Nineteenth-century American clergyman and abolitionist Henry Ward Beecher once wrote, “A tool is but the extension of a man’s hand, and a machine is but a complex tool.” These words presaged, by more than a century, a line of scientific research into “embodiment”: how humans’ wealth of sensory inputs—including the touch and visual perception involved in manipulating a tool—modify the sense of one’s physical self. Embodiment implies that when one holds a screwdriver, for example, the brain morphs its representation of a “hand” until that representation reaches all the way to the very tip of the tool.

But is this what really happens? Why should the brain somehow give up, even temporarily, its conception of a dexterous hand for that of a blunt instrument? Such questions inspired Tamar Makin, a neuroscientist at University College London. In research recently published in PLOS Biology and the Journal of Neuroscience, she examined whether people with prostheses—or London street cleaners wielding litter-grabbing metal poles—do, in fact, merge their identities with such tools. So far the results of Makin’s studies contradict Beecher’s intuitions about hand-tool embodiment, as well as some modern research that had suggested using an implement alters internalized body maps.

Makin’s work has focused on developing a better understanding of how the brain can best accommodate artificial limbs that operate in the most efficient manner possible—hence her interest in determining whether embodiment is a real thing. And she would like to eventually go beyond replacing body parts and try enhancing still intact ones. Her laboratory has been collaborating with the prosthesis designer Dani Clode to study how people learn to adapt to a “third thumb”—a robotic finger that users strap to their hand and control remotely with their big toes. A paper published this week in Science Robotics shows that people can learn to use the thumb to augment their hand function—but it also raises questions about whether such “plastic” changes in the brain could alter, for better or worse, a person’s neural representation of their biological hand. Makin and her graduate student Hunter Schone, who was co-lead author of the London litter pickers study, talked to Scientific American about what they have been doing recently—even while the pandemic gripped the world.

[An edited transcript of the interview follows.]

How did you get started with this work?

HUNTER SCHONE: We had a Ph.D. student in this lab named Roni Maimon-Mor. She was interested in this question of whether prosthesis users represent their prosthetic limbs as if they were a body part or more like a tool. So Tamar and Roni ran this study. They found really interesting results that basically showed that the more a user used a prosthetic limb, the more it was not represented in the brain like a hand or a tool.

TAMAR MAKIN: This was a perfect test case because someone might use a prosthesis for 14 to 16 hours a day. A prosthesis is a tool that is designed to substitute hand function. There's a very dominant perspective—and it stretches from philosophy to engineering to psychology and also to popular culture—that if we are using tools as experts, then the best way for our brain to represent them is as if they were a part of our body.

Can you describe why the prosthesis finding was surprising?

SCHONE: It all comes from this idea in the engineering community about embodiment. When [people] use a device (and, specifically for an amputee, where they’re using a prosthetic limb), do they subjectively feel like it’s a part of their body? Maybe that means, from a neural standpoint, that the brain begins to actually represent the device like a body part. The striking answer that we get across this work is that this idea of embodiment doesn’t really fit when you are considering what’s happening in the brain—at least in the parts of the brain that we were studying.

What led to the research with the litter pickers?

MAKIN: So the result of the work with the prostheses was very, very exciting. Hunter was keen to come up with a replication of this finding. [With the litter pickers], we thought [that the brain is smart and flexible enough, plastic enough, to come up with new solutions to represent expert tools. They can pick up really different shapes and weights—cups with fluid, cigarette butts. It’s actually quite impressive how good they are.  

SCHONE: Like a prosthesis that sort of substitutes for hand function, we tried to find a tool that’s an extension of the hand function—a different kind of tool—and we tried to find only experts and see if you can show, in this group, the same results as [those] for the prosthesis. We talked about recruiting dentists and surgeons and the kinds of tools they use. Roni thought litter-picker workers might be easier to find than trying to find a bunch of surgeons. You see them everywhere in London.

Man holding a litter picker tool at a park.
Litter-picker tool. Credit: Hunter Schone

What did you do in the experiments?

SCHONE: We were focused on the visual cortex. The interesting thing about it is that in visual cortex, you have a visual area of the brain that represents the hands. Nearby, there’s an overlapping area where tools are also represented. Here is sort of this perfect region to test this question, where you have hands and tools represented. We wanted to know what’s going on in this space when you become an expert with a tool.

We wanted to sort of simulate the litter-picker experience in the best possible way, because they can’t use the litter-picker tool in the scanner. So we made these videos: The videos showed a hand grabbing—or a litter-picker tool grabbing—an object or someone using a different type of grabbing tool such as a pair of tongs. We made 48 unique videos.

And what did you find?

