Chapter 8: Intelligence: Theory & Measurement

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement not replaced the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Welcome back to the Deep Dive.

Today, we are undertaking what is arguably the most fundamental and I think often the most challenging deep dive in human psychology.

Absolutely.

We are putting the foundational concept of intelligence right under the microscope.

Yeah.

We're working through a core textbook chapter, a really dense objective analysis of what intelligence is, how we scientifically measure it, and crucially where the capacity for it truly originates.

It's a monumental task and you're right the source material is rigorous.

I mean understanding intelligence is really key to understanding almost everything else about human behavior, our mental processes, problem solving, our whole ability to adapt.

Exactly, our ability to adapt effectively to this incredibly complex changing environment.

If you want to know how humans think, this chapter really gives you the basic blueprint.

Okay, so let's unpack this with a little historical context.

It's always a good place to start.

This concept, it's not new.

The source material tells us the term intelligence goes back over 2 ,000 years to Cicero and he used it in the sense of understanding.

And even before him you have Plato grappling with this.

He drew a really clear distinction between the cognitive aspects, what he called reason, and the emotional aspects.

The driving force.

The driving force of human action, yeah.

And that separation between cognition and emotion is one we, you know, we still use today.

Right.

So if we fast forward to the modern psychological foundation, we run into Sir Francis Galton and Herbert Spencer.

Yes.

They were instrumental in bringing the term into popular, and more importantly, scientific consciousness in the 119th century.

They explicitly linked intelligence to the ability of an organism to adjust effectively to its environment.

And Galton specifically, he's a titan here, he not only contributed to that initial concept, but he was also the first person to realize the potential of studying twins.

The natural experiment.

Exactly, that identical twins who are sometimes separated could serve as a natural experiment to investigate the role of heredity in this mental ability.

It was just a revolutionary methodological idea at the time.

And this whole historical line, I mean from Plato's reason to Spencer's adaptation, Galton's methodology,

it all sets the stage for the core definition we are using today.

A definition that was primarily formulated by Cyril Burt in the early 20th century.

And that definition, which is our intellectual compass for this entire deep dive, is intelligence as innate, general cognitive ability.

Three very loaded words.

Incredibly loaded, especially for a textbook.

So our mission then is to spend the next 30 plus minutes systematically dismantling and analyzing those three components, innate, general, and cognitive.

And we're going to do that by rigorously exploring the theories, the exact measurement methods, the biological basis, and of course the criticisms that are presented in the source material.

All right, let's get into it.

Let's start with that word general.

When we talk about intelligence in the field of psychometrics, we are almost always referring to the concept of G, or general ability.

And the central, and I'd say pretty bold hypothesis driving the concept of G, is that all cognitive problem solving and abstract mental activities.

So whether you're solving a word puzzle, calculating a complex equation,

figuring out spatial relations.

Right, all of them require the application of one single underlying mental ability.

They might require it to varying degrees, but that core mental horsepower is shared across all tasks.

And the power of G as a concept really rests on this universal empirical finding that the source material points out again and again.

Which is?

That all tests of cognitive ability, regardless of how specialized verbal, numerical, visual, spatial, memory, you name it, they always correlate positively with each other.

Always.

So if you score high on one type of test, you tend to score above average on all the others.

Exactly.

And that consistent ubiquitous positive correlation is exactly why the hierarchical model of intelligence works so well.

Right.

At the bottom, you have these very specific abilities, but binding them all together, crowning the entire structure, is that unified G factor.

It's really the synthesis of those ancient ideas of reason and the modern requirements of problem solving.

So that's general.

What about the second word in the definition?

Cognitive.

It sounds a bit abstract.

It does, but its root is actually pretty simple.

Cagittare in Latin, which just means to think.

So it refers to mental activity that's concerned with thinking.

Okay.

And within this framework, the key functions of cognition are what?

Being aware of external reality, processing information efficiently, solving problems.

And this specific phrase, which is key, the adduction of relations and correlates.

That's the operational definition used by a lot of psychologists, isn't it?

It is.

And that is our perfect transition to Charles Spearman, because Spearman provided the foundational framework that's used to define the entire field of cognitive activity by laying down three fundamental laws.

Let's start with the most basic one, then.

Okay.

The law of apprehension.

Apprehension in this context, it doesn't mean fear.

It means a person's power to grasp outer reality and their own internal consciousness.

Exactly.

And today we'd probably interpret this as the power to encode and transmit information.

And psychologists have devised a classic experiment, the reaction time study, to measure this raw encoding ability and see if it actually correlates with intelligence.

