Content uploaded by Lynne U Sneddon
Author content
All content in this area was uploaded by Lynne U Sneddon on May 30, 2017
Content may be subject to copyright.
93
DO PAINFUL SENSATIONS AND
FEAR EXIST IN FISH?
Lynne U. Sneddon*
Abstract
The detection of pain and fear in fi sh has been subject to much debate
and, since fi sh are a popular experimental model and commercially
important in both angling and aquaculture, many procedures that
fi sh are subjected to cause injury, fear and stress. These injuries would
give rise to the sensation of pain in humans but whether fi sh have the
capacity for pain is relatively under explored. Recent evidence has
shown that fi sh have the same neural apparatus to detect pain that
mammals and humans do, that their brain is active during a poten-
tially painful experience, that fi sh show negative changes in behaviour
and physiology and that this is reduced by administering a pain killer.
Experiments demonstrating the signifi cance of pain to fi sh have been
conducted and have shown that fi sh do not show appropriate fear and
anti- predator responses during a painful stimulation. This suggests
that they are dominated by the pain state confi rming its importance to
the fi sh. However, social context affects the aggressive behaviour of fi sh
when noxiously stimulated. In a familiar group, dominant trout perform
much less chasing of conspecifi cs yet this suspension in aggression is not
seen when placed in an unfamiliar group of fi sh. Therefore, responses to
pain are more complex and not simple refl exes. Together, these results
demonstrate that pain is an important stimulus for a trout and we
should seek to minimise and alleviate pain where possible. Studies have
demonstrated that fi sh are capable of exhibiting signs of fear including
avoidance behaviour and they may also anticipate fearful events. Recent
evidence shall be discussed with future directions suggested.
* * * * *
* The author is professor at the University of Liverpool, UK.
94
ANIMAL SUFFERING: FROM SCIENCE TO LAW
The capacity to experience negative states such as pain and fear
are integral to the doctrine surrounding animal welfare. To promote
positive welfare animals must be free from affective states that are
detrimental physically and mentally. Therefore, proving an animal
perceives pain and experiences fear provides convincing evidence
that their wellbeing can be compromised. From a moral perspective,
humans as intelligent ethical beings must ensure that the use of ani-
mals should be conducted in the most humane manner endeavouring
to maintain their health and welfare. The question of whether fi sh
possess the capacity for experiencing pain and fear is particularly
highly debated and dichotomised into those who whose opinions are
entrenched in a semantic argument over human brain anatomy1
versus scientists who conduct research producing data that demon-
strates fi sh exhibit adverse responses to pain and fear2. We have a
complicated relationship with fi sh, using them as a foodstuff, catch-
ing them for sport, employing them as experimental research models
and keeping them as pets and exhibits in public aquaria. Therefore,
understanding how the practices we subject fi sh to affects them is of
paramount importance if we are to meet minimum welfare standards.
Here, the central tenets of pain and fear that must be fulfi lled for an
animal to be considered capable of both negative affective states shall
be discussed, with evidence from scientifi c studies exploring these in
fi sh. The possibility of adverse welfare states in fi sh shall be discussed
in relation to current practices such as large scale fi sheries, angling,
experimentation and the ornamental pet trade.
PAIN: KEY TENETS
Tenet One: Neural apparatus
To detect a stimulus an organism must have the sensory system
attuned to that type of sensation whether it is olfactory (smell), gusta-
tory (taste), visual, auditory or painful. The receptors for each type of
sensory stimulation specifi cally detect these particular cues. There-
fore, receptors for perceiving potentially painful stimuli (termed noci-
ceptors) preferentially detect tissue damaging agents such as high
mechanical pressure, extremes of temperature and chemicals such
as acids. Nociceptors have been well studied in mammals and birds,
1. Iwama, G. K. 2007. The welfare of fi sh. Dis. Aquat. Organisms, 75, 155- 158.
2. Sneddon L.U. 2011 Pain perception in fi sh: Evidence and implications for the use
of fi sh. J. Consc. Stud.18, 209- 229.
LYNNE U. SNEDDON
95
but less so in non- mammalian vertebrates. Sneddon was the fi rst to
identify nociceptors in the rainbow trout in 2002 using neuroanat-
omy3 (Fig. 1) and by recording electrical activity from nociceptors on
the head of the trout (Fig. 2)4. Further research on the properties of
these fi sh nociceptors have demonstrated they are comparable with
mammalian nociceptors and respond to noxious heat and pressure
in a similar manner5. Topical application of many noxious chemicals
such as acetic acid, agents with low pH and carbon dioxide infused
water stimulate trout nociceptors. Any chemicals with such proper-
ties which fi sh encounter in their aquatic environment are, therefore,
likely to excite their nociceptors. In humans this would give rise to
the sensation of pain.
The possession of nociceptors must be accompanied by relevant
pathways from periphery and internal tissues to the central nervous
system so that the information is conveyed to the brain for processing.
