AQUAEXCEL2020 Ethics Advisor - Professor Felicity Huntingford
Members of the general public are increasingly concerned about fish welfare, particularly the welfare of the fish they eat, and, partly in response to this concern, the aquaculture industry has taken many steps to improve the welfare of farmed fish. These steps have been informed by an intensive programme of research into many aspects of fish health and welfare and have been carried out by scientists from a range of disciplines, many based in Europe and supported by EU funding. While making significant contributions to the welfare of farmed fish, such studies may themselves raise ethical issues, for example, if experimental studies (say those aimed at identifying the fish density that optimises welfare in a given species) require exposing fish to potentially adverse conditions (such as densities that might be below and above that optimum). Researchers must be mindful of these issues when planning their research, ensuring compliance with national and international regulations.
Professor Felicity Huntingford is the independent Ethics Advisor for AQUAEXCEL2020, ensuring that research is carried out to a high ethical standard. She is honorary Professor of Functional Ecology at the Institute of Biodiversity, Animal Health and Comparative Medicine at the University of Glasgow (UK) and one of the world’s pre-eminent experts on fish behaviour. Professor Huntingford has extensive experience researching fish behaviour and advising the aquaculture sector on welfare practices. She has been involved in numerous collaborative research projects funded by the EU and has spoken at many aquaculture workshops and conferences about the welfare of farmed fish.
We interviewed Professor Huntingford to find out more about her role as ethics advisor for AQUAEXCEL2020.
- What made you decide to work on welfare issues in the aquaculture sector?
It was a gradual process, rather than a straightforward decision. From the start of my research career, as a doctoral student in the early 1970s, I have worked on fish behaviour, concentrating on how individual fish respond to potential competitors (for example, fighting with rivals) and to potential risk (for example, responding to the presence of predators). I work primarily in the laboratory, using an experimental approach. This immediately raises ethical issues, because although aggression and avoiding predators are part of a fish’s natural behavioural repertoire, the particular interactions that I have recorded would not have happened, nor would they have happened in a restricted space, had I not staged-managed them. When working with sticklebacks (in my DPhil thesis and beyond), l quantified the responses of stickleback during short exposures to a territorial intruder confined behind glass and to a pike fed to satiation (so that it might approach potential prey but would not attack). This was primarily for reasons of experimental control, but it meant that the sticklebacks involved experienced no actual damage; all the same these sticklebacks (and to an extent the pike as well) were certainly stressed.
Two things happened in the 1980s that focussed my attention on ethics and fish welfare. The first was a famous and painful court case in the UK in which a reputable behavioural biologist was prosecuted under the UK’s Cruelty to Animals legislation (designed primarily to ban activities such as dog fighting and badger baiting) for laboratory studies of the development of predatory behaviour in mammals. I agreed to appear as a witness for the defence because I too worked on stage-managed predator-prey interactions in vertebrates. Somewhat to my relief, this turned out not to be necessary, because the researcher involved decided to plead guilty, after discussion with the judge, who made it clear that he had no doubt at all that the work had been motivated by absolutely appropriate scientific objectives. However, he identified two things about the way the experiments were designed as ‘wonton’ (an old-fashioned English word from the relevant act, one definition of which is ‘having no just foundation’), which was critical from a legal perspective. These were that, in his view, the tests to quantify hunting efficiency were longer than was needed to be able to answer the identified questions and also that no humane endpoints were in place. Looking back, I find it very interesting that this judge (a lawyer and not a scientist) could clearly see problems relating to what we now call refinement.
The case served as a wake-up call to the UK government, which responded by overhauling the legislation relating to the use of animals for scientific purposes, at that point more than a hundred years old and covering physiological but not behavioural experiments. Having been involved in the court case, I was included in the consultation process prior to the drafting of the new law, which required behavioural studies that might cause stress, distress or lasting harm to vertebrates (and cephalopods) to be carried out under licence, regulating behavioural research but also giving protection against prosecution. The case was also a wake-up call to the community of behavioural biologists and the UK’s Association for the Study of Animal Behaviour, who, together with the American Animal Behaviour Society, drafted ethical guidelines for the use of animals in behavioural research. These were published in and still inform acceptance of papers in their jointly-managed journal Animal Behaviour. Having been involved in the court case and in the consultation, I was asked to write the sections of these guidelines relating to aggression and predator-prey interactions. The guidelines were subsequently adopted by the Fisheries Society of the British Isles for research published in their Journal of Fish Biology, and so when this society decided to commission a Briefing Paper on fish welfare, they asked me to put together a team for this purpose. This Briefing Paper and the review arising from it generated sufficient interest and controversy (particularly about definitions of welfare based on what animals feel, which leads to questions about the extent to which fish are sentient) to keep me thinking, writing and talking about fish welfare up to the present day. So, I came to focus on fish welfare through concern about what I do to fish during my own research and not primarily about what other people do to them.
