The Number That’s Supposed to Tell You How Smart You Are
The Number That’s Supposed to Tell You How Smart You Are
By JM Talboo DebunkerBusters.blogspot.com
There is a number that can follow a person for years.
It can appear on a psychological report, a school record, a conversation between parents, a social-media post, or a website where somebody takes a test for fun.
IQ.
We're told that the number represents intelligence.
But what exactly does that mean?
What does a score of 75 represent?
What does 100 represent?
What does 130 represent?
And perhaps the more important question:
What doesn't it represent?
I'm not writing this article because I think IQ tests are a scam. I'm not writing it because I took an IQ test once and didn't like my score. And I'm not writing it because I think everyone secretly has a genius-level IQ.
I'm genuinely interested in the question.
Because the more I look into IQ testing, the more complicated the word intelligence becomes.
And that's where this gets interesting.
Let's Give IQ Tests Their Due
Before criticizing IQ tests, we should acknowledge something important:
They aren't measuring nothing.
Modern intelligence testing has a substantial research history behind it. Cognitive ability tests can measure things such as reasoning, working memory, processing speed, verbal abilities, and visual or spatial problem solving. Different tests emphasize different abilities.
And a general factor often called g emerges from correlations among many cognitive tasks.
That isn't imaginary.
IQ and related measures also have real predictive relationships with things such as educational performance, training outcomes, and aspects of occupational performance.
So the easy argument—
“IQ tests are meaningless.”
—isn't good enough.
If a measurement consistently captures something and that something predicts certain outcomes, there is probably something real being measured.
The question is whether we have gradually taken that legitimate observation and attached a much bigger meaning to it.
Because measuring something real isn't the same thing as measuring everything important.
Even Binet Warned Us
This is where the history of IQ testing gets fascinating.
Alfred Binet, one of the pioneers of intelligence testing, wasn't simply trying to create a number that would permanently classify human beings.
His original work was connected to identifying children who needed additional educational assistance.
And Binet himself warned against treating his scale as though it were a literal, exhaustive measurement of intelligence.
He wrote:
“This scale properly speaking does not permit the measure of the intelligence, because intellectual qualities are not superposable…”
That's a remarkable statement coming from one of the people responsible for creating the foundation of modern intelligence testing.
Binet also described intelligence as an extraordinarily broad concept involving things such as judgment, practical sense, initiative, and adaptation.
He didn't believe a child's test result should become a permanent sentence about what that child could become.
In other words, one of the pioneers of intelligence testing understood something we sometimes forget:
A test is a test.
It isn't automatically the thing itself.
Neil deGrasse Tyson's Question
Neil deGrasse Tyson has made a similar criticism in modern times.
His argument isn't necessarily that cognitive tests measure nothing. His criticism is much more interesting than that.
The basic point is that humans developed tests involving particular kinds of cognitive tasks and then attached the enormous word “intelligence” to the resulting measurement.
Once we call something an intelligence test, people naturally begin treating the score as though it tells them how intelligent a person is.
But perhaps the more scientifically precise question would be:
What abilities does this particular test measure, under these particular conditions?
That's a much less glamorous question.
But it might be a much better one.
The Math Problem
One of the interesting arguments surrounding IQ tests concerns mathematics.
Some people argue that mathematical reasoning belongs in an intelligence test because mathematics requires sophisticated reasoning.
Others argue that extensive mathematical education and experience can give somebody an advantage that has little to do with their underlying general cognitive ability.
Both arguments contain something worth considering.
And this gets personal for me—not because my own IQ score proves anything, but because my own educational history makes me think about the problem differently.
I've never particularly loved mathematics. I had a complicated educational history, including substantial time away from a conventional classroom. I also discovered that I could be very motivated to learn things I actually cared about while having a completely different attitude toward subjects that didn't interest me.
So if two people walk into the same test with completely different educational histories, interests, practice, motivation, and experiences, what exactly are we comparing?
General cognitive ability?
Knowledge?
Training?
Test-taking strategy?
Motivation?
Or some combination of all of them?
Probably some combination.
That doesn't make the test useless.
But it makes the number harder to interpret than the number itself suggests.
And There Are IQ Tests That Aren't Just Math Problems
It's also important not to build a straw man.
Modern intelligence testing isn't simply a collection of arithmetic questions.
There are nonverbal assessments, including tests based heavily on visual and abstract reasoning. Raven's matrices, for example, ask people to identify patterns and relationships in visual information.
That's important because it demonstrates that intelligence testing has evolved considerably.
The criticism can't simply be:
“IQ is just how good you are at school math.”
It isn't.
The better question is:
How broad is the collection of abilities being sampled, and how justified are we in calling the resulting construct “intelligence”?
