Jun 17 2024

Are Animals Conscious?

This is a great scientific question because it challenges how we ask and answer scientific questions. Are animals conscious? This is a question discussed in a recent BBC article that piqued my interest. They eventually get to a question that they should have opened with – how do we specifically define “consciousness”? We can’t answer questions about an alleged phenomenon unless we know what it is. Ideally we would have an operational definition, a list of inclusion and exclusion criteria that need to be met to fit the definition.

So I am going to start with this question – how do we define consciousness? I think there are at least two different contexts here. In medicine we use the term to refer to different states in people. We know, from our own experience, that humans are conscious, and from one point of view we define consciousness as what humans experience. We assume other fully functioning humans are conscious because we are, and there is no reason to think that other beings with brains similar to our own have a fundamentally different phenomenon driving their behavior. In fact part of consciousness is a theory of mind, which is the ability to think about what other beings think and feel.

So when we talk about consciousness in humans the question revolves around the health and functioning of the brain. Someone might be unconscious, or comatose, or vegetative. We label these as “disorders of consciousness”. We might also discuss consciousness in the context of healthy altered states, such as sleeping. Here we do have very specific technical definitions, based upon neurological examination. However, even here our definition is being challenged by new technology, such as functional MRI scanning, which may shows signs of subtle consciousness in someone who does not show signs on exam.

An entirely separate question is whether any non-human entity is or can be conscious. This includes machines (general AI) and animals. The challenge here is that we cannot base any conclusion on extrapolation from ourselves. We cannot experience what another entity is thinking or feeling. We can only observe their behavior. This has led some scientists to take an approach called behaviorism, which only seeks to understand and model behavior, and not to examine or speculate about internal states. In humans behaviorism has largely given way to cognitive psychology, which explicitly deals with internal cognitive states.

So – can we have a cognitive psychology of animals or are we limited to behaviorism? There is one difference with humans – language. Humans can tell you how they feel and what they think, because humans have language. It is controversial whether or not animals can learn sign language and communicate like humans, so let me put that question aside for this essay. Without language we only have behavior. The question therefore is – can we infer from behavior alone the internal cognitive state of a non-human?

There are two types of error we can make here. One is to anthropomorphize, to assume that an animal behavior which superficially seems analogous to a human behavior results from a similar underlying mental state. We tend to project human feelings and experiences onto other agents operating in the world. In fact, we are wired to so do. The other type of error is to dismiss non-human consciousness solely because of the absence of language, or because that consciousness is different than typical human consciousness.

This, in my opinion, leads to an inherent dilemma in asking the question about consciousness. The answer will be yes or no depending on how narrowly or broadly we define consciousness. The question, in a way, becomes meaningless (except at the extremes, rocks are not conscious while humans are the gold standard of consciousness). For everything between rocks and humans, starting with viruses, the best approach might be to consider consciousness as a continuum. Actually, I don’t think viruses are on this continuum, but bacteria might be (with a broad enough definition). By the time you get to multicellular creatures, we are seeing complex behavior that we could anthropomorphize and call consciousness, or broaden the definition enough to include their behavior.

Here is one example from the BBC article – bees have been observed to roll little balls for no apparent useful purpose. Some researchers believe the bees do this because they enjoy it, and therefore it is the equivalent of playing. I have a problem with this conclusion. I think it is a great example of the first type of error – anthropomorphizing. We can’t really know what the bees are experiencing, and there could be many reasons why their primitive behavioral algorithms encoded in their networks would include such behavior. It could easily be an incidental behavior, something the system does while idling. Or perhaps it reduces the probability that the bees will engage in other behavior that might be counterproductive. Or it is a tradeoff, a behavior that emerges when the algorithm is programmed for some other useful behavior. But it’s a real stretch to say the bees are “enjoying” the behavior and therefore they are conscious in a way that is closer to humans than we thought. I don’t buy it. Or at least, I don’t think that conclusion serves any scientific purpose.

Are bees conscious? Sure, if you define consciousness to include whatever bees experience. Does that help explain their behavior? I’m not convinced.

Because consciousness is a continuum, and we are all connected evolutionarily, the closer you get to humans phylogenetically the closer you get to human consciousness. Anyone who has owned a dog I think would have no problem believing that dogs are conscious to a high degree (although not human level). They communicate with us, they can read our behavior and have been shown to have an impressive vocabulary. They appear to have emotional states. They have dog consciousness. Even here, though, it is tempting to anthropomorphize, to go beyond doggie consciousness and assume human-level motivation behind their behavior. Again, it’s what we do.

When you get to primates then I think we have to be especially careful. Now we are so close to humans evolutionarily that we are also getting close to human-like consciousness. But I do think that human evolution does include a unique feature – the evolution of sophisticated language ability, that fundamentally altered the way we think. We have a lot of brain power that chimps, for example, simply don’t have.

Also, evolutionary branches that diverged off the one that ultimately lead to humans evolved different brains with different types of consciousness. Birds have bird-consciousness, which is different than mammalian consciousness. Cephalopods, like octopuses, have cephalopod consciousness, which is different than vertebrate consciousness. We should try to understand them on their own terms, and not dismiss consciousness because it is non-human, nor try to shoehorn their apparent consciousness into human analogues (like play).

It’s very challenging research, and it makes me think the behaviorists have a point. We can model behavior, but it is very tricky to infer internal states from that behavior. But I am open to research that tries to do just that, as long as they can make falsifiable hypothesis. We may never be able to know for sure, but we can at least come up with some useful testable hypothesis and some indirect inference. But again – we have to proceed very carefully.

No responses yet