National Geographic News
A photo of a chimp at a termite mound fishing with a probe and puncturing stick.

A chimp at Nouabalé-Ndoki National Park in the Republic of the Congo probes a termite mound with a stick.

Photograph by Ian Nichols, National Geographic Creative

Virginia Hughes

for National Geographic

Published July 10, 2014

Chimpanzees and other great apes are known for their intelligence: They can learn words, play with objects, and even seem to mourn the deaths of their friends. But just as for humans, cognitive abilities vary from one animal to the next.

Now, in one of the largest studies ever conducted on chimp cognition, researchers report that those individual differences are due in no small part to genetic makeup. The study appears Thursday in Current Biology.

Genes determine about half of the variability in chimp intelligence and environmental factors the other half, according to primatologist William Hopkins, of the Yerkes National Primate Research Center in Atlanta, Georgia, and colleagues. (Pictures: "How Smart Are Planet's Apes? 7 Intelligence Milestones.")

Research on animal learning has been focused almost entirely on the contribution from the environment. For most of the 20th century, scientists held that animals were like robots, behaving in predictable ways based only on environmental cues, such as reward and punishment. This new study adds to growing evidence that animals are not passive machines but rather are sharp, active thinkers.

Studies of humans have produced similar estimates to the primate study, suggesting that intelligence is approximately 50 percent heritable. But human development is heavily influenced by cultural factors, such as formal education systems, and so nature and nurture are difficult to tease apart, Hopkins said.

Being one of the closest relatives of humans, he said, "chimps offer a simpler way to think about that question."

Social Survival

Chimps can be surprising in their cognitive abilities, Hopkins noted. Several decades ago, a chimpanzee Hopkins was studying figured out that he could watch real-time video of himself on a nearby television monitor. A video camera was recording the chimp's actions.

The chimp, named Austin, opened his mouth wide to look at his teeth, but couldn't see them well.

"So he went and got a flashlight and shined it into his mouth to see farther down his throat," said Hopkins.

Austin was the smartest chimp Hopkins ever encountered, he said. But he also saw a lot of variability among the animals. "When you're out there working with them all the time, you definitely form some opinions about whether you think they're smart or not so smart," he said.

To find out how much of that variability is due to genetics, Hopkins and his team assessed the cognitive abilities of 99 captive chimpanzees. They used a battery of 13 tests measuring various manifestations of intelligence, such as how the animals dealt with the physical world, reacted to sound, and used tools.

The group of chimps tested had an expansive family tree, ranging from full siblings to fourth and fifth cousins. This allowed the researchers to calculate how well scores on cognitive traits aligned with genetic relatedness.

Two categories of tasks were significantly heritable: those related to spatial cognition, such as learning physical locations, and those that required social cognition, such as grabbing a person's attention. Some chimps are quite clever, making kissing sounds or clapping their hands to draw an experimenter's attention, Hopkins said. "This one is a real measure of intelligence and innovative behavior."

Performance on the two types of tasks did not correlate with sex or with whether the chimps were reared by their mothers or human caretakers, the researchers found.

The findings lend support to the so-called social brain hypothesis, which postulates that human intelligence evolved because it helped our ancestors to manage relationships in large and complex groups, Hopkins said.

That the study found spatial cognition to be heritable also makes sense, noted Josep Call, a comparative psychologist at the University of St. Andrews in Scotland, who was not involved in the study. "Thinking about space is extremely important for a number of animals. It's evolutionarily ancient," he said.

Puzzlingly, though, the study found that other cognitive skills, such as understanding causality and using tools, are not particularly heritable. "Why do they find it for space, why do they find it for social cognition, but they don't find it for tool use?" Call wondered. The ability to use tools, after all, would have also been an important skill for survival.

See footage of tool-making behavior that further blurs the line between humans and chimpanzees, our closest living relatives.

Environment Matters

Although Hopkins and colleagues found a strong genetic component to chimp intelligence, there were equally strong effects from environmental influences, which are malleable over time.

These results are similar to those in human studies, noted Ajit Varki, distinguished professor of medicine at the University of California, San Diego, who was not part of the new study.

What's more, the power of the environment may be underestimated by studies like these, Varki said. "In the impoverished and stereotyped setting of long-term captivity," he said, "the critical influence of environmental variability could be markedly blunted."

Virginia Hughes’s blog, Only Human, appears on National Geographic’s Phenomena site. Follow her too on Twitter.

Fornik Tsai
Fornik Tsai

Chimpanzees more and more like humans; humans understand more about chimpanzees don't know.

Fhgef Drerff
Fhgef Drerff

This is very interesting.  More studies on intelligence need to be done.