That's a much harder question.
The Number Can Be Real Without Being the Whole Person
Consider other numbers.
A bank account contains a number.
That number tells you how much money is in the account.
Website traffic contains a number.
That number tells you how many visits occurred.
A person's height is a number.
Their weight is a number.
Their age is a number.
Their marathon time is a number.
Their IQ is a number.
Numbers are incredibly useful because they simplify reality.
But simplification can become dangerous when we forget what was simplified away.
A bank balance doesn't tell you whether somebody is happy.
Website traffic doesn't tell you whether somebody's work is meaningful.
A marathon time doesn't tell you whether somebody is a good parent.
And an IQ score doesn't tell you everything about what a person can do.
Every statistic needs a question attached to it.
What question is the statistic answering?
What Is “Success”?
This becomes especially important when somebody says that IQ predicts “success in life.”
What is success?
Money?
Fame?
Career advancement?
Academic achievement?
Longevity?
Happiness?
Relationships?
Freedom?
Purpose?
Helping other people?
Spiritual development?
A person can be extraordinarily successful according to one definition and unsuccessful according to another.
Even longevity is a statistic.
I've joked before that when looking at somebody with enormous amounts of money, one thought that sometimes crosses my mind is:
You're not going to have that money forever.
Maybe I will live longer.
Then I've technically “won” the longevity metric.
But what did I actually win?
That's the problem.
Numbers are only meaningful relative to what we're trying to measure.
What About Spiritual Enlightenment?
Here's another example that makes the problem even stranger.
How would you measure spiritual enlightenment?
Years spent meditating?
Hours of meditation?
Self-reported happiness?
Compassion?
Ability to remain calm under suffering?
Reduction in ego?
Brain activity?
Changes in neural structure?
Researchers can measure some of those things.
Brain-imaging research involving experienced meditators has found differences associated with meditation practice and altered states of consciousness.
But even if we can find a reproducible neurological signature associated with a particular experience, that doesn't automatically prove somebody's interpretation of that experience.
At the same time, calling an experience “subjective” doesn't mean that nothing objective is happening.
If a person consistently reports a particular mental state and that state is accompanied by measurable physiological or neurological changes, then something is happening in the organism.
The remaining question is:
What exactly is happening, and what does it mean?
That's a scientific question.
And it illustrates a larger principle:
Measurement and definition aren't the same thing.
Fast Thinking Isn't the Same as Deep Thinking
Another criticism of IQ tests is that they reward speed.
There is some truth to the idea that many cognitive tests place time pressure on performance.
But speed itself can be a legitimate cognitive ability.
Rapid pattern recognition can be extremely useful.
So can slow, deliberate reasoning.
Consider writing a song.
I can sometimes generate rhymes very quickly. Years of practice have made certain rhyme structures, word combinations, multisyllabic patterns, and associations almost automatic.
But generating something quickly doesn't necessarily mean I've finished thinking about it.
Sometimes the first line is perfect.
Sometimes it's garbage.
The important part comes afterward:
Does it actually say something?
Can I examine it?
Can I find the deeper connection?
Can I improve it?
Can I recognize when a clever rhyme is just a clever rhyme?
Fast generation and deep evaluation are different processes.
And that's an important distinction when thinking about intelligence.
A person can become incredibly fast at something through practice.
That doesn't necessarily mean they are “thinking faster” in some universal sense.
And somebody who takes longer to arrive at an answer isn't necessarily thinking more deeply.
Knowledge Is a Double-Edged Sword
There's another paradox I've experienced myself.
If you know a subject extremely well, you can have an enormous advantage in a debate.
You know the history.
You know the terminology.
You know the counterarguments.
You know where the other person's argument is vulnerable.
You can probably make a much stronger case than somebody encountering the subject for the first time.
But winning the argument doesn't prove you're right.
It might simply mean you know more about arguing that position.
And expertise has another strange property.
Sometimes knowing a field extremely well makes you better at solving its problems.
But sometimes someone unfamiliar with the assumptions of the field can look at the problem and ask a question everyone else stopped asking years ago.
That doesn't mean outsiders are smarter than experts.
It means expertise and fresh perspective are different advantages.
A good critic can therefore be extremely valuable.
Someone who challenges your assumptions forces you to determine whether your argument survives outside your own head.
That's true in science.
It's true in mathematics.
It's true in journalism.
And it's certainly true when investigating controversial claims.
Evidence Can Be Right While the Conclusion Is Wrong
This is where another important distinction appears.
Imagine a researcher performs an experiment and finds a genuinely unusual chemical property in a material.
The observation can be real.
The experiment can be real.
