Six-year-old Johanna Gill puts a protective hand on her sister, Eva.
Read Caption
Martin Schoeller, National Geographic
Six-year-old Johanna Gill puts a protective hand on her sister, Eva.

What You Do Is Who You Are

We’re living in an age of genetic explanations. Consider a few headlines from the past week alone: Single Gene May Extend Lifespan by 25 Percent. Genes Show One Big European Family. Genetic Test Helps Identify Aggressive Prostate Cancer.

I certainly do my share of gene’splaining. But for all the buzz given to the studies that uncover telltale genes, rarely mentioned are the most obvious and familiar examples of how genes aren’t destinyidentical twins. They share the same genome and, usually, the same parents, same neighborhood, and same food. And yet, as anybody who’s ever met a pair knows, they are not the same person. Why?

“Ten years ago the prevailing theory was that there must be systematic differences in their environments,” says Eric Turkheimer, a professor of psychology at the University of Virginia. One twin is favored by her mother, say, or is bullied in school, or catches fewer colds. But studies looking for big, non-shared environmental influences have come up short. As Turkheimer wrote in a fascinating commentary in 2011: “Exactly what the nonshared environment consists of has been a matter of mystery and controversy for some time.”

This is a tough thing to study in people. After all, scientists can’t pluck a pair of newborn identical twins from the arms of their mother, raise them for years in absolutely identical environments, and watch how they diverge.

Scientists can do that with mice, though, and now they have. In today’s issue of Science, neuroscientist Gerd Kempermann and colleagues report that genetically identical mice raised in the same environment show striking differences in their brains and exploratory behaviors. What’s more, these individual differences — individual personalities, you might say — become more and more pronounced over time.

View Images
The “enriched” environment. From Freund et al., Science 2013.

The researchers started with 40 identical, inbred female mice, and implanted radio-frequency identification tags in their necks. They housed the animals for three months in what must be the mac daddy of all rodent cribs: a 6-foot-square, 6-foot-tall, five-level enclosure, outfitted with acrylic glass tubes, nesting boxes, plentiful food and water, and a sprinkling of plastic flower pots, cardboard pieces, wooden scaffolds, and other toys.

This “enriched environment” also had antennas throughout that tracked the precise movements of the mice. The researchers calculated the animals’ roaming entropy, or RE, which is the probability that a mouse is located at any given location at any given time. RE is related to physical activity, but not exactly the same. “On a treadmill your roaming entropy is much lower than on a forest trail,” Kempermann explains. But RE isn’t only about territory coverage, either. Flying from New York to Los Angeles, despite covering a huge territory, would be a low RE because you don’t cover the space in between. So: A high RE means the mouse covers a lot of territory and covers it thoroughly.

View Images
By Jodi Cobb, National Geographic

The scientists also tracked how the animals’ brains changed over the three-month period. They measured neurogenesis, or the birth of new neurons in the hippocampus, a region important for learning and memory. Neurogenesis is a useful measure of brain plasticity for two reasons. First, it’s very responsive to environmental changes. Second, it’s quantifiable. “Neurons, we can count,” Kempermann says.

When he was a postdoc in Rusty Gage’s lab at the Salk Institute, in California, Kempermann reported that adult mice given access to a running wheel made twice as many new neurons than did mice with no wheel. Within the running group, of course, some individuals used the wheel more than others. “We started to wonder whether there might be individual differences in the response, even if the talent, the genetic make-up, was the same,” says Kempermann, who now works at the German Center for Neurodegenerative Diseases in Dresden*.

And that’s exactly what his new study found. The mice showed individual differences in RE right from the get-go, but they became far more different from one another as time went on. The study also found a strong association between RE and neurogenesis: Mice with the most extensive exploratory behavior showed the most neurogenesis.

“What this study objectively and clearly shows is that the ‘nature vs. nurture’ tension goes right down to the level of brain circuits,” says Sam Pleasure, a professor of neurology at the University of California, San Francisco, who was not involved in the work.

What the study doesn’t yet explain, though, is the biological mechanism driving these individual differences. Environmental changes can affect DNA methylation, an epigenetic mark that affects how genes are expressed. So mice with lower REs may carry different methylation signatures than those with higher REs, for example. “If the study had gone that one step further, it would have been even more significant,” Pleasure says. Kempermann agrees, and plans to look for possible epigenetic explanations in future experiments.

View Images
By Jason T. Ramsay, via Flickr

We will never know how the same experiment would play out with people, but the findings make intuitive sense, given what we know of identical twins. The data might also explain some intriguing differences in human intelligence, according to Turkheimer. Despite the fact that intelligence has a strong genetic component, “people have been getting smarter on IQ tests for the last century,” he says, a phenomenon known as the Flynn effect, after James Flynn, who first documented it.

Flynn and economist William Dickens proposed an explanation based on behavioral feedback loops. The environment after the industrial revolution became more cognitively demanding, and in response, people adapted, leading them to change the environment so that it’s even more demanding and requiring higher intelligence. “This study suggests that changes of that kind might actually have a neural basis, which is a very exciting possibility,” Turkheimer says.

Turkheimer describes this whole idea quite beautifully in that 2011 commentary (which, it’s worth saying again, is really worth a read):

“The nonshared environment, in a phrase, is free will,” he writes. “Not the kind of metaphysical free will that no one believes in anymore, according to which human souls float free above the mechanistic constraints of the physical world, but an embodied free will, tethered to biology, that encompasses our ability to respond to complex circumstances in complex and unpredictable ways and in the process to build a self.”

Who we are, in other words, is our genes, yes, and it’s our environment, yes — but also what we do with them.


The study was a joint effort of the German Center for Neurodegenerative Diseases in Dresden, the Center for Regenerative Therapies Dresden, and the Max Planck Institute for Human Development in Berlin.

If you’re interested in further reading on this subject, check out National Geographic’s feature on twins, or a story I wrote for New Scientist about scientists using these sorts of feedback loops to design intelligent robots.