Do fish feel pain? Advances in a scientific debate spanning centuries
Fish is man’s most important source of high-quality protein, constituting around 17% of the animal protein consumed by the world’s population, according to the Food and Agriculture Organization (FAO) of the United Nations (FAO 2014). In 2010, fish provided more than 2.9 billion people with nearly 20 percent of their intake of animal protein (FAO 2014).
Fish farming, the culture of fish for human consumption, now provides more than 50% of the fish we eat and is rapidly intensifying and diversifying (Huntingford and Kadri 2014). In recent years, the public, particularly in industrialized countries, has started to value the welfare of farmed fish (Mejdell et al. 2007). Animal welfare can be defined as animals’ health based on physical, physiological, mental or emotional parameters. Other definitions state that an animal is faring well if it performs its species-specific full range of behaviors (Hewson 2003). However, in many cases, animals kept in captivity may not be able to perform their full set of behaviors, such as migration. Fish welfare is also increasingly integrated into national animal welfare legislation globally (Mejdell et al. 2007). The fish farming industry can no longer ignore this issue, as negative publicity may affect sales of farmed fish.
Among the key signs of poor welfare in animals, pain and stress are particularly pertinent. As human beings, we are keenly familiar with the experience of pain and stress in our species. It is inevitable for us to wonder: can other species also experience pain and stress in an analogous manner?
Indeed, we have been pondering that very question for centuries. René Descartes, a French philosopher, mathematician, and scientist who attended University of Poitiers in the seventeenth century, viewed animals as merely mechanistic automatons. Descartes, who experimented on animals extensively himself, claimed that animals lacked a ‘soul’, which he believed was necessary for higher cognitive capacities such as the experience of pain and suffering. (Hepple 2005) This debate continues to this day within the scientific communities. Some modern-day researchers argue that fishes are unlikely to experience pain due to limited evidence of behavioral and neurobiological responses to injurious stimuli (Rose et al. 2014), while others argue that fishes could potentially experience pain based on anatomical findings (Sneddon 2002).
In a twenty-first century hall of University of Poitiers, we learned about more recent developments of this ongoing scientific debate. Dr. Jonathan Roques strove to address this question in relation to fish, particularly teleosts. His research broadly spanned the impact of acute and chronic stimuli on fish welfare, in aquaculture practices.
Unraveling the question of the perception of pain and stress for teleost fish is certainly not straightforward. To endeavor to investigate fish welfare requires an understanding behind the mechanisms underlying pain perception, and ultimately, consciousness. The official definition of pain, established by the International Association for the Study of Pain (IASP) in 1979 is of “an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of tissue damage, or both” (Moayedi and Davis 2013).
The experience of pain has a physiological basis; in humans, awareness of pain and stress is controlled by the highly developed hippocampus, amygdala, cerebral frontal lobes, and neocortex (Apkarian et al. 2005). In contrast, the simplicity and relatively small size of the brains of fish has led many to doubt fish have the necessary cognitive complexity to experience pain or stress (Huntingford and Kadri 2014).
Nociception is defined as the detection of potentially harmful stimuli. Fish do possess the necessary brain parts (the pons, medulla, and thalamus) required for nociceptive processing (Sneddon 2004). Nociceptors are receptors that exclusively detect noxious stimuli (Sneddon 2002). The presence of nociceptors has been confirmed in rainbow trout head (Oncorhynchus mykiss). More recently, Roques and his team found nociceptors in the tailfin of the common carp (Cyprinus carpio) (Roques et al. 2010). Clipped tissue of the common carp was analyzed at the ultrastructural level in order to identify nerve fibres classified in mammals and rainbow trout as pain fibres. Nerve bundles were found in the carp tailfin clips. The presence of nerves with characteristics of pain nerves found in trout and mammals in the carp tailfin provides further evidence that fish could experience pain (Roques et al. 2010; Sneddon 2002). Pain perception requires more than nociceptive sensory machinery, however; it also requires the cognitive capacity to feel pain in a human sense (Roques et al. 2010; Rose et al. 2014).
Quantifying pain in humans is already a challenging task, as pain is inherently a subjective and private experience (Rose et al. 2014). As patients in a doctor’s office, we are often asked to self-report the pain intensity we experience on a numerical pain scale. For non-verbal humans, such as infants, facial expression is a large component of pain assessment. Neither self-reporting nor facial expressions are possible methods for pain assessment in most animals. So how can we truly know what animals are experiencing? Even Descartes recognized that animals were capable of registering physical sensations and reacting to them accordingly. However, he did not believe these behavioral reactions were accompanied by conscious experience; rather, animals showing symptoms of suffering were viewed as merely “mechanical robots [that] could give… a realistic illusion of agony.” (Hepple 2005)
Pain or stress in fish can only be evaluated via indirect parameters, including physiological and behavioral symptoms. Possible behavioral indicators of poor welfare in fish include erratic swimming, loss of appetite, and performance of persistently repeated actions (Huntingford and Kadri 2014). The activation of a physiological stress response induces various short-term physiological effects, including the release of the hormones adrenaline and cortisol from the inter-renal glands (Huntingford and Kadri 2014). Handling fish, as in experimental studies, elicits an inherent stress response which can be difficult to distinguish from mild pain responses (Roques et al. 2010). Thus, when studying pain in fish, it is necessary to have appropriate control groups in order to differentiate between stress and pain responses (Roques et al. 2010).