If the nociceptive inputs do not ascend higher than the refl ex centres
(dorsal root ganglion) of the spinal cord or the trigeminal ganglion
in the hindbrain then the perception of the painful stimulus is often
accompanied by a refl ex withdrawal but this instantaneous percep-
tion and response are collectively termed nociception. This does not
necessarily lead to pain but may do if the information is conducted
to higher brain areas in the forebrain and midbrain and leads to a
negative affective experience associated with any damage6. There-
fore, demonstrating that fi sh have pathways to the brain and that
the higher brain is active in response to potentially painful stimuli
is particularly important. The trigeminal (head) and spinothalamic
(body) tract that are involved in conveying pain information in mam-
mals have also been identifi ed in fi sh7. Studies have shown that stim-
3. Sneddon, L. U. 2002. Anatomical and electrophysiological analysis of the trigeminal
nerve in a teleost fi sh, Oncorhynchus mykiss. Neurosci. Letts., 319, 167- 171.
4. Sneddon L.U. 2003 Trigeminal somatosensory innervation of the head of the rain-
bow trout with particular reference to nociception. Brain Res., 972, 44-52.
5. Mettam JJ, McCrohan CR, Sneddon LU. Characterisation of chemosensory trigemi-
nal receptors in the rainbow trout, Oncorhynchus mykiss: responses to chemical
irritants and carbon dioxide. J. Exp. Biol. 2012;215(4):685-93. Ashley, P. J., Sned-
don, L. U. & McCrohan, C. R. 2006. Properties of corneal receptors in a teleost fi sh.
Neurosci. Letts., 410, 165- 168. Ashley, P. J., Sneddon, L. U. & Mccrohan, C. R. 2007.
Nociception in fi sh: Stimulus- response properties of receptors on the head of trout
Oncorhynchus mykiss. Brain Res., 1166, 47-54.
6. Sneddon L.U. 2011 Pain perception in fish: Evidence and implications for
the use of fi sh. J. Consc. Stud.18, 209- 229.
7. Review in Sneddon L.U. 2004 Evolution of nociception in vertebrates: com-
parative analysis of lower vertebrates. Brain Res. Rev. 46, 123- 130.
96
ANIMAL SUFFERING: FROM SCIENCE TO LAW
ulation of the body with noxious mechanical and thermal cues does
ascend up through the spinal cord to forebrain areas8. Many critics of
fi sh pain have stated that fi sh are only capable of nociception and that
noxious stimuli simply evoke a refl ex response restricted to hindbrain
and spinal cord. Therefore, studies have set out to measure activity
in the brain of fi sh using a variety of techniques including electri-
cal recordings with electrodes placed in key brain areas7 of goldfi sh,
Atlantic salmon and rainbow trout; global gene expression determin-
ing molecular changes in trout and carp brain9; and using imaging
technology such as functional magnetic resonance imaging (MRI) in
common carp5,10. The results are convincing that painful stimulation
elicits activity in the forebrain and midbrain as well as hindbrain
and spinal cord, and that this change in activity differs from that of
innocuous, non- painful stimulation. Therefore, higher brain areas are
active during pain in fi sh. All of this evidence demonstrates that fi sh
do indeed possess the neural machinery to detect and respond to pain.
Tenet Two: Behavioural and physiological responses
Pain often motivates us to alter our behaviour to promote heal-
ing and recovery as well as preventing further pain by behavioural
alterations such as guarding behaviour where one does not use a
limb or painful area. It is crucial to demonstrate that an animal’s
behaviour is detrimentally affected by a painful stimulation and
that this is not an instantaneous refl ex response. Prolonged adverse
changes in behaviour after a painful event suggest suffering or dis-
comfort. Behavioural responses to pain are also accompanied by
measureable physiological reactions such as an increase in respira-
tion rate. Therefore, determining whether alterations in behaviour
and physiology occur after potentially painful stimulation provides
insight into the subjective experience especially if these responses
8. Dunlop, R. & Laming, P. 2005. Mechanoreceptive and nociceptive responses in
the central nervous system of goldfi sh (Carassius auratus) and trout (Oncorhyn-
chus mykiss). Journal of Pain, 6, 561- 568; Nordgreen J, Horsberg TE, Ranheim
B, Chen ACN. Somatosensory evoked potentials in the telencephalon of Atlantic
salmon (Salmo salar) following galvanic stimulation of the tail. J. Comp. Physiol.
A Neuroethol. Sens. Neural Behav. Physiol. 2007;193(12):1235-42.
9. Reilly S.C., Quinn J.P., Cossins A.R. & Sneddon L.U. 2008 Novel candidate genes
identifi ed in the brain during nociception in common carp. Neurosci. Letts. 437,
135- 138.
10. Verhoye M., Mettam J.J., Van der Linden A., McCrohan C.R., & Sneddon L.U.
The Use of Functional MRI to Assess Changes in Brain Activity During Noxious
Stimulation in the Common Carp (Cyprinus carpio). MS submitted.
