Of course, there's lots of evidence that fish aren't anything like as stupid as are popularly supposed. So, having recently read the excellent Other Minds (a short book by a naturalist/philosopher about the mental lives of octopus), the first thing I read from my Christmas list this year was What A Fish Knows. Both of these books are top of their game, lively and engaging from start to finish. I give them both 10/10. As they both appear to be hugely successful, are readily available and are short reads, I'll keep this brief - they're well worth reading in their entirety.
Other Minds does an excellent job of dispelling popular notions of how octopus intelligence is often sensationalised, without ever dismissing the fundamental point that octopus are indeed intelligent. For example, they've been known to turn off lights outside their tank by squirting water at them. The octopus probably isn't anywhere near clever enough to know that that will work in advance, but simply demonstrating its standard behaviour of squirting water at anything it doesn't like. All the same time, they do show genuine problem-solving behaviour - using shells for camouflage, waiting until their keepers aren't watching to plan their escape, and opening screw-top jars. They are undoubtedly thinking about what they're doing, not just flailing their many arms around wildly.
Being a philosopher, the author of Other Minds is particularly keen to examine what the intelligence of his subjects means for such notions as free will. For example, cephalopods demonstrate split-brain behaviour. Bizarrely, if a pigeon learns something with one eye and then sees it again with the other, it won't won't respond the same way. Its memories are stored in different sides of the brain. This isn't the case for humans (except in cases of surgery), whereas the octopus is a sort of halfway-house in which some information is shared across the brain, but not all.
This has weird implications for what it means to be a thinking entity. Either we're highly mechanistic constructions after all, or, if we have something of a more spiritual animating force, then it apparently can't access both parts of the octopus brain at once. Goodness only knows what the conscious experience of an octopus is like - and it's even weirder considering that the octopus brain is hugely different to those of vertebrates.
Let's keep this light and move on to What A Fish Knows. It turns out that a fish knows... everything. Or at least, as a group, fish display the same range of diversity of traits as any other animal group. The author does an excellent job of keeping things concise but thorough, breaking down what a fish knows into its emotions, what it senses, thinks, desires, and how it behaves. Like Other Minds, the author covers basic biology as well as the mental aspects. And the physiology of a fish turns out to be quite important in understanding how they think.
Inhabiting a world quite alien to our own, under the same eternity of selection pressures they've evolved the same degree of characteristics as any other animal group. They sense a world we can never inhabit in ways we quite literally can never imagine. Unless we start enhancing our bodies with silicon sensory apparatus, we will never know what electroreception or hydodynamic imaging feel like. Perhaps they're similar to our own senses. or perhaps sensing an electrical field is as different from the other senses as sight is to hearing. We may never know - so much for Mary's room.
Fish don't just have an array of powerful senses at their disposal, they also demonstrate problem-solving behaviour. They compete and co-operate socially, even doing the equivalent of pointing to items of interest to their fellow fishes. They use tools (rocks to open shells) and build elaborate nests to attract mates and raise young.
Fish even co-operate in a sort of democratic way. As well as assessing the environment for themselves, they also respond to what other fish are doing. It turns out to be quite difficult to fool a fish - and, rather unlike human decision-making, this seems to scale well for larger groups. The larger a shoal, the less likely it will follow a robotic fish towards a source of danger : one idiot isn't enough, although if enough fish commit, others will follow.
It's fair to say that these aquatic residents think and feel, even if they don't always express it in mammal-like ways. They show signs of distress and undoubtedly experience pain; any residual reluctance in the biological community to accept this apparently coming from a long-standing historical bias*. Trout, if you want a suitably ridiculous factoid to blurt out during one of those bits in a conversation when everyone goes quiet, fake orgasms. And their memory ? Well in goldfish it lasts at least a year, not three seconds.
* The view that animals don't really think or feel seems to be a particularly strong blind spot for certain scientists. It's a particularly hard bias to overcome, since we can't even prove to each other that we ourselves are sentient.
One thing I did wonder about memory was about spontaneous recall. Once, many years ago, a dog of mine lost her toy in some woods just beyond the field behind my house. The next day she very suddenly became excited and jumped around by the garden gate. Eventually we let her out and she disappeared into the woods like a shot, returning swiftly with her lost toy. I cannot imagine that there was anything going on in her mind that wouldn't resemble the human processes of suddenly remembering something out of the blue, having a distinct mental concept, an inner life on which to act.
