How does this bird know how to sew?

Vox reader Stefanos Nasiopoulos asks: How do instincts work in animals? Do they use the same mechanism as memories? How are they different from learned behavior? For example, when the tailorbird is actually sewing leaves to form a nest, does it understand what it’s doing? Or does it just feel a compulsion to do it without knowing why?

In the forests of Asia lives a small bird with a very apt name — the common tailorbird. These animals, which are covered in a coat of green, gray, and chestnut feathers, are known for their ability to sew leaves together to create a nest.

The birds poke holes in leaves and then, using their beak as a needle, stitch them together with bits of spider web, plant fibers, or other string-like materials. It’s pretty incredible to watch.

A good question is, how the hell does a bird know how to sew? Was it taught by its elders? Or is it instinct?

The same question applies to a number of other animal behaviors: Beavers building dams, squirrels burying nuts, cats always falling on their feet.

Earlier this year, a Vox reader reached out looking for answers about how animal instincts work. I was curious too. Partly because I need to know what makes my dog (a shepherd-pit-bull mix) so fixated on squirrels.

So I called up Mark Blumberg, a neuroscientist at the University of Iowa who’s written books about animal instinct — a term he says is confusing and thoroughly unhelpful.

What follows is our conversation, which has been edited for length and clarity. And if you want more, listen to the episode of our science podcast, Unexplainable, called “Basic instinct,” in which Blumberg lays out his argument against the simple idea of instinct.

When I think of instinct, I think of my dog lunging at a squirrel — a seemingly thoughtless behavior. Is there a scientific definition of the word?

There are many, and that’s one of the problems — people use different definitions. So you could talk about it as something that is inborn, unlearned, genetically programmed, is typical of the species, emerges without prior experience, and so on.

I don’t use the word instinct. Because once you have that many definitions, the word no longer has a useful meaning. When people say, “It’s in our DNA,” that’s just a way of saying that it’s habit — that we do it reflexively or habitually or unthinkingly or unconsciously. You can use all these words, but none of them get you closer to understanding what’s actually going on.

I use the phrase “species-typical.” It’s a behavior that is typical to a species. It doesn’t presume anything about where it came from or what’s controlling it or whether learning is involved or whether genes are involved.

Let’s talk about some specific behaviors: The tailorbird makes a nest by stitching leaves together. When this bird, or any bird, is building its nest, does the bird understand what’s going on? Does it know what it’s doing?

I don’t know what it knows, and I’m not sure anybody can know what it knows and why it’s doing what it’s doing. There are some amazing behaviors out there — and the tailorbird’s may be one of them — where animals are doing things that are so complicated that you just need to study them to figure out all the dimensions that give rise to these behaviors.

I had a border collie, and it [is known to] have this herding instinct. Herding in a border collie requires a great deal of training. But here’s the thing: If you Google “rabbit herding sheep,” you will see something that’s pretty amazing — which is a rabbit with a herd of sheep. And every time the rabbit moves, the sheep move.

What’s the lesson to take from that? People tend to think of [habitual behaviors, like herding] as rooted in the mind. That’s why people ask questions about what birds are thinking when they’re building a nest. But it’s not just in the mind. We’re putting instinct inside the mind, but we’re forgetting, in this case, about the sheep. The sheep have their own responses. They can be manipulated. When you look at behavior, you have to think about it within a system.

So you have this bird doing this stitching thing. The only way you know that instinct exists is because those materials exist in the world for it to manipulate.

Is that to say that maybe there’s something on a genetic level that makes certain dog breeds predisposed to certain behaviors, but there’s so much more than just DNA that shapes behavior?

First, all dogs show the basic behaviors that border collies show when it comes to herding. Some dogs do certain things, in part, because they have certain bodies that are good for it. The reason why pit bulls are considered dangerous by many people is that they have the muscular machinery to do damage. And if you train them to be damaging, they will be damaging. But I’ve known Doberman pinschers and pit bulls that are as docile as a chihuahua.