LYNNE U. SNEDDON
97
are aversive and normal behaviour is suspended11. In rainbow
trout, individuals injected subcutaneously by acetic acid, a noxious
chemical, exhibit a dramatic increase in opercular (gill cover) beat
rate that remains elevated for 3 to 6 hours compared with sham
handled individuals and those injected with non- noxious saline12.
Many animals reduce activity after a painful event, and rainbow
trout and zebrafi sh do indeed show a substantial reduction in active
behaviours11. Trout and carp also perform anomalous behaviours in
response to subcutaneous injection of acetic acid into the frontal lips.
These behaviours are not seen in control animals and are reduced
by the use of an analgesic or painkiller, morphine. Therefore, these
behaviours are specifi c to pain responses in these fi sh. Reduction
of pain- related responses by the use of analgesics in mammals is
accepted as fi rm evidence that whatever alterations occurred were
indicative of pain. Analgesic drugs are now being tested in fi sh with
studies exploring side effects using robust behavioural and physi-
ological indicators that can easily be measured by the animal carer.
In the case of reduction in activity and elevation of opercular beat
rate when stimulated noxiously, lidocaine administered at the same
site signifi cantly reduced these adverse changes after 30 minutes
but other drugs injected intramuscularly did not13 (Fig. 3). Carprofen
did eventually have some action reducing some of these responses,
however, buprenorphine was not effective in trout. These studies
demonstrate that it is indeed nociceptive pathways that are respon-
sible for the changes in behaviour and that normal feeding, swim-
ming and physiological functions are detrimentally affected. These
alterations would be acceptable as indicators of pain in mammals
and since they are prolonged and are not immediate, short- lasting
refl ex responses, these suggest that there is a negative experience
associated with painful stimulation in fi sh.
11. Sneddon, L. U. 2009. Pain perception in fi sh: Indicators and endpoints. Ilar Jour-
nal, 50, 338- 342. Sneddon L.U. 2011 Responses to Nociception or Pain in Fish. n:
Farrell A.P., (ed.), Encyclopedia of Fish Physiology: From Genome to Environment,
volume 1, pp. 713–719. San Diego: Academic Press.
12. Reilly, S. C., Quinn, J. P., Cossins, A. R. & Sneddon, L. U. 2008. Behavioural analy-
sis of a nociceptive event in fi sh: Comparisons between three species demonstrate
specifi c responses. Applied Animal Behaviour Science, 114, 248- 259. Ashley, P. J.,
Ringrose, S., Edwards, K. L., Wallington, E., Mccrohan, C. R. & Sneddon, L. U. 2009.
Effect of noxious stimulation upon antipredator responses and dominance status
in rainbow trout. Animal Behaviour, 77, 403- 410.
13. Mettam J.J., Oulton L.J., McCrohan C.R. & Sneddon L.U. 2011 The effi cacy of
three types of analgesic drugs in reducing pain in the rainbow trout, Oncorhyn-
chus mykiss. Appl. Anim. Behav. Sci. 133, 265- 274.
98
ANIMAL SUFFERING: FROM SCIENCE TO LAW
Tenet Three: Consciousness
To be able to suffer, one must be consciously aware of this nega-
tive mental state, such that when experiencing pain we must know
we are in pain and that we hurt. This conscious awareness is impos-
sible to measure but as humans who are capable of communicating
complex concepts through language we can convey whether we are in
pain to one another and the degree of such pain as well as intensity,
duration and so on. However, humans can exaggerate and if a person
has no means of communication we cannot know how much pain they
are in. This is very much the context that we operate in when assess-
ing animal pain. How can we get into the animal mind? Unless one
has been an animal it is impossible to know what they experience
and unscientifi c to dismiss the capacity for pain or fear since this is
impossible to measure directly. However, clever experimentation with
consideration for the behaviour, ecology, life history and evolution of
an animal can open up routes into obtaining meaningful information
on the subjective experience of animals. Fish live in a very different
world to humans and have evolved to meet the demands of an aquatic
life. Critics of fi sh pain anthropomorphise pain stating that animals
must have the same brain anatomy as humans to be capable of con-
sciously experiencing pain. Yet many studies have shown that fi sh
are capable of complicated behaviours with a relatively smaller brain
and they are one of the most successful animal groups. As discussed
previously, the fi sh forebrain and midbrain are active during painful
stimulation and fi sh exhibit complicated changes in behaviour and
physiology that are ameliorated by painkilling drugs. Therefore, fi sh
are likely to experience pain but it may be more primitive than that
experienced by mammals, however, it not feasible to measure this
so we cannot make a fi rm conclusion. Welfare assessment should be
based upon sound scientifi c approaches. Certainly, many scientists
take the view that as we know very little about the neurobiology of
consciousness in humans that it would be foolish to make judgements
on animals using this criterion14. Yet other scientists are of the opin-
ion that we know enough regarding consciousness in animals and
that the absence of the human neocortex does not prevent an animal
having some form of conscious awareness nor experiencing negative
affective states associated with suffering15.