The measurements can be real.
And yet the final explanation can still be wrong.
You can correctly establish:
“Something unusual is happening.”
without having established:
“I know exactly why it is happening.”
That's an incredibly important distinction.
A conspiracy hypothesis can contain genuine scientific questions.
It can contain legitimate physical evidence.
It can contain documents that really exist.
It can contain measurements that really happened.
And still arrive at an incorrect conclusion.
The reverse is also possible.
A claim can be dismissed because somebody calls it a “conspiracy theory,” even though individual propositions within it deserve investigation.
The label doesn't settle the question.
Evidence does.
Calling something a conspiracy theory doesn't make it false.
Calling something a scientific theory doesn't make it automatically true.
The same principle applies to IQ.
Calling something an intelligence test doesn't automatically make its score the complete measurement of human intelligence.
What About Street Smarts?
Here's where I think the IQ conversation gets particularly interesting.
Most people have heard the terms:
book smart
and
street smart.
We don't have a single universally recognized number that tells us how much street smarts somebody possesses.
But does that mean the ability doesn't exist?
Consider a simple real-world situation.
Someone can have excellent academic abilities and understand complicated material from a textbook while being surprisingly poor at judging what will actually happen when dealing with a physical situation in the real world.
Someone else might have less academic achievement but years of practical experience dealing with tools, machinery, people, hazards, unpredictable environments, and changing circumstances.
Those are different kinds of knowledge and ability.
That doesn't prove one person is globally smarter than the other.
But it does demonstrate that competence is multidimensional.
And perhaps that's the more interesting question.
When someone successfully handles an unfamiliar real-world problem, what exactly are we watching?
Pattern recognition?
Experience?
Risk assessment?
Improvisation?
Practical knowledge?
Working memory?
Planning?
Judgment?
Adaptability?
Maybe all of them.
What Would Kaku's IQ Test Look Like?
Physicist Michio Kaku has raised a related idea: perhaps an interesting test of intelligence would involve putting people into unfamiliar situations and asking them to generate realistic possible scenarios.
Imagine being stranded somewhere unfamiliar.
Imagine a major system suddenly stops working.
Imagine having limited resources.
What do you do first?
What could go wrong?
What are the realistic possibilities?
What happens three steps after your first decision?
That's a very different kind of cognitive challenge.
And it raises an uncomfortable question:
If intelligence is supposed to tell us how capable someone is, why don't we routinely test people on their ability to handle unfamiliar real-world problems?
Perhaps because those abilities are much harder to standardize.
And that's exactly why standardized tests are attractive.
They give us numbers.
But Standardization Is Both the Strength and the Limitation
A standardized test has enormous advantages.
Everyone receives similar tasks.
Everyone can be compared.
Results can be analyzed statistically.
Researchers can study reliability and prediction.
That's incredibly useful.
But standardization also means deciding beforehand which abilities count and which abilities don't.
The moment we construct a test, we create a particular window through which we look at the human mind.
The window can be scientifically useful.
But it remains a window.
What Happens When We Label Children?
This is where the issue stops being merely philosophical.
Imagine a child being told—or having their parents told—that their IQ is 75.
That number can sound devastating.
Now imagine discovering that the child has dyslexia and was struggling to read the questions.
A parent asks:
“Who read the questions to him?”
Suddenly the interpretation changes.
The point isn't that the child necessarily has a high IQ.
The point is that before interpreting a score, we have to understand what happened during the measurement.
Was the child able to understand the instructions?
Were there language barriers?
Was attention impaired?
Was the person exhausted?
Was there anxiety?
Was there unfamiliarity with the testing environment?
Was the person motivated?
Were there educational disadvantages?
Were there sensory, developmental, or learning factors that affected particular tasks?
Clinical psychologists can take many of these factors into account when interpreting cognitive testing.
But a number printed on a page can easily escape its original context.
And once it does, it can become a label.
The Danger of Turning a Snapshot Into an Identity
Suppose somebody measures your temperature once.
You get 101 degrees.
That tells you something useful.
But it doesn't mean:
“You are a 101-degree person.”
An IQ score is obviously different from temperature, but the analogy demonstrates the danger of confusing a measurement with an identity.
A score describes performance on a particular measurement.
It doesn't automatically tell us:
- how curious somebody is
- how creative they are
- how courageous they are
- how compassionate they are
- how persistent they are
- how practical they are
- how adaptable they are
- how much they know
- how much they are willing to learn
- how wisely they use what they know
- or what they might become
Some of those things may be related to cognitive ability.
Some may interact with it.
Some may be largely independent.
But they aren't all contained in a single number.
Education Changes the Question
There's another fascinating piece of evidence here.