We forget that behaviors grow up in bodies under environmental experiences and explicit learning, and all of those things are going to shape them. It’s not a singular phenomenon. There have been studies of personalities and dogs, and they’re finding that almost all the assumptions that people make about dog personalities based on genetics are wrong.

So is there no genetic basis to the habitual behavior we see in animals, beyond how genes shape their bodies?

Let me say this very clearly: I am not saying that genes are not playing roles in these sorts of processes. Of course, genes are playing a role. The fundamental problem is that we want to dichotomize the problem — nature, nurture; learned, unlearned; genes, environment. All of those dichotomies are false. This stuff is complicated. It’s not rocket science. It’s harder.

Back to this bird: If you had to guess, what are the different processes shaping that elaborate behavior?

It’s going to be a smorgasbord. If you think about something like herding instincts or attacking squirrels, a lot of these complex behaviors can appear at birth in the form of these very slight biases. An animal might have predispositions to perceive the world in a particular way, to behave in a particular way with the body that it has. And then over time, it cascades and gets stronger. It is going to be a complex, subtle unfolding of these initial biases in the system that lead animals to do the things that they do.

For example, all animals must develop a strong preference for their own species. You could say there’s an “instinct” to be able to identify members of your own species. But where does that come from? It comes from early interactions with members of your own species. People have done experiments in which they confuse the animals by having other species in the litter. And what happens? Those animals develop an attraction to the other species.

You also might think that animals instinctively avoid poisonous foods. But there are many cases of observational learning, where animals will interact with other animals that may have just eaten poisonous food. They’ll smell that food on their breath. They’ll be connected with other things that make them want to avoid that smell, and so they won’t eat poison, even though they’ve never interacted with the poison before.

And when animals are engaged in these species-typical behaviors, what some people call instincts, do they have an awareness of their actions? Is a beaver making a decision to build a dam or just compulsively doing it?

At some level, it’s impossible to answer that question. But think about your own behavior and habits, like brushing your teeth in the morning. Are you thinking about what you’re doing with every stroke of that brush? Or when you’re combing your hair or pressing the break while driving a car. Are you aware of what you’re doing? And I think the answer for many of those things is no.

I don’t think that makes the behavior any less thoughtful in the sense that there’s a lot going on in the body that you’re just simply not keeping track of. These are called procedural memories — like riding a bike — which are things for which we have trouble using language to describe. Now, humans have language. So we can divide the world of memory into things we can talk about and those that we can’t talk about. But [most] animals don’t have language, so they’re in the world of procedural memory all of the time.

What are your thoughts that don’t have words? That might be what it’s like to be a bird building a nest. I have no idea, but that’s about as close as I can get.

Does it ultimately matter if we know where certain behaviors come from? Should we care?

First, we’re curious creatures. We want to understand ourselves, and we want to understand the world around us. We’re willing to go out and take pictures of stars and galaxies and send multibillion-dollar telescopes into space. What is that doing for us? Well, it’s damn cool and very important to understand our place in the universe. It’s important that we understand our place as animals. We are animals. We need to understand our place in the animal kingdom. We need to understand why we do what we do.

And from a strictly narcissistic perspective, it is in our advantage for us to be able to understand behavior. Everything is behavioral. Just like we have a diversity of species, there’s a diversity of behavior. If we get it wrong, we end up making assumptions about where we should be focusing our attention in humans to address behavioral problems or to improve behavioral outcomes or teaching outcomes.

And I’m guessing that understanding where behaviors come from can also help keep some animals alive, right? Understanding how certain animals navigate, for example, might help us pinpoint how humans are interfering with that ability.

Absolutely. Climate change is going to have huge effects on an animal’s ability to survive and adapt. You have a lot of animals, like turtles and alligators, whose actual biological sex is determined by the temperature at which they were incubated. And now you have changes in these temperatures. What is that going to do to the actual ratios of males and females in those species so that they can continue to reproduce? That’s another place where this sort of understanding becomes important.