14. Dawkins M.S., 2012. Why Animals Matter. Animal consciousness, animal welfare,
and human well- being.Oxford: Oxford University Press, pp 224.
15. http://fcmconference.org/img/CambridgeDeclarationOnConsciousness.pdf.
LYNNE U. SNEDDON
99
Consciousness studies often employ the mirror self recognition
test where animals are able to recognise themselves in a mirror. These
have failed for fi sh and the species tested often react to their own image
by attacking it seeing the refl ection as an intruder or competitor. How-
ever, we must consider the evolution and ecology of fi sh – when would
they come into contact with their own mirror image? Terrestrial ani-
mals would come to water bodies to drink and would see their own
refl ection but this is precluded by living under water. This difference
in ecology would infl uence whether mirror self recognition would work
and explains why fi sh have not evolved to recognise themselves in this
way16. However, fi sh can recognise themselves through smell and con-
sidering how fi sh live in a world where light is fi ltered out at depth,
a reliance on other forms of communication are especially important.
Cichlid fi sh can recognise their own odour distinct from others but also
distinct from closely related kin17. Therefore, this is evidence for self
recognition and the ability to discriminate one owns smell from others.
In terms of pain, one of the central pieces of evidence is whether
animals will self medicate with painkillers when in pain or are will-
ing to pay a cost to access such pain relief. Many studies in birds and
mammals have shown animals will eat food dosed with analgesics
upon experiencing painful stimuli. However, fi sh suspend feeding
until they have recovered from a painful event. In order to determine
whether fi sh will pay a cost to accessing pain relief, zebrafi sh were
given access to two chambers, one of which was enriched with gravel,
plants and a live shoal behind a transparent barrier. The other cham-
ber was made unfavourable by being barren and brightly lit. Fish
selected the enriched chamber to spend most time in and when they
had selected the chamber six consecutive times they were assigned
to a noxiously stimulated group which had acetic acid injected sub-
cutaneously and a control group with innocuous saline injected. Half
of each group were then re- tested and continued to spend most time
in the enriched chamber. However, when an analgesic was added to
the unfavourable chamber only fi sh experiencing pain spent time
in this chamber shifting their preference. This demonstrates that
fi sh sought analgesia and were willing to pay the cost of being in a
brightly, lit barren area where their pain was reduced18 (Fig. 4). This
16. Lev- Yadun S. & Katzir G., 2012 Mirror images: Fish versus terrestrial animals.
J. Theoret. Biol. 182- 184.
17. Thunken, T., Waltschyk, N., Bakker, T. C. M. & Kullmann, H. 2009. Olfactory
selfrecognition in a cichlid fi sh. Animal Cognition, 12, 717- 724.
18. Sneddon L.U. Zebrafi sh shift preferences to access analgesia. MS submitted.
100
ANIMAL SUFFERING: FROM SCIENCE TO LAW
is compelling evidence for a negative affective component when fi sh
experience a painful event.
FEAR: KEY TENETS
Tenet One: Neural apparatus
As with pain, the neural machinery required to detect and react
to fear- causing stimuli must be comparable with the mammalian
brain circuitry. Fear is generally sensed as an external threat to the
whole animal. For example, the predator test is a standard fear para-
digm in experiments where an animal is exposed to the sight, odour
or some other cue of a predator that elicits a fi ght or fl ight response.
Thus fear stimuli are psychological threats to the survival of the
whole animal and fear motivates the animal to make an appropriate
defensive response such as freezing, hiding or fl eeing. Fear can either
be innate or unlearned whereby the stimulus elicits a fear response
without the animal previously being exposed to the stimulus (e.g. the
predator test) or fear can be learned and in many experimental stud-
ies animals are provided with a non- threatening cue or conditioned
stimulus (CS) such as an innocuous light or sound paired with the
presentation of a fear causing stimulus such as chasing or confi ne-
ment (unconditioned stimulus; US) a few seconds later. After repeated
trials of the CS-US the animal learns to respond to the CS or innocu-
ous cue by showing a fear response in the absence of the actual fear
stimulus. Rodent models have been employed in such paradigms
investigating the neuronal circuitry and the mammalian amygdala
and hippocampal regions are particularly important in mediating
emotions especially fear learning and memory19. Experiments in
fi sh have shown comparable behaviours, cognitive mechanisms and
brain areas that are homologous to the fear circuitry in mammals.
For example, the dorsomedial telencephalon in the forebrain area of
goldfi sh has identical functions in fear conditioning as the amygdala
of mammals mediating fear responses and learning whereas the gold-
fi sh dorsolateral telencephalon is homologous to the mammalian hip-
pocampus involved in spatial learning and retrieval of memories20.
19. Ashley P.J. & Sneddon L.U. 2008. Pain and fear in fi sh. In: Fish Welfare, Branson
E.J. (Ed), Oxford: Blackwell Publishing, pp 61-77.
