How fish think and feel, and why we should care about their welfare

Abstract

People often judge animal intelligence by how much their behaviour resembles ours. The lack of expression on fish faces and their very different lifestyles make it easy for people to denigrate them as unfeeling, simple creatures. But recent studies imply that fish intelligence and sentience is comparable to that of other vertebrate groups.

Full text

Fish with Feelings
Sea cucumbers Native problems Masters of the forest
AUTUMN 2016 VOL. 53 NO. 1 $12.50 INC GST

THEM & US
People often judge animal
intelligence by how much their
behaviour resembles ours. The lack
of expression on fish faces and
their very different lifestyles make
it easy for people to denigrate
them as unfeeling, simple creatures.
But recent studies imply that
fish intelligence and sentience
is comparable to that of other
vertebrate groups. This Australian
blenny (Ecsenius australianus) is a
denizen of the Great Barrier Reef
and Coral Sea. Photo: Robin Jeffries
10 | Wildlife Australia | AUTUMN 2016

Don’t be fooled by their lack of expression. Behavioural ecologists Culum Brown and
Catarina Vila Pouca reveal fish to be thinking, feeling animals that deserve a better deal.
AND WHY WE SHOULD CARE ABOUT THEIR WELFARE
Watching fish swim around a home aquarium is soothing
and said to be good for human health. Although pretty,
they seem to be rather dull, simple creatures. But
consider this – if you feed them daily at the same time and same
place, they soon learn the routine and wait for you there. And if
you feed them at one end in the morning and the opposite end
in the evening, they also learn this routine. There may be more
going on in their brains than you think.
One group popular in aquariums are the rainbowfish (genus
Melanotaenia), which are native to Australia and New Guinea.
They are definitely smarter than they look and lead complex
social lives. They recognise each other, and females prefer to
shoal with individuals they know rather than strangers. The
males have a social hierarchy, and court females by displaying
to them and chasing them. One of us (Culum) has spent many
hours in the laboratory and the field putting them through
their intellectual paces. When taught to swim through a hole to
avoid getting trapped in a net, they proved to be quick learners,
needing no more than five attempts (about an hour’s training).
Those in groups of five did better than those in pairs, showing
they learned from each other. And even though rainbowfish
live only two or so years in the wild, they still remembered the
escape route a year later, without having seen the apparatus
in the meantime. They also learned, like Pavlov’s famous dogs,
to associate a signal (a light) with food – outshining rats by
learning the association within 14 trials compared to about 40
for rats (aided by a tone) – and to avoid particular locations or
microhabitats associated with a predator.
Rainbowfish clearly have brain power – but they are not at all
exceptional for fish. Since Culum started that research 20 years
ago, evidence has piled up to overturn the reputation of fish as
dull, unfeeling creatures with simple behaviours.
No one could fail to be impressed by the Machiavellian social
intelligence of cleaner wrasses, for example, which rivals that
of many primates. They maintain ‘cleaning stations’ on coral
outcrops, where they remove parasites and dead skin for a
wide range of clients.
Studies by Redouan
HOW FISH
THINK AND FEEL
It might not be obvious in an aquarium but rainbowfish have good memories, a capacity for social learning, and many other mental capacities. They also lead
complex social lives. This crimson-spotted rainbowfish (Melanotaenia duboulayi) is native to streams in eastern Australia. Photo: Culum Brown
They are definitely
smarter than they
look and lead
complex social lives.
wildlife-australia.org | Wildlife Australia | 11

