Australasian Science: Australia's authority on science since 1938

Fishy theories about pain in fish

By Geoff Russell

There's more to sensory perception than the complexity of an animal's brain circuitry.

Brian Key makes it sound so simple (Australasian Science April 2016, People feel pain in a particular part of our brain which fish simply don't have. End of story. But this only creates a bigger mystery. We've known where people feel pain and the essentials of both fish and human anatomy for decades. So why, I wonder, is there a debate? Key acknowledges the debate at the beginning of the article but doesn't explain it. Are those who think fish feel pain simply ignorant of basic piscean anatomy and the pain centres in people? No and no.

Clues to this mysterious debate are contained within Key's article itself. First, his computer analogy. It's simply wrong. A low end computer can do any computation that a supercomputer can do; it just takes longer. The complexity of the circuits is irrelevant. Alan Turing proved this result back in the 1930s; long before anybody made a computer of any kind! It's not quite as film worthy as breaking German U-boat codes, but it's a result with deep practical significance. The complicated instructions used by programmers in their various programming languages are translated by other programs (called compilers) into simpler instructions and when those hit the CPU (central processing unit) they are further translated into a stream of even simpler instructions. The bottom line here is that you can buy many different computers which appear to work the same, but which have very different machine instructions and circuits underneath.

So if computers are relevant at all, it's to show the opposite of what Key thinks; namely that it is at least plausible for fish to use different brain circuitry for pain than we do.

This plausibility is enhanced by Key's note about the difference between what happens in a fish with most of its fore-brain removed and people with fore-brain damage. Given that fully functional fore-brains are so critical for complex human behaviour, how do fish manage? The obvious answer is that fish use different methods and structures to achieve similar results.

Simply citing brain hardware differences may be enough to separate us from carrots and bricks, but fish certainly have a very sophisticated brain and exhibit all the behaviours that we quite reasonably associate with conscious intentional activity. The scientific study of fish, other than their anatomy and taste, is relatively new. A 2015 review ( reveals a behavioural richness that will surprise many:

... [fish] have excellent long-term memories, develop complex traditions, show signs of Machiavellian intelligence, cooperate with and recognise one another and are even capable of tool use ... Emerging evidence also suggests that, despite appearances, the fish brain is also more similar to our own than we previously thought. There is every reason to believe that they might also be conscious and thus capable of suffering.

Complex brains and intelligence have evolved multiple times independently in the animal kingdom ( with some extraordinary differences in form. Why would pain behaviour have evolved multiple times but only be accompanied by pain perception in us and close relatives like mice and monkeys as Key admits? There are serious suggestions that zebra fish be used in human stress research ( Subject these fish to unpredictable unpleasant events and they do what we do, have impaired cognitive function, increased anxiety behaviour and exhibit similar physiological stress responses. Why would such an elaborate analogue to human stress evolve without perception? And why evolve perception to unpleasantness but not to extreme unpleasantness?

It's precisely the perception of pain and unpleasantness that gives pain its protective effects. We avoid physical damage because it hurts. Mutations in the SCN9A gene ( will render people insensitive to pain and without perception, pain is no longer aversive. But we know that fish avoid pain (; presumably for exactly the same reason as we do; it hurts.

Attributing machine-like qualities to animals is called mechanomorphism ( In the seventeenth century, French philosopher Rene Descartes sliced and diced dogs claiming that they behaved as if in pain, rather than being actually in pain. Hundreds of years later, some people still say the same thing with fish. They merely behave as if conscious, as if remembering, as if recognising each other, as if painkillers work. At some point it becomes sheer obstinacy to refuse to recognise consciousness in these animals, and if they are conscious and have pain receptors and behaviours similar to our own, then denying pain perception because of hardware differences looks like a decision based in self interest; governed perhaps more by taste buds than objectivity.

Geoff Russell is the author of CSIRO Perfidy ( He has qualifications in mathematics, philosophy and years of experience scrutinising research protocols on the Animal Experimentation Ethics Committees at Flinders Medical Centre and the Department of Primary Industries in South Australia.