20. Portavella M., Vargas J.P., Torres B. & Salas C. 2002. The effects of telencephalic
pallial lesions on spatial, temporal and emotional learning in goldfi sh (Carassius
auratus). Brain Res. Bull. 57, 397- 399.
LYNNE U. SNEDDON
101
The mammalian habenula is an evolutionarily highly conserved
diencephalic brain structure subdivided into medial and lateral
regions (MHb and LHb, respectively). The LHb sends efferent neurons
to monoaminergic neurons and has been implicated in the control of
aversive learning and emotional behaviours. The MHb projects to the
interpeduncular nucleus (IPN), and regulates fear responses. The
zebrafi sh dorsal habenula (dHb) also connects with with the interpe-
duncular nucleus (IPN) and is equivalent to the mammalian medial
habenula. Anatomically the habenula system in zebrafi sh is similar
(Fig. 5) and studies have sought to address its function by silencing
this system during fear responses. Genetic inactivation of the dHb
resulted in zebrafi sh that froze rather than the normal fl ight response
to a conditioned fear stimulus (Fig. 6), suggesting that the dHb- IPN
pathway is important for controlling fear responses21.
Tenet Two: Consistent behavioural response
Fear responses should generate a coherent set of behavioural
and physiological reactions. Measurements of startle, freezing and
other defensive behaviours can be coupled with physiological param-
eters such as heart rate and release of stress hormones, for example,
cortisol. Studies in fi sh have demonstrated a consistent response to
threatening stimuli such as avoidance of novel objects, freezing to
reduce conspicuousness; escape or fl eeing behaviours; thigmotaxis
where the fi sh swims next to tank walls avoiding open, central areas;
sinking to depth; fast start swimming and diving responses; and anti-
predator behaviours22. Many rodent tests of fear and anxiety are now
routinely applied to fi sh species such as open fi eld, novel object, clas-
sical conditioning, avoidance learning, predator cues and scototaxis
(preference for darker areas). Combined with studies on pain, fear
responses can be evaluated as to whether pain or fear is more impor-
tant. In rainbow trout, fi sh show a classic anti- predator response to
alarm substance by performing increased escape responses and also
hiding under cover. When trout were given a pain stimulus they did
not perform correct fear responses and did not increase their use of
21. Agetsuma M., Aizawa H., Aoki T., Nakayama R., Takahoko M., Goto M., Sassa T.,
Amo R., Shiraki T., Kawakami K., Hosoya T., Higashijima S. & Okamoto H. 2010.
The habenula is crucial for experience- dependent modifi cation of fear responses
in zebrafi sh. Nat. Neurosci. 13, 1354- 1356.
22. Review in Maximino C., Marques de Brito, T., Waneza da Silva Batista A., Her-
culano A.M., Morato S. & Gouveia Jr A. 2010. Measuring anxiety in zebrafi sh: A
critical review. Behav. Brain Res. 215, 157- 171.
102
ANIMAL SUFFERING: FROM SCIENCE TO LAW
cover nor perform escape reactions demonstrating in this context pain
was the imperative23 (Fig. 7).
Tenet Three: Anti- anxiety drugs
The fi nal key criterion that animals must fulfi l is demonstrat-
ing that anti- anxiety drugs reduce any fear responses such as those
described above. Many agents are used to decrease fear and anxi-
ety including benzodiazepines, opioids, cholinergic and serotonergic
agents. Benzodiazepines are a major class of drugs used to treat
human anxiety disorders and have been shown to reduce fear in
mammalian models. Benzodiazepines act by enhancing the action of
a neurotransmitter, GABA (gamma- aminobutyric acid) which has an
inhibitory infl uence thus exerts a sedatory effect. Binding sites for
these drugs are found in comparable brain areas of fi sh and several
experiments have shown they reduce fear responses in zebrafi sh24.
Piracetam, a derivative of GABA, is prescribed to reduce clinical anxi-
ety in humans. Chronic administration of piracetam also reduces fear
behaviour in zebrafi sh where fi sh spend more time in a white area in
a scototaxic (light versus dark chamber) test25 (Fig. 8).The opioider-
gic system has a key role in the modulation of human and animal
fear. Fish possess a functional opioidergic system, including both
opioid peptides and their receptors akin to the mammalian system.
Opioid administration in zebrafi sh in a fear test reduced the amount
of erratic, fl ight swimming23. Serotonergic mechanisms are not only
implicated in depression but also animal anxiety. Selective serotonin
reuptake inhibitors (SSRIs) are potent modulators of brain serotonin
and many of these drugs have been employed in mammalian studies
seeking to reduce fear. Zebrafi sh have a well- developed serotonergic
system but this is not anatomically nor genetically identical, how-
ever, many fi sh serotonin receptors have similar expression patterns,
23. Ashley P.J., Ringrose S., Edwards K.L., Wallington E., McCrohan C.R. & Sned-
don L.U. 2009 Which is more important in fi sh: pain, anti- predator responses or
dominance status? Anim. Behav. 77, 403- 410.