Bshary have revealed that cleaners recognise each of their regular
clients, sometimes numbering more than 100, and treat customers
differently according to categories. For example, they give priority
to visitors (the locals have nowhere else to go, so will wait), and
they only cheat on non-predators (by nipping some skin for its
nutritious mucous). Reputation and public relations are important
to cleaner wrasses. They cheat less when being watched by
potential clients or when they face competition for their services,
and if a client leaves in a huff after being bitten, they seek to
reconcile by chasing after and giving it a ‘back rub’. In fish a back
rub can reduce stress hormone levels just as it does in humans.
What we are learning about fish conflicts with widely held beliefs
that justify public disinterest in how they are treated. A major
reconciliation is needed – of the moral status of fish with the
evidence of their intelligence and capacity to suffer.
Fish and animal welfare laws
Animal welfare laws are shaped by public perceptions of animals’
intelligence and consciousness. The greatest concern is felt for
great apes, and primates in general, because
they are considered smarter than other animals,
and most like us in their capacity to suffer.
But as research continues to reveal that the
mental capacities of animals of many types are
much greater than realised, the compassion
has spread, particularly to other mammals.
Animal welfare laws are now much less biased
to certain species, and generally recognise that
all vertebrates warrant welfare concern – except
fish. Fish are mostly either excluded from welfare
laws (Western and South Australia for example exclude fish from
the definition of ‘animal’) or fishing is exempted from welfare laws.
Every major commercial animal production system nowadays
has some animal welfare requirements – inadequate though they
are – except those involving fish. Only under laws for scientific
institutions are fish treated like other vertebrates in requiring
welfare assessments and certain levels of care.
The problem is that most people think of fish as ‘lower’ species,
more primitive than other vertebrates, and good mainly for eating
or as low maintenance pets. More fish are eaten and more are
kept as pets than any other type of animal. Because they lack
recognisable facial expressions and we don’t hear them vocalise,
they don’t generate much empathy, and few of us get to see them
behaving or interacting in their natural environment. As a result,
they are mostly excluded from our ‘moral circle’. We don’t consider
their needs and wants, and fail to understand the cruelty inflicted
on them. Billions of fish are caught each year in huge trawling nets.
Most are crushed to death when the net is pulled up, and others
suffer trauma from the pressure changes. If they survive that, they
suffocate on the deck.
The other problem is that the fishing industry has powerful
lobbyists and the commercial value of these practices is high.
Fishing is also a popular recreation – and not regarded in the
same way as other forms of hunting. It is even permitted in many
national parks, where it is illegal to pick flowers.
Primitive?
The fallacy that ancient forms of life are primitive underpins the
ill-treatment of fish. Scientists used to understand evolution as a
linear progression from inferior to superior forms, with humans
at the top of the evolutionary ‘ladder’. This idea comes from the
time of Aristotle but remains popular today and is still taught in
some schools. Fish have an ancient lineage – they have existed for
over 500 million years and all other vertebrates can be traced to a
fish ancestor that lived around 360 million years
ago – but ancient does not equate to primitive.
With more than 32,000 species, more than all
other vertebrate species combined, fish are a very
successful and highly diverse group of animals.
They have continued to evolve, and most fish
species alive today emerged around the same
time as humans. Evolution tends to be highly
conservative, preserving important traits through
time, so it is not surprising that fish and humans
have many features in common, including in
our brains and behaviour, due to our shared ancestry. We are
essentially fish with a few tweaks.
Intelligent?
Fish have a sensory system as good as or better than our own
– the smelling ability of sharks, for example, is about 10,000
times more sensitive than ours, and with four or more cones in
their eyes, compared to three in humans, fish see more colours
than we do. But how do they process and act on all the sensory
information they obtain? Understanding the cognitive capacities
of animals is important for animal welfare, for intelligent animals
are more likely to suffer from mistreatment because they can
remember and anticipate stressful and painful events. Here are
just a smattering of examples showing that fish intelligence has
been vastly underestimated.
The problem is
that most people
think of fish as
‘lower’ species.
Frillfin gobies (Bathygobius soporator) live in shallow intertidal rocky areas on both sides of the Atlantic Ocean. By roaming over their neighbourhood
at high tide, they build up a mental map of their surroundings that allows them to escape predators at low tide by jumping into neighbouring pools
without being able to see their destination. Photo: Kevin Bryant
12 | Wildlife Australia | AUTUMN 2016

