IMAGING THE PAST

Andrew Curry

What stone toolmaking and neuroscience can tell us about what it means to be human.

Our ancestors stuck with Acheuleanesque hand axes longer than any other tool type. The distinctive axes were used for chopping and cutting all across Africa, Asia, and Europe for nearly 1.6 million years—from their appearance around 1.7 million years ago to 100,000 b.c. The tools support a rich tale. “I would argue that this is the most important period in human evolution,” says Putt. “Anatomy, behavior, cognition—all point to a more human type of being.”

The acheulean transition has sparked a long-running debate: What role, if any, did toolmaking play in our evolution? This is where it gets even more interesting. Early hominids must have been capable of making tools before they began making them. “You can’t have the behavior occurring before you have the capacity to produce that behavior,” says Emory University evolutionary anthropologist Dietrich Stout.

But once toolmaking began, it may have been driven by other abilities, coevolving with language, social interaction, advance planning, and other behaviors. Hominids who were able to quickly learn to make useful tools and then teach their offspring the skill would have had a better chance of passing their genes on. Natural selection, the theory goes, would favor the toolmakers who were able to both learn and communicate. “The idea is that stone tools and language coevolved in humans,” says Natalie Uomini, an archaeologist at the Max Planck Institute for the Science of Human History. “It’s one of the biggest open questions in human evolution right now.”

Over the past two decades, archaeologists have used neuroscience research to try to move beyond speculation. The field has progressed in fits and starts since the 1980s, when machines were first used to look at human brain activity in real time. Anthropologists suggested the technology’s potential as a research tool as early as 1990, but it took another decade for Emory’s Stout to publish the results of the first stone tool experiments that used a brain scanner.

Part of the delay had to do with the limitations of the technology. The most powerful and common brain scanning devices are functional MRI (fMRI) machines. MRI scanners use massive magnets that require people to lie absolutely still during scans, making direct observation of toolmaking challenging. Researchers have come up with some creative workarounds. In one early attempt, Stout had study participants make flint tools for 45 minutes. He then put them in a PET scanner, which measures the accumulation of glucose, the body’s basic fuel, in different parts of the brain. “The thought was that glucose would accumulate in the neurons that were most active,” Stout says.

Experimenting with modern people to draw conclusions about the deep past poses other problems. Stone toolmaking isn’t a common skill in modern times, for example, but Stout wanted to capture the brain activity associated with making the tool, not learning how to make it. So before they could be scanned, people needed to learn to knap flint—easier said than done. We may sneer at the supposed simplicity of “Stone Age” technology, but volunteers in one of Stout’s experiments needed an average of 167 hours of practice to produce passable Acheulean hand axes.

Today, a pile of broken rock 10 feet across and five inches deep—more than 3,000 pounds of chipped flint—outside Stout’s office at Emory testifies to the thousands of hours volunteers have put into learning to make tools. “I can explain golf in a few minutes, but getting your swing down can take years,” Stout says. “Toolmaking is difficult to do, comparable to a sports skill or playing a musical instrument.”

In other experiments, Stout trained participants to knap flint and then scanned them in an MRI machine while they watched videos of someone else making tools. Using such workarounds, Stout, Uomini, and others pinpointed areas in the brain that were particularly active during Acheulean toolmaking: the right inferior frontal gyrus, for example, a region that experiments have shown helps with impulse control and juggling multiple tasks.

Stout also showed that toolmaking is capable of rewiring the brain. In another experiment, brain scans showed that the more a person practiced knapping flint, the more brain matter built up in the regions responsible. And in a recently published study, Stout worked with primate specialists to show that the brain circuitry connecting the right inferior frontal gyrus to the rest of the brain was much more developed in humans than in chimps.

Altogether, Stout argues, there is powerful evidence that toolmaking helped shape the modern human brain. What Stout’s experiments haven’t shown yet is whether language was also a required part of the package. “It’s very controversial whether early hominin species had language,” says Uomini. “For me, the big question is, How do language and stone toolmaking draw on similar brain areas? Are they overlapping brain areas, or not?”

If they are, it would be reasonable to guess that language goes a long way back. Is it possibly a critical clue to our evolution as a species and what differentiates us from other animals, including our close primate relatives? John Gowlett, an anthropologist at the University of Liverpool, argues that the ability to plan chains of actions—such as those that go into forming sentences or making Acheulean tools—is an essentially human trait. “Animals like us are unleashed by language. We have past, present, and future,” Gowlett says. “Other animals are tied to the present.”

For the last eight years, Putt has been working to test the idea that tool use and language involve the same brain areas. Early on, she turned to a lab on the University of Iowa campus dedicated to using a new brain scan technology called functional near-infrared spectroscopy (fNIRS). Unlike the more familiar fMRI, which requires people to lie motionless in a huge apparatus, fNIRS is portable and allows subjects to move and work while wearing the scanner, whose sensors are nestled against the scalp using a special snug cap. The technique works by shining infrared light through the scalp and skull into the brain—certain wavelengths pass easily through skin and bone—and measuring how much light is absorbed.

The technique works because blood that is carrying oxygen soaks up light at a different rate than blood that has already delivered its payload of oxygen to active brain cells. Brain regions that are working stand out from regions at rest. “If you’re using one part of your brain more actively, it’s going to be more oxygenated,” Putt says. “We can measure in milliseconds what’s happening and pinpoint, with great accuracy, where it’s happening.”

Over the course of several years, Putt trained two different groups of volunteers to make Acheulean-style stone hand axes. Both groups watched a video of Alex Woods shot to show only his hands at work. The first group watched him while listening to his instructions; the second watched the same video, but with the sound turned off. At first, Woods was deeply skeptical. “Sit down with a bunch of students, with technology they hadn’t even seen, and try to get them to make it with no verbal instruction whatsoever?” he recalls thinking. “There’s no way.”

To his surprise, both groups managed to craft rudimentary hand axes. Once they had mastered the basics, Putt had them make the tools while wearing fNIRS sensors. Putt was looking for differences between the two groups. Her theory was that if language centers in the brain lit up regardless of whether participants learned with the sound on or off, it would be a strong clue that tool use and language were somehow related.

Instead, scans of dozens of participants showed that the brain activity of those who learned without hearing the instructions— without language, in other words—looked different. In the toolmakers who learned just by watching Woods, the brain’s language centers were quiet; in the scans of people who learned while listening, language centers lit up. What both did have in common was activity in areas that connect with working memory and the areas that process sounds and images. The results were counterintuitive. Learning to use and make Acheulean tools was not closely linked to language centers, as evolutionary anthropologists had assumed. “We’ve showed you can make a hand ax without language,” Putt says.

Rather, Putt’s scans suggest that in the absence of verbal instruction and language, the brain relies on a combination of working memory and motor control to make tools. Research has shown it’s the same network of neurons we use to make music. “Piano playing uses almost the identical network as toolmaking,” Putt says. “You’re coordinating your hands while keeping in mind all of these sub-goals.”

It’s an exciting result. Experiments on modern humans may never conclusively prove how our distant ancestors thought. But the insights archaeologists and neuroscientists are generating together are contributing to our understanding of what makes us human. And with neuroimaging techniques developing fast, their potential to answer questions about our deep past has barely been tapped. “The big picture is to look at other points in time. Were Neanderthals as smart as humans? We can test that, by comparing their tools to the Upper Paleolithic tools humans were making,” Putt says. “We’re focusing on one small snapshot, but humans have been evolving for millions of years.”

By Andrew Curry in "Archeology", USA, volume 71, n. 2, March/April 2018, excerpts pp. 43-45. Digitized, adapted and illustrated to be posted by Leopoldo Costa.