24. Review in Stewart A., Wu N., Cachat J., Hart P., Gaikwad S., Wong K., Utterback
E., Gilder T., Kyzar E., Newman A., Carlos D., Chang K., Hook M., Rhymes C.,
Caffery M., Greenberg M., Zadina J. & Kalueff A.V. 2011. Pharmacological modu-
lation of anxiety- like phenotypes in adult zebrafi sh behavioral models. Progress
in Neuro- Psychopharmacol. Biol.l Psychiat. 35, 1421–1431.
25. Grossman L., Stewart A., Gaikwad S., Utterback E., Wu N., DiLeo J., Frank K.,
Hart P., Howard H., Kalueff A.V. 2011. Effects of piracetam on behavior and
memory in adult zebrafi sh. Brain Res. Bull. 85, 58-63.
LYNNE U. SNEDDON
103
binding, and physiological properties compared with mammals. As
with rodent and human clinical studies on the use of SSRIs, clear
anxiolytic action or diminished fear responses of chronic fl uoxetine
has been recorded in zebrafi sh23. The cholinergic system relates to
the sympathetic and parasympathetic nerve fi bres or neurons in
which acetylcholine (ACh) is the neurotransmitter liberated at a
synapse. Cholinergic receptors are of two types: nicotinic receptors,
which are situated in striated muscles and muscarinic receptors,
which are situated in parasympathetically innervated structures.
Low choline levels have been related to high anxiety in humans,
therefore, attention is now turning to the cholinergic system as a
new target for reducing fear. Zebrafi sh administered with nicotine
(nicotinic- cholinergic agonist) were more active and spent less time
at the bottom compared with untreated fi sh who displayed a classic
fear response of freezing and remaining on the bottom of the tank in
a novel tank test. Thus, the neurobiological mechanisms of fear and
the impact of selective drugs to reduce fear in humans and mammals
are also apparent in fi sh.
IMPLICATIONS FOR THE USE OF FISH
Fish do fulfi l the criteria for both animal pain and fear and if we
are to accept that many of the procedures we apply during our use of
fi sh are likely to cause tissue damage giving rise to the sensation of
pain and may also be considered life threatening then these are likely
to evoke fear. The experimental data for both pain and fear in fi sh are
accepted for mammals yet why do some authors reject these when
the same indicators are presented? Perhaps this is due to the varied
functions that fi sh serve as a foodstuff, sport, pet and experimental
model. If one enjoys catch and release angling as a means of leisure
this does involve hooking the fi sh causing injury as well as suffocating
the fi sh in air to retrieve the hook. One could argue that if the fi sh is
killed quickly and humanely and used for food then there is a benefi t
to the human that outweighs the cost to the fi sh. However, catch and
release does involve of course the return of live fi sh to the water body
for the pleasure of the fi sher at the expense of any impact upon the
wellbeing of the fi sh. Fish are also farmed in high densities to pro-
vide protein for our growing populations and many of the practices
such as vaccination, size grading, handling, and slaughter will often
result in damage or situations which may result in fear. Studies are
exploring ways of improving fi sh welfare in aquaculture which would
improve economic return on healthy, well grown fi sh. The amount of
104
ANIMAL SUFFERING: FROM SCIENCE TO LAW
fi sh caught at sea outnumber the number of terrestrial animals used
for food and not only are the target species of fi sh caught but many
unwanted species are captured and discarded in the process. Fish are
also a popular experimental model and as described above much of the
experimental data collected is very similar to mammals26 and small
species such as zebrafi sh are much easier and economically cheaper
to maintain than traditional rodent models. Finally, the ornamental
pet trade in both freshwater and marine fi sh is an important indus-
try with fi sh now being the third most popular pet behind cats and
dogs. The purchase of an aquarium set-up and addition of fi sh does
not require any licensing or training.
The impact upon fi sh welfare does ignite contentious debate,
however, how can we reduce the impact we humans have? Is it fea-
sible for humans to stop eating fi sh? If you accept that fi sh suffer
pain and fear when caught by current fi shing practices, it may be a
decision that is made by the individual. Alternatively, more welfare
friendly solutions could be proposed by improving fi shing gear, cap-
ture methods, refi ning the procedure of capture and slaughter so it
less invasive, causes less damage and enhances welfare27. Can the
time between capture of fi sh at sea and discarding be reduced such
that the chances of survival are better? The public drive improve-
ments in animal welfare and it is public opinion that would pro-
vide the strongest motivation for enhancing fi sh welfare during
large scale fi sheries. The public are willing to pay more for animals
farmed under better welfare28 or from sustainable stocks, therefore,
this could and has been applied to the source and method of catch-
ing wild fi sh29. Clearly, there are improvements that may be made to
current fi shing methods, specifi cally, reduce the time spent fi shing
so that fi sh are landed more quickly, reduce the injuries sustained
to the fi sh by improving equipment; use of quick, effi cient humane
killing techniques on board; and reduce bycatch30. These may be
26. Sneddon L.U. 2011 Cognition and Welfare in Fish. In Fish Behaviour and Cogni-
tion, 2nd edition (eds Brown et al.) Oxford: Blackwell Publishing, Chapter 17.