The other thing that happened to move my attention to fish welfare was a request in the mid-1980s, when I had just started to work on Atlantic salmon, for myself and colleagues to monitor the behaviour of farmed salmon in sea cages. The request came from the (then) Scottish Salmon Growers Association, in response to concern from their members that aggressive monopolisation of food might be causing injuries and uneven feed distribution in their stock; this turned out to be absolutely correct. This study was the start of a long-term interest in adding behavioural biology to the portfolio of disciplines working to promote welfare-friendly aquaculture.
- What does your work as Ethics Advisor for AQUAEXCEL2020 involve?
In AQUAEXCEL2020 my role involves screening all planned research projects, both the TransNational Access applications and the Joint Research Activities, for ethical issues, mainly focussing on compliance with the 3Rs – Replacement, Reduction and Refinement. It also involves analysing all the ethical assessments on an annual basis, to give an overview of how well AQUAEXCEL2020–funded research complies with the 3Rs. I then present this information to AQUAEXCEL2020 partners at the annual meeting. It’s important to note that my role does not involve making decisions about the scientific quality of the proposed research.
- What do you consider to be good welfare practice for AQUAEXCEL2020 researchers when designing and carrying out their experiments? Do you have a strong example from the project so far?
All AQUAEXCEL2020 partners work within the 3Rs framework in accordance with the relevant European, national and institute regulations relating to the use of vertebrates in scientific experiments. For example, the TNA application form requires information on any authorisation required for the proposed work, an ethical analysis of the procedures to be used and an account of steps to be taken to meet the requirements of each of the 3Rs. My job as ethical advisor has mainly involved scrutinising this ethics section alongside the accompanying detailed description of research procedures. In general, applicants have handled the ethical issues well. In 2019, 26% of TNA applications raised no ethical concerns regarding the use of live finfish. In 58% of cases that did raise concerns, these were dealt with entirely satisfactorily. The remaining 42% of those cases required some further information, but these issues were resolved satisfactorily.
Research funded by AQUAEXCEL2020 provides some excellent examples of best practice in relation to the 3Rs. One partner has developed microorganism-free strains of the small crustacean Artemia for replacement of live fin fish as experimental subjects, to investigate disease transmission and resistance processes. One TNA-funded study of the effects of stress on vulnerability to disease combined in vitro studies of virulence in cultured bacteria exposed directly to corticosteroids (among other substances) with in vivo studies of disease acquisition in Artemia exposed to such corticosteroid-treated pathogens. Important findings here that are potentially generalisable to finfish are better understanding of what determines bacterial virulence and identification of potential anti-virulence treatments. The following shows an impressive concern by the applicant for ethical issues: “The model organism Artemia is not covered by laws and regulations of animal experiments. We will however, always seek to treat the organism with as much care and respect as possible, and try to keep the number of organisms used to a minimum.”
On the subject of reduction, many TNA researchers proposed work on samples stored from earlier studies - sensible forward planning that reduces the need for repeated procedures. Other effective steps to reduce the number of live fish used in experiments are illustrated by a study of the effects of temperature on progression of a bacterial disease of Rainbow trout, tracking disease progression in individually identified fish (using PIT tags); this increases statistical power and allows valid conclusions to be drawn using fewer fish. For some research aims, experimental fish must be held at densities comparable to those experienced by their farmed counterparts, so that findings can be generalised, making the number of treated subjects inevitably higher than the minimum for statistical validity. This was the case in a study exploring the use of probiotic lactobacilli in improved weaning diets for pike-perch larvae; here reduction was achieved by carrying out preliminary in vitro tests to reduce the number of experimental treatments.
This same application for work on pike-perch larvae dealt well with one important aspect of refinement where experiments involve longer impositions of procedures. Specifically, this is the need to monitor fish frequently, assessing their status with respect to well-defined indicators of welfare, with a view to terminating all or part of the study should unduly adverse consequences be observed. In another project, identification and use of humane endpoints are well-illustrated in a toxicological study of interactive effects of exposure to copper and microplastics on development in larval and post-larval seabream. A scoring scheme was set up before the experiment, with several indicators (poor body condition, abnormal swimming and activity) being screened daily and combined into a poor welfare score. Fish with scores above a pre-determined threshold would be humanely killed.
- What are the most important welfare considerations when designing research involving fish?