SCHONE: We put everyone in the scanner and showed them the videos. And we looked at the activity in this area of the brain that is an overlapping visual area for hands and tools. And we looked at how each of these things was represented. So we used an analysis that compares differences in activity patterns—which means you can see things that in the brain that are either represented similarly or differently.

We did an analysis of each of these different categories (hand, litter-picker tool, tongs). We compared how similar the activity pattern was when a participant saw the litter-picker tool, compared with seeing the hand—or seeing the litter-picker, compared with seeing the tongs.

The idea of embodiment is that as you use a tool, your brain relates to it more like an actual body part. That would mean that the hand and the litter-picker tool would be more similarly represented. But in fact, we saw the exact opposite effect, and that’s the same results that we saw in our prosthesis study.

MAKIN: This is the first set of studies to really directly test this assumption of how the brain represents a tool by decoding the brain activity of people. This has never been done before. So this is the most serious attempt to identify embodiment—and [to find] very conclusive evidence against it.

But it doesn’t seal the deal against embodiment. Embodiment is a multifaceted phenomenon. If you ask the brain, or at least the visual cortex, of the people who use the prosthesis more, their organs embody the prosthesis less. But if you ask people about their subjective experience of how they perceive the prosthesis after long periods of use, the more they use it, the more they say it feels like part of their body.

The gap between how people experience the prosthesis and how the brain represents it could be explained by the fact that we’re just focusing on one specific part of the brain. For us, it was the perfect brain area. But it’s one brain area of a massive, massive, very complicated network of brain areas. If we’re just looking at one piece of the puzzle, it could be a completely different picture for the rest of the brain. That could be one explanation.

I personally don’t think this is the case. I don’t see the benefit for the brain to mimic or build on the sensorimotor infrastructure that we have for the hand, to control or represent something that is so different from a hand. The way we move every one of our fingers separately is very different from the very rigid, single movement carried out with a litter-picker tool and vastly different from what you can do with your prosthesis.

If the prosthesis and the litter picker were not extensions of the body, how were they represented in the brain?

MAKIN: If you look at people who are experts in picking out birds or cars, they start creating a kind of new expertise in the brain. So they dedicate more brain resources to representing birds and cars. And they’re coming up with more refined representations of this new thing than normal—something your brain or my brain wouldn’t be able to pick up. In the same way, you can become an expert in using a prosthesis. You understand the various features and how they’re important. You start coming up with this expert representation of it. But it’s not a representation of your own body, it’s a representation of an object. That’s what we found through our [functional magnetic resonance imaging] studies of prosthesis users and litter pickers.

Can you say more about where you are going with your work?

MAKIN: I’m interested in how we can help engineers design the best prosthesis, because at the moment, we have a crisis: engineers are designing more and more complicated prostheses, which become more and more expensive, more and more fancy.

And I think a potential problem is that designers are very much enchanted with the notion of embodiment. They try to design prostheses that work similarly to the way the body works. They assume that the closer the prosthesis is to the hand, the easier it will be for the brain to replace the hand with a prosthesis. Our research says, “Don’t bother. Your brain is plastic. Your brain is going to come up with the best solution to represent a prosthesis. The brain knows that it is not a hand.” This is exciting because it invites engineers to think outside the classical box of how to design prostheses. They can think about new engineering approaches. Who said the prosthetic limb needs to look like a hand and not like a tentacle from an octopus?

What else are you doing in this area?

Credit: Dani Clode Design and the Plasticity Lab, University College London

MAKIN: This also opens up very exciting opportunities for motor augmentation. We’re working with a robotic finger to give able-bodied people with five fingers an extra body part they’ve never had before. The brain doesn’t have more resources to support this new body part because we’re not genetically designed to have six fingers. But if the brain is plastic enough to come up with a new solution for prostheses, it should also be plastic enough to allow us to take advantage of these new technologies—such as a third thumb. We are very lucky to work with Dani Clode, who designed this augmentation device to extend motor capabilities of people. And we’re doing lots of studies with this device in order to understand what happens to the brain of the user when they use an extra thumb.

Trained users of the thumb demonstrated great dexterity. But is more research needed to determine what happens to motor coordination in the natural hand if people use artificial thumbs all the time?

MAKIN: Using the thumb changed the way people coordinated their movements, and we believe this triggered brain plasticity in the way the hand is represented. This means that in order to take advantage of the thumb—extending the motor repertoire—something has to change in the way we use and represent our bodies. This is definitely an important consideration that hasn’t really been raised before, with huge safety considerations. If we ask a factory worker to use an extra body part in the factory, does this mean they will be more clumsy when driving home after removing the device? Much more research is needed to gain a better understanding of the balance between extra body part use, brain plasticity, and motor control of both the body and the device.