical research is dictated by their status as protected or non-protected animals in the Animals (Scientific Procedures) Act 1986, rather than by direct evidence of sentience or ethical concern for animal lives and experiences. 

The biomedical community does not recognise an ethical dilemma in replacing a lower number of mice with a higher number of zebrafish <= 5dpf, as non-protected animals are not considered to hold intrinsic moral value. In this way, in the eyes of the biomedical research community, the legal status of the animals dissolves the evident ethical tension. Therefore, the researchers do not feel the need to justify decisions around the use of zebrafish at <= 5 dpf in terms of the wellbeing of the zebrafish. 

The Animals (Scientific Procedures) Act 1986 is unfit for purpose, as it offers legal protection to animals on the basis of measures that are weak proxies for their moral status, particularly in light of new research in animal welfare science. What matters morally – and what ought to be captured in law – is not an animal's capacity for independent feeding, but rather their capacity for suffering. This Act utterly underserves its aim of providing an ethical framework for animal experimentation that reduces harm to animals, which calls for serious reform to this piece of legislation.


Despite the Animals (Scientific Procedures) Act 1986 being the primary piece of legislation that protects animals used for experimentation in the UK, its focus on only protecting vertebrates (plus cephalopods), with species that are not mammals, birds reptiles only protected from the stage of independent feeding shows it to be severly underserving animals. Significant groups of animals which recent evidence suggests likely have the capacity for pain, such as decapods, are left entirely unprotected. Further, all fish and amphibians are protected only after the arbitrary mark of capacity for independent feeding. When a species is unprotected, it is legal for researchers to do whatever they want to however many individuals of that group they desire. This can lead to enormous buckets of suffering. 

One vital recommendation arising from this analysis is the need to reassess the threshold at which animals gain legal protection. This criterion should be reevaluated, ensuring that animals are protected based on scientific evidence of their capacity for suffering, rather than arbitrary developmental milestones. This would also force this ethical dilemma into the light, unable to be dissolved by the legal status of zebrafish larvae as unprotected animals.

Further, there is a need for the 3Rs to evolve into a robust and coherent framework to guide researchers through the ethical implications of replacing sophisticated animals with less sophisticated ones. This evaluation should consider factors which are morally relevant, such as an animal's capacity to experience pain and should guide researchers through ethical dilemmas arising from emergent tensions between the core principles. 

A reformed framework could provide clearer guidance on when to transition from using a low number of mice to a high number of zebrafish, considering both scientific evidence and ethical considerations. This would allow researchers to make more informed interspecies welfare trade-offs, reducing the chance of intuition, biases and inaccurate priors leading to unnecessary harm. 


Bercier, V., Rosello, M., Filippo Del Bene and Céline Revenu (2019). Zebrafish as a Model for the Study of Live in vivo Processive Transport in Neurons. Frontiers in Cell and Developmental Biology, [online] 7. doi:https://doi.org/10.3389/fcell.2019.00017.

Cornet, C., Calzolari, S., Miñana-Prieto, R., Dyballa, S., van Doornmalen, E., Rutjes, H., Savy, T., D’Amico, D. and Terriente, J. (2017). ZeGlobalTox: An Innovative Approach to Address Organ Drug Toxicity Using Zebrafish. International Journal of Molecular Sciences, 18(4), p.864. doi:https://doi.org/10.3390/ijms18040864.

Curtright, A., Rosser, M., Goh, S., Keown, B., Wagner, E., Sharifi, J., Raible, D.W. and Dhaka, A. (2015). Modeling Nociception in Zebrafish: A Way Forward for Unbiased Analgesic Discovery. PLOS ONE, 10(1), p.e0116766. doi:https://doi.org/10.1371/journal.pone.0116766.

de Abreu, M.S., Giacomini, A.C.V.V., Echevarria, D.J. and Kalueff, A.V. (2019). Legal aspects of zebrafish neuropharmacology and neurotoxicology research. Regulatory Toxicology and Pharmacology, 101, pp.65–70. doi:https://doi.org/10.1016/j.yrtph.2018.11.007.

Gaffney, L., J. Michelle Lavery, Schiestl, M., Trevarthen, A., Schukraft, J., Miller, R., Schnell, A.K. and Fischer, B. (2022). A Method for Improving Interspecies Welfare Comparisons. doi:https://doi.org/10.20944/preprints202210.0012.v1.

Lopez-Luna, J., Al-Jubouri, Q., Al-Nuaimy, W. and Sneddon, L.U. (2017a). Impact of analgesic drugs on the behavioural responses of larval zebrafish to potentially noxious temperatures. Applied Animal Behaviour Science, 188(188), pp.97–105. doi:https://doi.org/10.1016/j.applanim.2017.01.002.