None of the memory experiments described in the book seem to relate to this. Rather, they're about habituating fish to certain conditions, e.g. learning that there's food in a red tube but not a black one, stuff like that. Months later they still "remember" this. But is this really memory, or just a conditioned response ? For example it's been claimed that ant colonies possess "memories" that outlast individual ants, but the experiments there seem to be more about changing the ant's behaviour depending on the conditions. That's not necessarily at all the same as having a mental image of a lost toy and its location, something specific which allows agency and intent - you might as well say that if you slash a tyre, the car "remembers" its traumatic experience and demonstrates this by driving badly.
No-one seems to have come up with a test for spontaneous recall in fish, though I'll grant that this is much harder than testing habituation. Though perhaps at least one observation does demonstrate this. Fish that live in rockpools sometimes have to escape to new pools as they dry out. Fish that have swum over the pools at high tide are more successful than those which don't, suggesting they have true mental maps - i.e. true memories, not mere habits to general circumstance. You can't generate a conditioned response merely by looking at something once.
Anyway, it seems very unlikely indeed that all fish behaviour is pure instinct. Some of the memory tests are more specific, and some behaviours require novelty, not just applying learned responses. And fish can learn from watching each other too : archer fish become much more accurate at shooting water jets at prey after watching more experienced fish first. Tigerfish learn the best hunting techniques (leaping out of the water to catch birds) by watching each other. These are so specific behaviours, requiring such careful and precise actions to achieve the desired results, that it seems to be inevitable that they are thinking about what's going on, mentally planning their actions and imagining the perspective of their peers.
The author also points out that size makes no difference to intelligence. Although there doesn't seem to be the equivalent of a fish chimpanzee, i.e. one with a broad range of problem-solving skills, even the smallest fishes can display behaviours ordinarily associated with larger animals. And they don't just apply them in very limited cases, but more generally. Cleaner wrasse, when presented with two coloured plates of food, can learn that it's better to eat first from the colour corresponding to the one from which food is removed soonest. They even solve this problem faster than chimps. And some fish can recognise not just other individual fish, but also humans.
To me, this kind of thinking blurs the line between instinct and genuine intelligence. The fish are clearly doing something more than mechanistic responses. As the author points out, "brain size be damned, if it's critical to a species survival then that species will most likely be good at it." It makes sense for a cleaner wrasse to understand that disappearing food must be eaten quickly, but it takes more than gut instinct to generalise this from visiting sharks to coloured plates. That's true intelligence, in my book.
What of the slower-learning chimps ? That doesn't necessarily make them overall stupider, because chimps (who can solve some puzzles faster than humans) can do things that fish can't. When we see an animal doing something apparently very stupid, like albatross who can't work out if a chick is theirs if it falls out of the nest, we should perhaps be forgiving - their world is some ways very different to ours. We have absolutely no conception of what it's like to soar thousands of miles across the sea under our own power, and I daresay that if we found ourselves with the body of an albatross or a little fishy, we'd be dead within a week. Perhaps the responses of an animal will only ever make sense from the perspective of that animal, their world being just too different for us to ever fully comprehend it.
Or perhaps not. The most interesting thing for me was not the amazing senses or complex behaviour of fishes, but a fundamental similarity they share with humans : fish suffer from optical illusions. That classic one where one line looks longer than the other even though it isn't ? Yep, fish have the same problem. Fish can be trained to select the longest line for a food reward, but when then presented with the illusion, they respond much as humans do - incorrectly. I find this astonishing. These radically different creatures, mostly endothermic, silver-scaled denizens of the deep dark depths, separated by hundreds of millions of years of evolution from me, are at least in some ways thinking in a very similar way. Fish have mental lives and intelligence - they are not thoughtless automatons, nor are they so wholly alien. If we can't yet know the mind of a fish, we do at least know that they have minds.
What a Fish Knows
This is the most delightfully charming work of nonfiction I have read in a long time! I have heard on numerous occasions that fish are nothing more than "water vegetables", lacking feelings of pain, memory, familial love and social structure.
Beautiful article! Thanks for sharing!
ReplyDelete