Why Standardized Tests Fail to Gauge Intelligence in People on the Spectrum

Three years ago, Philip Reyes, a seventh grader with nonverbal autism, learned to express himself through Rapid Processing Method (RPM), a learning and communication method for those with severe autism. Through this strategy, those who are minimally verbal can spell out what they want to say by pointing to a letter board.

And through it, Reyes has shown himself to be a fantastic writer. Here’s just part of a compelling article he wrote about his life experiences:

[Before I learned RPM] Teachers were well meaning but believed I could not understand much of anything because I could not talk or write to communicate that I was smart and understood everything going on around me. I became like a pet to train, as everyone tried to make me act normally with candy rewards….

Do not underestimate us. Our minds are bright but many times our bodies cannot follow the instructions from our minds. For example, my mind might want me to hold a pencil to write a note. But nothing works. My hand can’t grasp correctly. I can’t coordinate the fine movement in my hand to form letters. I can’t find my hand in space. I only make soft scribbles.

People see my body acting awkwardly. They mistake my body’s actions for evidence that my mind is dull. I believe many of my autistic peers have the same problem.

Cases like Reyes’s have cropped up over the years: people unable to communicate through speech have gotten their hands on a keyboard and written things that shocked others with their intelligence. See, this severely understudied community is commonly thought to be intellectually disabled—to have an IQ below 70.

But there’s evidence that may not always be the case.

In reality, the issue of intelligence is pretty complex when it comes to autism, and some experts think there’s at least a chance we’ve underestimated the nonverbal population. Intelligence is tricky to measure and has, perhaps, been measured the wrong way in times past. But for the sake of those with smarts buried deep, we need to start digging a little further and consider rethinking some beliefs we’ve held on to.

Measuring intelligence on the spectrum

Especially for those with severe autism, it’s hard to accurately gauge intelligence for several reasons. These include factors like sensory issues, misunderstanding test instructions, or simply thinking about the situation differently than neurotypical people might. Additionally, a person with autism might do poorly on a test but express proficiency with the same skills in other contexts.

The test used could also have an impact. The Wechsler Intelligence Scale for Children (WISC) is commonly used to evaluate intelligence in kids. A significant portion is verbal, and many of the questions evaluate knowledge of social norms. So the test unintentionally yet inherently stacks the odds against those with autism. As the old adage states, we can’t judge a fish by its inability to climb a tree, but that’s essentially what the WISC does for kids on the spectrum—especially those who are nonverbal.

As Reyes explained in his article, sensory-motor difficulties may also come into play for some with nonverbal autism. That is, their brain works but their bodies won’t cooperate and do what they want them to, when they want them to do it. This understandably affects testing performance. Ido, another blogger with nonverbal autism, experienced this exact phenomenon during his childhood ABA therapy:

I was asked to demonstrate my understanding of basic concepts by pointing to flashcards arrayed in front of me….My mind might be screaming, ‘Touch tree! Don’t touch house!’ and I would watch, like a spectator, as my hand went to the card my hand, not my brain, wanted. And down in the data book it would be marked that I had not yet mastered the concept of tree.

So it’s hard to measure intelligence on the spectrum, and many people—both verbal and nonverbal—remain underestimated because standardized tests are not created for them. However, when those on the spectrum are given intelligence tests other than the WISC, the results are remarkably different.

For example, one small study gave kids with severe autism four tests: the WISC, Raven’s Colored Progressive Matrices (which is a picture-based intelligence test), the Children’s Embedded Figure Test, and a visual search test.

Neurotypical kids often score pretty consistently across both verbal and nonverbal tests. In contrast, the kids with severe autism were unable to finish the WISC but did significantly better on the latter tests. On the Raven test, more than half achieved an IQ score of 75 or higher, and 10 percent achieved an IQ score of about 120: that’s considered moderately gifted!

What does this all mean?

This is not to say that some on the spectrum do not have intellectual disabilities. We still can’t be entirely sure about intelligence because a) the nonverbal population has been severely understudied, and b) intelligence is difficult to measure with standardized tests. But it is to say that blanket statements are unfair.

And there are consequences of assuming someone on the spectrum is unintelligent—especially if they have nonverbal autism, since they may not be able to speak up and defend themselves.

First of all, it insults people. When someone has an average or above-average intelligence, they may feel angry, frustrated, or hurt when people talk to them differently or when people talk about them as if they are not in the room. The latter can even spark behavioral problems, and it’s no wonder.

Second of all, it locks up potential. Consider how many scientists or writers could slip through our world—a tragedy that may take off with an incorrect assumption about cognitive ability. The world could benefit from these people’s insights and wisdom; yet we may never realize it because they’re locked away in a mind we don’t think can hold such treasures.

What can we do to avoid this tragedy?

As the authors of one case study stated, we should make “the least dangerous assumption,” which “suggests that when there is no absolute evidence [of cognitive ability], it is essential to make the safest and most respectful assumption that would be least dangerous to the individual if proven to be false.”

Soma Mukhopadhyah, the inventor of RPM, did just that with Tito, her nonverbal son. She taught him a lot, either by reading extensively to him, talking directly to him, or exposing him to interesting places. She assumed he was smart, and she was right; Tito is now a published writer.

And that kind of treatment is what those like Reyes probably want, too. As he succinctly wrote in another article: “To help me best, please see me as an equal human being as you. That means treat me the way you want to be treated.”