Memory: Contrary to the myth that fish can remember things
only for three seconds, like Dory from the film Finding Nemo, they
actually have exceptional memories. Some fish avoid hooks for
as long as a year after being caught once. Rock-dwelling gobies
are champions of spatial memory. They roam in rock pools at
high tide and build a mental map of their surroundings that
allows them to jump into another pool if they are disturbed
at low tide. Even after being moved 30 metres they can return
home, and they retain their mental map for weeks after being
moved. Port Jackson sharks, native to Australian waters, quickly
learn to associate air bubbles with a food reward and remember
this association for longer than a month.
Recognition of self and others: Most fish live in social groups, so it
is not surprising they can recognise individuals. Guppies easily
learn the identities of up to 15 others. Fish prefer to shoal with
familiar individuals, and are better able to learn from each
other and avoid predators than when they are with strangers.
There is recent evidence that some fish are capable of self-
recognition, regarded as one of the key criteria of consciousness.
Timo Thünken and others showed that male cichlids recognise
their own odour, which they preferred to the odours of brothers
or unrelated males. Most fish can also determine how closely
related they are to other individuals.
Social learning and traditions: Social traditions are thought to
be responsible for the migration routes of many species. Gene
Helfman and Eric Schultz showed this was the case for the daily
feeding routes of French grunts. These inhabitants of tropical
reefs in the Western Atlantic and Caribbean Sea hide by day
among sea urchins and migrate together along the same route
each night to their feeding grounds. When the researchers moved
some to new empty sites, they headed off in the same direction
as if they were at home and failed to find food. But if they were
transplanted into sites with resident grunts, they followed the
locals to their traditional foraging ground. Commercial fishing
that targets older fish may deplete such cultural knowledge, and
could explain recent shifts in fish movements like the location of
cod spawning grounds.
Social intelligence: The complexity of some fish interactions –
including cooperation, deception and reconciliation – implies
a sophisticated social intelligence in some species. One
fascinating example of cooperation is team hunting by groupers
and giant moray eels, reported by Redouan Bshary. It is initiated
by a grouper rapidly shaking its head at an eel at rest in its rock
cavity. Although normally a night hunter, the eel more often
than not follows the grouper and they hunt together, the moray
fossicking among the coral and the grouper cruising over the
surface. The combination is so efficient that groupers catch five
times as many prey than when foraging alone. Fish also tend
to cooperate with each other in more dangerous deeds. When
approached by a predator, sticklebacks and guppies usually
form pairs to inspect the predator and obtain information about
its identity and intentions, advancing towards the predator in
turns. If one partner chickens out and hangs back, the other fish
remembers that and refuses to cooperate in future inspections.
This shows that fish punish others when they break social rules.
Tool use: Not so long ago, tool use was one in a long list of
skills thought to be unique to primates. In 2006, while diving
on the Great Barrier Reef, Scott Gardner heard a loud crack. He
found a black-spotted tuskfish with a cockle shell in its mouth
that it whacked against a rock from alternate sides until the
shell broke, releasing its contents. Gardner’s photos were among
the first documented evidence for tool use in fish. The broken
shells scattered around the rock suggested a habit, and similar
middens can be found all over the Great Barrier Reef. It turns
out that tool use in the wrasse family is common – a number of
species crush sea urchins against corals and rocks to break off
spines and access soft body parts. Cichlids and catfish often glue
their eggs to leaves and small rocks and carry them around when
their nest is threatened. The archerfish’s remarkable ability to
hunt insects on land by squirting water at them – taking account
of the difference in light refraction between air and water – is
also akin to tool use. They have to learn to do this and benefit
from watching others.
Sentience and pain perception
Despite fish displaying all sorts of intelligent behaviours,
there is still controversy about the question crucial to animal
welfare – do fish feel pain? Although definitions are constantly
debated by scientists and philosophers, it is generally accepted
that ‘sentient’ animals experience pleasure and pain, while
‘consciousness’ includes being intelligent, sentient and self-
aware. But, unlike intelligence, they are virtually impossible
to measure. We cannot know what is going on inside another
person’s mind let alone another animal. Instead, we rely on
various physiological responses and behaviours that are
assumed to imply sentience and consciousness – such as
changes in stress hormones with painful events and recognition
of self in a mirror.
There is no doubt that fish have all the hardware for pain
perception. Their pain receptors resemble ours, which is not
surprising since we inherited them from a fish-like ancestor.
Social traditions
are thought to be
responsible for the
migration routes of
many species.
French grunts (Haemulon flavolineatum) rely on cultural knowledge of
pathways to their feeding sites. Photo: Rob Atherton (bbmexplorer.com)
THEM & US
wildlife-australia.org | Wildlife Australia | 13

