Why is our society so invested in seeing intelligence as something that’s written into our genetic code? I googled “intelligence test online,” and without even clicking on anything, you can see a common narrative around intelligence:

— “Take our free IQ test and find out your true IQ score!” boasts one site.
— “Get your IQ score quickly and find out how smart you are,” screams another.
— “This is a very accurate IQ test. Will test your brain in 38 questions,” yells yet another.

Each of these sites tries to entice you with a claim that their test is the true measure of who you really are. In this common narrative, your true IQ score, how smart you are, is a precise and quantifiable quality that — yes! — science can reveal.

One of the best examples of this narrative comes from this group, which has their very own intelligence test to determine whether you’re worthy. The very logos of the society and their test are squishy and organic: they subtly remind you of molecules, genomes, DNA. The message? You’re either among the Chosen, or you’re not.

How would you feel being a card-carrying member of this elite club, knowing not only that you have what it takes, but that it’s encoded into your molecular structure? There’s a funny link between our conceptions of intelligence — about who has it and who doesn’t — and our self-worth. When we find something we are good at, it helps to think about our ability as fixed and immutable, because it’s a guarantee that your membership card cannot be revoked.

The motivation to see what we are good at as an immutable quality can become sinister when we talk about group differences. In several previous blogs (see here, for example) I have covered the effects of negative stereotypes on performance. But when talking about gender differences in math achievement, or racial differences in intelligence, do members of the “groups on top” benefit from seeing those differences as immutable and not subject to change?

My colleagues and I put this idea to the test in a paper published in the Journal of Experimental Social Psychology (Mendoza-Denton, Kahn, and Chan, 2008). In one study, we looked exclusively at Asian-background participants, who are stereotyped as being good at math. We led half of them to believe that scientists had in fact found that the stereotype was true; the other half was led to believe that the stereotype was false. Within each of these groups, participants were led to believe that math ability was either innate, or the result of hard work and effort. Thus, we ended up with four groups:

a) stereotype confirmed, and math performance is innate;
b) stereotype confirmed, and math performance is effort-based;
c) stereotype disconfirmed, and math performance is innate;
d) stereotype disconfirmed, and math performance is effort-based.

We then had all of the participants take the same math test. The result? The Asian-background test-takers who believed their group was on top, and who also believed that their math skills are innate outperformed all the other groups. They actually benefitted from believing that their high ability could not be revoked.

As it turns out, there is another social category for whom stereotypes of math ability applies: gender. We did the study again, and this time looked at the performance of both men and women after leading them to believe that the stereotype of men being better at math than women had either been confirmed or disconfirmed by scientists. As in the first study, within each group half were led to believe that math performance is innate, and half that math performance results from hard work. We found — again — that the favorably-stereotyped group performed best when they believed the source of their advantage is fixed. But we also found that precisely in these same conditions, the unfavorably-stereotyped group performed worst.

Same exact test, same exact questions, but when these tests are framed as assessments of innate group differences, we see the most dramatic divergences in performance.

The tragic thing about this is that these very moveable performance differences are then used as “evidence” of innate differences — invoking the sanctity of the “IQ test” as an X-ray into the brain or the genetic code (see, for example, here). Beliefs about the innate nature of human abilities are tools to perpetuate and magnify achievement gaps — a pretty scary thought when we consider that immutable notions of ability are so deeply embedded into our educational structures (e.g., tracking by “ability levels”). It doesn’t have to be this way — check out this post, for example — but we need to understand that revoking valued membership cards is going to be met with resistance.