Lopez-Luna, J., Al-Jubouri, Q., Al-Nuaimy, W. and Sneddon, L.U. (2017b). Impact of stress, fear and anxiety on the nociceptive responses of larval zebrafish. PLOS ONE, 12(8), p.e0181010. doi:https://doi.org/10.1371/journal.pone.0181010.

Lopez-Luna, J., Qussay Al-Jubouri, Waleed Al-Nuaimy and Sneddon, L.U. (2017c). Reduction in activity by noxious chemical stimulation is ameliorated by immersion in analgesic drugs in zebrafish. The Journal of Experimental Biology, 220(8), pp.1451–1458. doi:https://doi.org/10.1242/jeb.146969.

Nils-Jørgen Knudsen Dal, Speth, M., Johann, K., Barz, M., Beauvineau, C., Jens Wohlmann, Fenaroli, F., Gicquel, B., Griffiths, G. and Alonso‐Rodríguez, N. (2022). The zebrafish embryo as an in vivo model for screening nanoparticle-formulated lipophilic anti-tuberculosis compounds. Disease Models & Mechanisms, 15(1). doi:https://doi.org/10.1242/dmm.049147.

Rothenbücher, T.S.P., Ledin, J., Gibbs, D., Engqvist, H., Persson, C. and Hulsart-Billström, G. (2019). Zebrafish embryo as a replacement model for initial biocompatibility studies of biomaterials and drug delivery systems. Acta Biomaterialia, [online] 100, pp.235–243. doi:https://doi.org/10.1016/j.actbio.2019.09.038.

Rupprechter, S.A.E. (2021). MicroRNAs as biomarkers in anti-tuberculosis drug-induced liver injury: a translational study from zebrafish to humans. Edinburgh Medical School Thesis and Dissertation Collection. doi:https://doi.org/10.7488/era/1446.

Scholz, S. (2013). Zebrafish embryos as an alternative model for screening of drug-induced organ toxicity. Archives of Toxicology, 87(5), pp.767–769. doi:https://doi.org/10.1007/s00204-013-1044-2.

Sneddon, L.U., Lopez-Luna, J., Wolfenden, D.C.C., Leach, M.C., Valentim, A.M., Steenbergen, P.J., Bardine, N., Currie, A.D., Broom, D.M. and Brown, C. (2018). Fish sentience denial: Muddying the waters. Animal Sentience, 3(21). doi:https://doi.org/10.51291/2377-7478.1317.

Teame, T., Zhang, Z., Ran, C., Zhang, H., Yang, Y., Ding, Q., Xie, M., Gao, C., Ye, Y., Duan, M. and Zhou, Z. (2019). The use of zebrafish (Danio rerio) as biomedical models. Animal Frontiers, [online] 9(3), pp.68–77. doi:https://doi.org/10.1093/af/vfz020.

Vliegenthart, A.D.B., Tucker, C.S., Del Pozo, J. and Dear, J.W. (2014). Zebrafish as model organisms for studying drug-induced liver injury. British Journal of Clinical Pharmacology, 78(6), pp.1217–1227. doi:https://doi.org/10.1111/bcp.12408.

Zon, L.I. and Peterson, R.T. (2005). In vivo drug discovery in the zebrafish. Nature Reviews Drug Discovery, 4(1), pp.35–44. doi:https://doi.org/10.1038/nrd1606.





More posts like this

Sorted by Click to highlight new comments since:

Executive summary: This research examines how biomedical researchers justify using large numbers of zebrafish larvae instead of fewer mice in experiments, highlighting tensions between the principles of Replacement, Reduction, and Refinement in animal research ethics.

Key points:

  1. The Animals (Scientific Procedures) Act 1986 protects vertebrates and cephalopods, but fish are only protected after independent feeding, allowing unregulated use of zebrafish larvae <=5 days post-fertilization (dpf).
  2. Researchers justify using large numbers of unprotected zebrafish <=5 dpf as full "Replacement" of protected mice, despite evidence suggesting zebrafish can likely experience pain at this stage.
  3. The high numbers of zebrafish used are seen as an advantage for statistical power, justified under "Reduction" by decreasing future use of mice.
  4. Legal status, not evidence of sentience, dictates the moral consideration given to animals, dissolving the ethical tension of using many "lower" versus fewer "higher" animals.
  5. The Act's protection criteria are arbitrary and unfit, failing to minimize animal suffering. Reform is needed to base protection on evidence of sentience and suffering capacity.
  6. A revised ethical framework is required to guide researchers on welfare trade-offs when replacing "higher" with "lower" animals, considering factors beyond legal status.



This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.

Curated and popular this week
Relevant opportunities