27. Metcalfe, J.D. 2009. Welfare in wild- capture marine fi sheries. Journal of Fish Biol-
ogy 75, 2855- 2861.
28. Glass, C.A., Hutchinson, W.G. & Beattie, V.E. 2005. Measuring the value to the
public of pig welfare improvements: a contingent valuation approach. Animal
Welfare 14, 61-69.
29. http://www.fair- fi sh.ch/english/; http://www.msc.org/.
30. Sneddon L.U. & Wolfenden D.C.C. 2012 How do large- scale fi sheries affect fi sh:
Pain perception in fi sh? In Sea The Truth: Essays on Overfi shing, Climate Change
and Pollution, Nicolaas G. Pierson Foundation.
LYNNE U. SNEDDON
105
conveniently considered under the following categories but these
are not restrictive:
– Reducing the initial numbers of non- target species (bycatch)
captured.
– Increasing the survival chances of discarded bycatch.
– Bycatch should be included in fi shing quotas.
– Reducing the duration of the capture experience.
– Mitigating the stressful experience of slaughter for target
species.
– Adjusting fi shing practices to exclude the use of live bait fi sh.
Many public and government bodies now consider fi sh to be
capable of perceiving pain and as a consequence suffer when injured.
Regulations are strict when considering the use of fi sh in scientifi c
experimentation (e.g. Scientifi c Procedures Act in the UK31; guide-
lines on scientifi c research in USA32). Farmed fi sh are also subject to
scrutiny and the European Food Safety Authority (EFSA) also con-
sider fi sh to be capable of suffering when subject to poor welfare33.
The Norwegian Scientifi c Committee for Food Safety have proposed
enhancements to recreational catch and release angling to minimise
pain and poor welfare during the practice of catching fi sh for sport
or food by individuals34. Therefore, we should apply these principles
of diminishing the impact of large scale fi sheries on fi sh welfare by
demanding better methods of fi shing. Some authors misguidedly
suggest that it is acceptable to treat wild fi sh in any way and have
little or no regard for their wellbeing as we should consider ourselves
as predators35. However, natural predators only kill to satiate their
hunger and stop once satisfi ed. They do not kill many other non- target
animals in the process of killing the fi sh that they consume and they
do not massively disrupt and destroy the environment when doing so.
To deliberately cause injury and suffering is unethical and as moral
beings we have a duty of care to animals that we place in the com-
31. http://www.homeoffi ce.gov.uk/science- research/animal- research/.
32. http://www.nap.edu/catalog.php?record_id=12526.
33. http://www.efsa.europa.eu/en/scdocs/doc/ahaw_op_ej954_generalfi shwelfare_
en.pdf.
34. http://english.vkm.no/dav/42f495efaf.pdf.
35. Diggles, B.K., Cooke, S.J., Rose, J.D. & Sawynok, W. 2011. Ecology and welfare of
aquatic animals in wild capture fi sheries. Reviews in Fish Biology and Fisheries
21, 739- 765.
106
ANIMAL SUFFERING: FROM SCIENCE TO LAW
pletely unnatural environment of fi shing equipment. The scientifi c
evidence that fi sh are capable of pain perception and of experiencing
fear cannot be ignored and we must factor this into our treatment of
fi sh regardless of the context.
ACKNOWLEDGEMENTS
I am grateful to the organisers Martine Lachance and Thierry
Auffret Van Der Kemp for inviting me to participate in the LFDA
and GRIDA symposium Animal suffering: from science to law. I wish
to thank David C.C. Wolfenden for his useful comments on this text
and NC3Rs for Research Grant funding.
FIGURE LEGENDS
Figure 1. Section of the maxillary branch of the trigeminal
nerve of the rainbow trout showing the presence of A-delta and C
fi bres that may act as nociceptors (×1000, scale bar=2 µm. Adapted
from Sneddon, L. U. 2002. Anatomical and electrophysiological
analysis of the trigeminal nerve in a teleost fi sh, Oncorhynchus
mykiss. Neurosci. Letts., 319, 167- 171 by kind permission from
Elsevier).
Figure 2. Electrophysiological recordings from a nociceptive
receptive fi eld on the trout face showing responses of nociceptors to
heat stimulation. The instantaneous fi ring frequency (IFF) is dis-
played in the centre as scatter graphs This illustrates sensitization
of a mechanothermal receptor to heat following noxious chemical
stimulation. The fi ring response to ramp and hold heat stimulation
is shown (A) before and (B) 9 min after subcutaneous injection of 1%
formalin < 1 mm from the receptive fi eld. Upper trace shows heat
stimulus, middle trace plots instantaneous fi ring frequency (IFF)
and lower trace shows extracellular single unit recording from the
trigeminal ganglion. Thermal threshold remains the same but fi ring
frequency is greatly increased following formalin injection. (Adapted
from Ashley P.J., Sneddon L.U. & McCrohan C.R. 2007 Nociception in
fi sh: stimulus–response properties of receptors on the head of trout
Oncorhynchus mykiss. Brain Res. 1166, 47-54 by kind permission
from Elsevier).