The beauty of the subject is that it separates motor speed from processing speed.

How does it do that?

Well, first they measure raw RT.

The subject simply presses a key when a single light goes on.

This baseline time is remarkably fast, something like one fifth of a second.

Okay.

But here's where it gets interesting.

The researchers introduce complexity.

They start to complicate the task by adding choices, which in turn adds bits of information that have to be encoded and processed.

Right.

So if you have two possible lights that could go on, you now have one bit of information.

Four possible lights is two bits and eight possible lights is three bits of information.

And crucially, the subject doesn't know which light will go on, so they have to process the uncertainty that's inherent in the task.

Now, when they measure the raw RT, that's the one light, zero bits of information, there's generally no difference between a subject judged bright and a subject judged dull.

They can both press the key equally fast.

Exactly.

Their motor speed is the time.

But when you start adding those bits of information, the reaction time of dull subjects increases disproportionately compared to bright subjects.

Ah, so the gap widens as complexity increases.

It widens significantly.

The duller the person, the steeper the increase in their reaction time for every additional bit of information that has to be encoded and transmitted to the brain.

So this difference is the signature of intelligence correlating with that elementary ability to encode information efficiently.

It suggests that G is fundamentally about the speed and efficiency of this initial neural transmission.

That processing speed is essential, but the real meat of intelligence, of course, is what we do with that information once it's encoded.

And that brings us to Spearman's second law.

The law of adduction of relations.

This is defined as the power to bring to mind relations that essentially hold between two or more items of content or ideas.

So it's that spontaneous understanding of a connection.

If I present you with the words black and white or high and low.

The relation of oppositeness just immediately comes to mind.

It's adducibly brought to mind.

You don't have to consciously calculate the link.

It's an inherent power of the mind.

And that second law is indispensable for the third one.

Absolutely.

The law of adduction of correlates.

This states that when a person has one idea together with a relation, they possess the power to bring up the correlative item.

And this is the basic structure that defines pretty much every analogy test and really a huge number of IQ test items.

It is.

Take the classic example.

Black is too white.

As high is too high.

Okay.

So first you use the second one, the adduction of relations, to establish that the relationship between the first pair is opposition.

Then you use the third law, the adduction of correlates, to apply that same relation to the third word high, which generates the correlative answer low.

Exactly.

Intelligence tests, therefore, are explicitly constructed to measure how successfully you can discover these relations and correlates across all kinds of different materials, words, numbers, shapes, and crucially across a massive gradient of complexity.

The examples provided in the source material really drive this home.

At the simplest end, stuff you'd use for younger children, you have things like tall, short, up.

Or those simple odd man out tasks, like figuring out which item doesn't fit in the sequence car train violet buses.

Violet, obviously.

The others are modes of transport.

Right.

But the complexity can jump dramatically, requiring many more steps and a much deeper abstraction of the relationships involved.

I'm thinking of that complex analogy given in the text.

Beethoven,

Schoenberg -Diggins.

Yes, that's a great example.

The correct answer offered is

To solve this, you have to first discern the relationship between Beethoven and Schoenberg.

It's not just that they're both composers.

No.

It's that Schoenberg was a revolutionary figure who fundamentally broke the classical structures established by Beethoven.

He was a radical departure.

Okay.

So the relationship is one of revolutionary foundational divergence from a classic master in the same artistic field.

Exactly.

So you then have to apply that very relation to Dickens, a classic literary master.

The correlative figure must be the revolutionary literary figure who fundamentally broke the classic narrative structures that Dickens established.

And that would be James Joyce.

That is James Joyce.

That reasoning is just orders of magnitude more intricate than simple opposition.

And this abstract relational reasoning is absolutely not limited to verbal tasks.

Not at all.

Psychologists use complex non -verbal items like the figure matrices or diagrammatic analogies you see in the source figures.

These require the exact same process, discovering relations and correlates, but they use visual symbols, rotations or patterns instead of words.

Right.

So you might be presented with, say, a three by three grid with a missing square.

And to solve it, you might have to track multiple dimensions at the same time.

For example, maybe the number of lines increases by one as you go across while the shading rotates 90 degrees as you go down.

And the little symbol inside the shape alternates between a star and a circle.

It'll hold all that in your head.

Right.

You have to mentally track and manipulate all that complex information.

And because researchers want to reduce the role of chance or just random guessing, they tend to use many of these less complex problems to test these laws reliably rather than relying on one single super hard problem.