But do fish respond to pain in an emotional, cognitive sense?
When we experience pain, there is both a reflex response to stop
or avoid the painful stimulus and a conscious component of
feeling and remembering, which motivates us to avoid painful
stimuli in future. Some biologists say the physical and emotional
experiences are separate, and that fish lack the latter. But if the
point of pain is to reduce the chances of injury, it is more likely
that pain detection and emotional and cognitive responses have
evolved as an integrated system. Fish have brain structures with
functions similar to other vertebrates, including areas thought to
have emotional and cognitive functions, so we strongly suspect
they respond emotionally to pain.
Several studies imply that fish feel pain. For example, rainbow
trout injected in the lips with bee venom or acid rub their lips on
the tank, breathe heavily, and rock from side to side. They lose
their appetite and their attention is impaired. They fail to take
refuge from predators and dominant fish become less aggressive.
The administration of analgesics to fish reduces the symptoms
associated with pain, just as it does in mammals. Some show
pain-based learning, associating certain objects or smells with
harm and trying to avoid them. Their stress levels increase when
anticipating painful events, and they are willing to pay a cost for
pain relief. Lynne Sneddon found that zebrafish in pain choose
to spend time in a barren chamber with analgesics rather than in
the enriched chamber they usually prefer.
Just last year Sonia Rey and others found that fish also show
‘emotional fever’, an increased body temperature in response
to stressful situations, that is widely regarded as indicative of
sentience. The temperature of zebrafish rose by 2–4 degrees
Celsius when they suffered confinement.
Mind the ethical gap
It is time to extend our empathy beyond the water’s edge. Fish
ought to be treated in the same way as other vertebrates, for their
cognitive capacities are comparable – they have good memories,
recognise other individuals, live in complex communities where
they cooperate with and learn from each other, develop cultural
traditions, and have abilities such as tool use. It is also clear they
feel pain and have the capacity to suffer. These are all good reasons
to include fish in our ‘moral circle’ alongside livestock, companion
animals and wildlife such as dolphins.
Public concern for animal welfare has brought some changes
to the treatment of livestock and pressure for much more. We
hope that people will also start demanding humane treatment
for fish. Achieving this is a daunting challenge, for it will require
many changes in public perception, government agencies,
fishing industries, and recreational angling. It may not be too
hard to convince aquaculture industries to reform – for they
can probably sell humanely produced fish for a premium price
(like free range eggs), but it is difficult to conceive of change in
industrialised commercial fishing practices. We could return to
small scale, sustainable operations similar to artisanal fishing,
but this will reduce catches and increase prices (which would
also better reflect the true environmental cost of taking fish from
the sea).
Now that we understand much better the impressive cognitive
abilities of fish and their capacity for suffering, the first step is to
change public opinion. Perhaps it’s time to trade fishing rods for
masks and snorkels.
READING: Brown C. 2015. Fish intelligence, sentience and ethics.
Animal Cognition
18 (1):1–17 Brown C, Laland K, Krause J. (Eds) 2011.
Fish Cognition and Behaviour
.
Wiley-Blackwell Brown C. 2012. Tool use in fishes.
Fish and Fisheries
13(1):105–15
Brown C, Warburton K. 1999. Social mechanisms enhance escape responses in shoals
of rainbowfish,
Melanotaenia duboulayi
.
Environmental Biology of Fishes
56(4):455–59
Brown C. 2016. Fish pain: an inconvenient truth.
Animal Sentience
2016.058
Bshary R, Gingins S, Vail AL. 2014. Social cognition in fish.
Trends in Cognitive Science
18(9):465–71 Bshary R, Brown C. 2014. Fish cognition.
Current Biology
23:R947–50.
DR CULUM BROWN is an Associate Professor at Macquarie University
with a fascination for Australian native fish. For years he has studied
the behavioural ecology of fishes with a special interest in learning
and memory. Recently, Culum has been involved in the debate about
awareness and pain perception in fish, and has been trying to change
public perception of fish intelligence and increase empathy towards them.
CATARINA VILA POUCA is a PhD student in Culum’s lab. She is interested
in the cognition and behaviour of sharks and rays. She has worked
with blue sharks in Portugal and is now studying the spatial and social
behaviour of Australia’s Port Jackson shark.
No one could fail
to be impressed by
the Machiavellian
social intelligence
of cleaner wrasses.
Their brains may seem small but bluestreak cleaner wrasses (Labroides dimidiatus) display an impressive
Machiavellian intelligence. They cheat on their clients when it’s safe to do so, try to make up with clients they
offend, and give priority service to those that won’t hang around. These cleaner wrasse are cleaning a snapper in
Madagascan waters. Photo: Philippe Pourtallier
It has been proposed that
primate intelligence was driven
in large part by our sociability.
Fish are among the most sociable
of animal groups and examples
have emerged of cross-species
cooperation. Here a green moray
eel (Gymnothorax prasinus)
and greyface moray eel (G.
thyrsoideus) share a lair near
Green Island, South West Rocks,
NSW. Photo: Richard Ling
14 | Wildlife Australia | AUTUMN 2016

Once thought to be unique to humans, tool use is regarded as a sure sign
of intelligence. Diver Scott Gardner photographed this black spot tuskfish
(Choerodon schoenleinii) on the Great Barrier Reef using a rock as an anvil
to open a cockleshell. Photo: Scott Gardner
A revelation of recent years has been that giant moray eels (Gymnothorax javanicus) and groupers (Plectropomus
pessuliferus) hunt cooperatively, with the grouper summonsing the eel with a specific head signal. There are likely to be
many more such examples of inter-species cooperation. Here, a moray eel defends a morsel of food from a young potato
cod. The eel then allowed another cod – known by local divers as its ‘buddy’ – to eat the food. We are a long way from
being able to interpret such behaviours. Photo: Robin Jeffries
THEM & US
wildlife-australia.org | Wildlife Australia | 15