Research has found that additional education is associated with improvements in performance on intelligence tests, with evidence suggesting that education itself can have causal effects on measured cognitive ability.
That doesn't mean one extra year of school magically adds a fixed number of IQ points to every person.
It means something much more interesting may be happening:
The environment can change cognitive performance.
Nutrition matters.
Development matters.
Education matters.
Experience matters.
Practice matters.
Sleep matters.
Motivation matters.
Genes matter.
And the brain itself changes with experience.
So if all of these factors interact with cognitive performance, what exactly does an IQ score represent?
Perhaps it represents something like:
How this person performed on this standardized sample of cognitive tasks at this point in their development, under these conditions.
That's a much more modest statement than:
This number tells us how intelligent this person is.
And maybe the modest statement is closer to the scientific truth.
What About Genetics?
Genetics clearly matters to cognitive ability.
But “heritable” doesn't mean “unchangeable.”
Human intelligence is influenced by enormous numbers of genetic factors, each generally contributing very small effects, interacting with environmental circumstances.
That's true of many human traits.
A genetic contribution doesn't mean that environment is irrelevant.
And environmental influence doesn't mean genetics is irrelevant.
The real picture is much more complicated:
genes → development → nutrition → education → experience → practice → environment → brain changes → performance
And eventually:
performance → test score.
Look at that chain.
How many things happened before the number appeared on the page?
Maybe the Real Question Is What Happens After the Test
Suppose two people receive IQ scores of 115 and 135.
What can we legitimately conclude?
We can say the testing data indicate a difference in performance on the measured cognitive abilities.
Depending on the test and circumstances, that difference may have meaningful predictive value.
But can we conclude that the person with 135 will make better decisions?
Will they be happier?
Will they be wealthier?
Will they be more creative?
Will they be kinder?
Will they be better at surviving a crisis?
Will they be better at understanding people?
Will they be wiser?
Will they be more spiritually developed?
Will they be more successful?
Not necessarily.
Some of those outcomes may correlate with cognitive ability.
Some may not.
And that's the entire point.
Correlation is not identity.
Here's an Experiment I'd Like to See
There's another question that interests me.
What happens when you dramatically change the incentive structure around education?
Imagine a voluntary research study.
First, establish baseline cognitive measurements.
Then randomly assign students to different incentive structures.
One group receives the normal educational incentives.
Another group receives substantial, guaranteed financial rewards for academic performance.
The exact amounts aren't the point here.
The point is the manipulation.
Would the relationship between baseline cognitive ability and academic achievement change?
Would students who previously performed below their measured ability suddenly perform dramatically better?
Would high-performing students improve too?
Would everybody improve?
Would grades improve without corresponding improvements on novel cognitive tasks?
That last question would be especially important.
Because there is a huge difference between:
getting better at school
and
becoming better at learning
and
becoming better at solving unfamiliar problems.
If money dramatically improves grades but not performance on novel problems, perhaps we learned something about motivation and academic performance rather than intelligence itself.
That's exactly the kind of experiment worth doing.
Not because we already know the answer.
Because we don't.
So, Are IQ Tests Good or Bad?
Maybe that's the wrong question.
A hammer is good at driving nails.
It is terrible at measuring the temperature of a room.
That doesn't make the hammer a scam.
It means we have to understand what the instrument is designed to do.
IQ tests may be excellent tools for measuring certain cognitive abilities.
They may provide useful information about learning and performance.
They may have meaningful predictive power.
They may help clinicians identify particular cognitive patterns.
All of that can be true.
And it can simultaneously be true that:
an IQ score is not a complete measurement of human intelligence.
Those statements aren't contradictory.
The Bigger Lesson
This is ultimately why I find the IQ debate so interesting.
It's not really about whether a number is “real.”
It's about what we do with the number.
A measurement can be scientifically useful and socially misused.
A statistic can be accurate and still be interpreted incorrectly.
A test can measure something important without measuring everything important.
A person can have enormous knowledge and still be wrong.
A person can lack formal education and still possess extraordinary practical ability.
A person can win a debate without having the truth.
A critic can expose weaknesses in an argument without having the correct alternative explanation.
A scientific theory can be wrong.
A conspiracy theory can be wrong.
A conspiracy theory can also contain a true observation.
And an intelligence test can measure real cognitive ability without being the complete definition of intelligence.
Maybe the question we should have been asking all along isn't:
“How intelligent are you?”
Maybe it is:
“What does this particular measurement tell us about you—and what doesn't it tell us?”
Because there is a profound difference between measuring a person...
and reducing a person to a measurement.
And especially when we're talking about children, that difference matters.
The number should be information.
It should never become a prophecy.