Broadly speaking, these are the same as with research on other vertebrates. When planning a project involving live fish, the first question is – does the information already exist in the literature? If not, then, could in vitro, ex vivo or in silico approaches answer the question, or could existing stored tissue be used? If live animals must be used, could simpler, less sentient, invertebrate animals be used? These are all questions relating to Replacement. If use of live fish is unavoidable, then it is important to undertake a thorough analysis of potential negative effects from procedures on health and welfare. This includes both acute procedures such as handling, blood sampling, and killing, and chronic procedures such as exposure to possibly adverse environmental conditions, experimentally-induced infections and new diets. Having identified such potential adverse effects, research procedures need to minimise the number of fish exposed for achieving the research goals (Reduction), and also to minimise such adverse effects (Refinement). Refinement can be achieved through various measures, including carrying out acute procedures under anaesthesia or keeping longer-term treatments as short as possible.
- Why do you think the 3Rs are so important in aquaculture research?
The 3Rs framework is a really clever way of highlighting and protecting the welfare of all animals used in research, including fish in aquaculture research. This is because it provides a clear and incisive way of focussing on all the various steps described in the previous section that must be undertaken to avoid any adverse effects on the welfare of live subjects and, where this is not possible, to minimise such adverse effects.
- Do you think fish welfare plays a major role in the public’s view of aquaculture?
The answer here depends on which sector of the public is concerned, on the intensiveness of the aquaculture, and the kind of animals – the public are less likely to show concern for crustaceans than for finfish, for example. Having said that, questionnaires indicate that members of the public are increasingly concerned about fish welfare. Several studies suggest that European consumers will pay considerably more for fish farmed under good welfare conditions, hence the success and importance of welfare accreditation schemes.
- How do you believe aquaculture research has changed over the past 10 years in relation to fish welfare?
I am not sure about the exact time scale, but certainly protecting the welfare of fish used as subjects for research is now firmly established as an important aspect of experimental design, where in the past it might not necessarily have received so much attention.
In addition, the welfare of fish in its broadest sense (including good body condition and freedom from disease) has itself become the subject of considerable attention in aquaculture research. A quick Web of Science search shows that in 1989-1990 of 127 published research papers with aquaculture as a key word, none was pulled out as involving fish welfare; the equivalent figures for 2019-2020 are 4319 research papers, 125 of which are highlighted by a welfare search. This increased interest in fish welfare has been accompanied by an expansion of the techniques available for assessing welfare; to mention just a few, these include more precise ways of measuring established indicators (stress hormones for example), new automated, hands-off monitoring systems (of movement patterns, for example) and application of the powerful resources of molecular biology/transcriptomics.
In the great majority of studies in aquaculture research, welfare is conceptualised and measured in terms of effective functioning rather than in terms of what fish feel, entirely appropriately since effective functioning is (relatively) uncontroversial and we can measure it now. However, interest in and understanding of the behavioural and cognitive complexity of fish has grown considerably in the past 10+ years, not least because of research by AQUAEXCEL2020 partners. This deeper understanding provides a rich background against which to view current measures of fish welfare and, perhaps in the future, will provide tools for probing how farmed fish feel as well as how well they function.
8. Is there something surprising you can tell us about fish welfare that most consumers would not know?
I would suppose that most consumers would not know that, as ectotherms, fish in nature and, given appropriate conditions, on fish farms, regulate their body temperature by taking advantage of thermal gradients. In nature, they can use this capacity, for example, to increase the efficiency of swimming and digestion and to increase their rates of growth and maturation. Of particular relevance in aquaculture, fish are able to ‘self-medicate’ when infected by a pathogen, stimulating an effective immune response by moving to warmer water, and since poor health impacts on welfare, improving their own welfare.
In my experience, people in the UK at least are always surprised when they hear just how smart fish are, that they learn readily and remember well, use tools, build complex structures, cooperate in complex ways with members of their own and other species, go to the aid of conspecifics that are in trouble and, in some cases and up to a point, can count the number of fish in a school. I was surprised myself by the next example, so I suppose consumers would be as well. A study published last year applied a psychological tool (called the cognitive bias test) developed to identify optimism and pessimism in humans and already used effectively on non-human mammals to a species of fish, for the first time. The cichlid species they used is monogamous and pairs for life. The study showed that if females are given no choice but to mate with the less preferred of two males (which is known to result in lower reproductive success), their affective state changes from positive (optimistic) to negative (pessimistic). Having been made aware of this kind of complexity, people often ask what this means for fish welfare; I do not know any simple answers to this challenging question.
We thank Professor Felicity Huntingford very much for talking to us about the ethics aspects in AQUAEXCEL2020and aquaculture in general.