In a 2010 study, Roques and his fellow researchers assessed the behavioral and stress-endocrine responses in Nile tilapia (Oreochromis niloticus) to a procedure considered to be potentially painful for fish: clipping the fishes’ tailfins. A tailfin clip was chosen as an experimental pain stimulus due to the occurrence of fin damage in both natural conditions and as a result of aquaculture practices, the presence of potentially nociceptive nervous fibers in fish fins, and the reproducibility of the procedure. Through its experimental design, the study carefully differentiated between the stress response associated with handling fish and the response to the clipping procedure. In a control group, fish were handled in the same manner as fish whose fins were clipped without administering the fin clip itself. The researchers monitored the fishes’ swimming activity, several stress parameters (plasma cortisol, glucose, and lactate), and the release of mucus content from mucus cells in the gills. The tank in which fish were kept was half-covered, leaving the tank with a ‘light’ half and a ‘dark’ half. Roques’ team found behavioral responses to the presumed pain stimulus in Nile tilapia: clipped fish increased their swimming activity and spent more time in the light section of the tank. A short-lived physiological response was also observed in response to the clipping procedure: the Nile tilapia gill’s mucus cells released their content up to one hour after the stimulus. These results revealed a clear and distinct response to the fin clip stimulus, which is likely to be painful to fish as predicted. (Roques et al. 2010)
A significant source of stress for farmed fish is linked to the process of transporting fish from aquaculture setups to slaughter houses. Roques and his colleagues studied the effects of simulated commercial transportation on the stress physiology of African catfish (Clarias gariepinus). Throughout the transportation process, fish can be subject to a variety of stressors, including handling stress, water movement, density changes, noise, vibrations, and poor water conditions. Blood samples were taken from fish before and after simulated transportation in order to analyze plasma cortisol, glucose, and non-esterified fatty acid (NEFA) levels. Other parameters used to assess the stress-related effects of the transportation process were gill histology and the number of skin lesions before and after transport. The fish were divided into three study groups: unhandled, transported, and control (with control fish being handled but not transported). Plasma cortisol levels varied between the three groups; control fish had elevated plasma cortisol levels compared to unhandled fish and transported fish had the highest plasma cortisol levels of all study groups. In transported fish, cortisol levels remained significantly elevated for 48 hours after treatment, while for control fish it took only six hours to return to baseline cortisol levels. Transported catfish were also found to have significantly more skin lesions than unhandled animals. Thus, overland transport is a stressor that can have evident physiological effects in fish, and should be considered when managing fish welfare in an aquaculture setting. (Manuel et al. 2014)
Continuing research in this realm will lead the way towards the establishment of more ethical practices in the aquaculture industry and greater societal awareness of the health and welfare of fish.
Efosa N. and Vulova S IMAE 2015-2017
Apkarian, A.V., M.C. Bushnell, R.D. Treede, and J.K. Zubieta. 2005. European Journal of Pain 9: 463-484.
FAO. 2014. The State of World Fisheries and Aquaculture 2014. Rome. 223 pp.
Hepple, B. 2005. Nuffield Council on Bioethics, London, United Kingdom.
Hewson, CJ. 2003. The Canadian Veterinary Journal 44: 496–499.
Manuel, R., J. Boerrigter, J. Roques, J. Heul, R. Bos, G. Flik, and H. Vis. 2014. Fish Physiology and Biochemistry 40: 33-44.
Mejdell C., V. Lund, and T. Halstein. 2007. Journal of Commonwealth Veterinary Association, Anniversary Celebrations 23(2): 21-26.
Moayedi, M. and K.D. Davis. 2013. Journal of Neurophysiology 109: 5-12.
Roques, J.A., W. Abbink, F. Geurds, H. Vis, and G. Flik. 2010. Physiology & Behavior 101: 533-540.
Rose, J. D., Arlinghaus, R., Cooke, S. J., Diggles, B. K., Sawynok, W., Stevens, E.D. and Wynne, C. D. L. 2014. Fish and Fisheries 15: 97–133.
Sneddon, L.U. 2002. Neuroscience Letters 319: 167-171.
Sneddon, L.U. 2004. Brain Research Reviews 46: 123-130.