LYNNE U. SNEDDON
107
Figure 3. The percentage change in (A) activity and (B) oper-
cular beat rate (OBR) performed by rainbow trout 30 minutes after
they were injected subcutaneously with saline or a noxious sub-
stance, 0.1% acetic acid (Acid) or acid combined with intramuscular
injection of 0.1mg/kg buprenophine (0.1 Bup) or 5mg/kg carprofen
(5mg/kg Car) or injected at the same site as the acid with 1mg
lidocaine.(1.0 Lid). The grey line represents the impact of saline
(control) treatment whereas the black line represents the impact of
pain (acid injection; adapted from Mettam J.J., Oulton L.J., McCro-
han C.R. & Sneddon L.U. 2011 The effi cacy of three types of anal-
gesic drugs in reducing pain in the rainbow trout, Oncorhynchus
mykiss. Appl. Anim. Behav. Sci. 133, 265- 274 by kind permission
from Elsevier).
Figure 4. Time spent in either a favourable or unfavourable
chamber by zebrafi sh that were injected subcutaneously with saline
(Control) or injected with 1% acetic acid (Acid) when analgesia was
present (+ Analgesia) or absent (-analgesia) in the unfavourable cham-
ber. When analgesia was present zebrafi sh spent more time in the
unfavourable chamber (*P<0.001; Sneddon, MS submitted).
Figure 5. Diagrammatic representation of the zebrafish
habenula (Hb) system. Asymmetric pathways from dorsal Hb (dHb)
to the interpeduncular nucleus (IPN) and parallel pathway from
ventral Hb (vHb) to the median raphe (MR). Dorsal oblique view on
the left and sagittal view on right. Red, dHbL (lateral)-d/iIPN path-
way; green, dHbM (medial) -v/iIPN pathway; blue, vHb-MR pathway
(Adapted from Agetsuma M., Aizawa H., Aoki T., Nakayama R., Taka-
hoko M., Goto M., Sassa T., Amo R., Shiraki T., Kawakami K., Hosoya
T., Higashijima S. & Okamoto H. 2010. The habenula is crucial for
experience- dependent modifi cation of fear responses in zebrafi sh. Nat.
Neurosci. 13, 1354- 1356. Suppl. Info.).
Figure 6. Examples of the control (a) and dHbL- silenced
(b) zebrafi sh locomotion trajectories during retrieval sessions, before
(20 s, red dotted lines), during (8.5 s, red solid lines) and after (20 s,
blue lines) the conditioned stimulus (CS) exposure. Silenced fi sh did
not show the classic fear conditioned response (Adapted from Aget-
suma M., Aizawa H., Aoki T., Nakayama R., Takahoko M., Goto M.,
Sassa T., Amo R., Shiraki T., Kawakami K., Hosoya T., Higashijima S.
& Okamoto H. 2010. The habenula is crucial for experience- dependent
108
ANIMAL SUFFERING: FROM SCIENCE TO LAW
modification of fear responses in zebrafish. Nat. Neurosci. 13,
1354- 1356).
Figure 7. (a) The median (interquartiles) change in percentage
time spent active in bold and shy fi sh injected with either saline (con-
trol) or acid (acid) from before to after the addition of alarm substance
(predator cue). (b) The median change in duration of time spent under
cover by bold and shy fi sh in the control and acid groups from before to
after the addition of alarm substance. The arrows indicate the impact
of pain upon these behaviours (*P < 0.01. N = 24; Adapted from Ashley,
P. J., Ringrose, S., Edwards, K. L., Wallington, E., Mccrohan, C. R. &
Sneddon, L. U. 2009. Effect of noxious stimulation upon antipredator
responses and dominance status in rainbow trout. Animal Behaviour,
77, 403- 410 by kind permission from Elsevier).
Figure 8. Behavioural effects of chronic piracetam (200 mg/L
for 7 days; n = 20–23 per group) on adult zebrafi sh tested in a light–
dark box (day 8) showing time spent in white chamber (#P<0.05;
Adapted from Grossman L., Stewart A., Gaikwad S., Utterback E., Wu
N., DiLeo J., Frank K., Hart P., Howard H., Kalueff A.V. 2011. Effects
of piracetam on behavior and memory in adult zebrafi sh. Brain Res.
Bull. 85, 58-63 by kind permission from Elsevier).
Figure 1
A-alpha
A-beta
A-delta
C fi bre
LYNNE U. SNEDDON
109
Figure 2
Figure 3
110
ANIMAL SUFFERING: FROM SCIENCE TO LAW
Figure 4
Figure 5
LYNNE U. SNEDDON
111
Figure 6
Figure 7
112
ANIMAL SUFFERING: FROM SCIENCE TO LAW
Figure 8