Scott Sinnock
Scott Sinnock

I am reminded of the thoroughly discredited blasphemes by Murray and Hernstein's Bell Curve in the 90's and Shockley and Jensen's article in the Harvard Educational Review in the late 60's. It seems their ghosts still haunt us.

Josh Dubnau
Josh Dubnau

 This is an interesting study, and in many ways a nice blog post (Virginia always does a good job!).  But i have one major bone to pick:  

It is simply wrong to state that intelligence is 50% genetic and 50% environmental.  This is probably the most common misconception about the meaning of the term "heritability".  Heritability does NOT measure what fraction of a trait is genetic.  It measures how much genetic variability contributes to the variability of that trait in that population.  This sounds like a subtle distinction, but it is MAJOR.  First of all, consider a thought experiment.  What would happen if we made a herd of dolly's sheep (clones) that are genetically identical.  Would they produce exactly equal quantities of milk?  Of course not, because we all know that environment will influence milk production.  So would we conclude that milk production has no genetic contribution?  of course not!  To quote my old friend Tim Tully "Of course its genetic.  They are making milk for christ sake! (well, thats a rated G version of the quote anyway)"  So this little example demonstrates that the quantity we call "heritability" does not measure genetic contribution to a trait in any meaningful or absolute sense.  "Heritability" does not mean "inheritable".  "Heritability" depends on the particular constellation of genetic variants in THAT POPULATION. And how those particular variants influence the trait.

But there is a second and even deeper issue, which is that even when genes DO influence a trait, they do so by interaction with each other AND by interaction WITH THE ENVIRONMENT! so a given gene may or may not influence a trait depending on the environment, and depending the other gene variants that happen to be in that particular animal.  And the magnitude of the gene's effect size will also vary depending on environment and the other genes too!  This almost certainly contributes to the so called "missing heritability" problem that runs rampant in human genetic studies.

Bottom line:  its an interesting study and blog post.  But the claim to quantify % of genetic influence on intelligence is dead wrong, and actually very dangerous.

Josh Dubnau, Assoc. Prof. CSHL, geneticist.

AJ Weberman
AJ Weberman

@Scott Sinnock you wouldn't be commenting without Shockley having put that grid on the diode. I remember reading about it in Popular electronics in the 1950's.

Fhgef Drerff
Fhgef Drerff

@Scott Sinnock The truth is hard to take for some close minded people that we all have different levels of intelligence.

Virginia Hughes
Virginia Hughes

Thanks, Josh. That is an important distinction, and you'll notice that on the first mention of heritability I defined it in terms of the variability in a population. After that I think it's acceptable to use the more conventional language as shorthand, though obviously you don't agree. :) --Ginny Hughes

Scott Sinnock
Scott Sinnock

@Josh Dubnau  It seems we love to quantify and love to see things as black and white, both lead to the desire to separate and quantify nature and nurture. We, to me, are 100% both.

Scott Sinnock
Scott Sinnock

@Fhgef Drerff  And even perhaps, as those two reports suggest from the simplest interpretation of test results, "intelligence" correlates with different groups of inbred people, i.e. races, or to use the modern genetic term, haplogroups. But I must note also that different types of "intelligence" (logic vrs music say) would be expected to likewise vary among groups, but not necessarily with the same groups always showing the higher scores, like the Asians in math. Tsk tsk tsk to mention such "false" possibilities which have been so thoroughly discredited by many because they conflict with our cherished political and moral theories of equality and justice.

Josh Dubnau
Josh Dubnau

@Virginia Hughes Thanks for your comment Ginny.

Yes, i do disagree actually.  The title says "Like humans, genes drive half of chimp intelligence".  Thats wrong.  And even in this population its not true in all environments.  The statement is limited to that population and that environment. There is are GXGXE (gene by gene by environment) interaction terms in reality, and those contributions are dropped from the equation in the calculation H^2 (heritability).  So the claimed conclusion is wrong in every meaningful way.  I really liked your tweet about the mis-reporting of this study in the Independent.  You are correct that they totally miss the point when they say that "Nature rather than Nurture" governs intelligence.  Your piece is obviously not in the same league of ms-statement.  And i know that you are not pushing a genetic deterministic view.  But i think that the slope is slippery.  Its really important to be clear about this.  Its also worth stating that i know many "trained geneticists" who do not fully understand this issue (not naming anyone, just saying).  So i dont blame you.  This is a subtle issue, but an important one.

I must admit that i have not yet read the CB paper, so i dont know if the authors over-state their findings or if they just didnt communicate it well to the press.  I promise to go read it carefully.




Featured Article

Latest From Nat Geo

See more photo galleries »

The Future of Food Series

See more food news, photos, and videos »