Which would just be too vulnerable to error or a lucky guess.

Precisely.

Okay.

So now that we've rigorously defined intelligence through these cognitive laws from Spearman, the immediate challenge becomes, how do we actually put a number on this invisible internal process?

That challenge was solved by Alfred Binet in Paris, working at the Sorbonne in the early 1900s.

His work was absolutely groundbreaking because he was the first to construct an intelligence test that correlated robustly with external real world criteria.

Like teacher ratings and actual school achievements.

Exactly.

And Binet's genius was in his realization that he could leverage a natural observable difference.

Older children are, on average, smarter than younger children.

He could use age as his yardstick.

And he constructed tasks to capture this age -related growth.

A prime example is the figure copying test.

This test has about 10 figures arranged by difficulty, starting with simple shapes like a circle.

And then moving up through a cross, a square, a triangle, and ending with really complex figures like a cube or a box divided by multiple diagonals.

Right.

And the instruction is simply copy the figure.

But what's crucial, as the source points out, is that success on these developmental items is not easily coached.

You can't just teach a four -year -old to draw a perfect cube.

It's practically impossible to teach a young child to succeed on a drawing item that is fundamentally beyond their natural developmental ability.

And this resistance to simple instruction is what makes the test correlate so strongly with general intelligence and less with a specific cultural or educational background.

And this test, and many others like it, allowed Binet to introduce the pivotal concept of mental age, or MA.

Mental age is the age level whose average performance a child's performance equals, regardless of the child's actual chronological age, their CA.

So let's break that down.

If the average seven -year -old in your sample group can draw the rectangle, but fails on the complex cube, then any child who achieves that same level of drawing proficiency has a mental age of seven.

Whether their actual birthday age, their CA is five, which would make them intellectually advanced, or 10, which would make them intellectually slow, their performance is anchored to the average seven -year -old mind.

Right.

And the initial Binet -Simon scale used a 30 tests, ranging from the most elementary stuff, like simple eye -head coordination in babies, all the way up to complex verbal tasks for older children.

Like distinguishing between abstract words like liking versus respecting.

Now, Binet initially focused on the simple difference between MA and CA, but that presented a pretty serious problem.

It did.

A two -year gap, say, an MA of eight and a CA of six is incredibly diagnostic in

But that same two -year gap at a later age, say, MA 14, CA 12, it means something entirely different.

Because intellectual development slows down as we approach adulthood.

The absolute difference just doesn't account for the developmental rate.

So the solution came from the German psychologist W.

Stern.

He suggested using a proportional relationship instead of an absolute difference.

He proposed dividing the mental age by the chronological age.

And just like that, the intelligence quotient was born.

The IQ.

They multiply the quotient by 100 to get rid of the decimals and keep the numbers clean.

And by statistical and historical definition, the average person, where MA equals CA, has an IQ of 100.

Let's walk through the math because it really reveals the genius of Stern's approach.

Okay, take child A, mental age of eight, chronological age of six.

The IQ is eight divided by six times 100, which is 133.

So this child is developing intellectually at a rate significantly faster than average.

Right.

Now take child B, mental age of eight, chronological age of 12.

The IQ is eight divided by 12 times 100, which is 67.

This child is developing at a rate much slower than average.

The IQ perfectly captures that relative rate of intellectual growth.

It does.

And when we look at the entire population, IQ follows that standard bell curve distribution, a lot like height or weight.

And it's important for you to know this distribution.

Roughly 50 % of the entire population scores between 90 and 110.

The extremes are statistically rare.

Very rare.

Only about 1 % of the population scores 140 or over, which is often considered the genius level.

And similarly, only 1 % scores 60 or below, which falls into the range of mental defect.

This distribution raises the critical question of constancy.

I mean, how stable is this score over a lifetime?

That's a great question.

An infant's IQ is notoriously inaccurate.

Infant tests relate more to physical and motor development than to true cognitive potential.

But the source material notes that by around age six, we can get a reasonable first estimate of the final adult IQ.

And after age eight, the IQ is remarkably steady and resistant to change, barring some major external interference like brain damage or a severe prolonged illness.

What about on the other end of the spectrum?

Does it decline?

Deterioration is also slow.

It typically begins after age 60 and is often linked to physical changes like arteriosclerosis, which impacts blood flow to the brain.

We should also address the technical point about adult IQ.

Right.

Since mental age stops growing around age 16, you don't suddenly get twice as much MA at 32 as you did at 16.

The quotient formula breaks down.

If we kept the strict formula, everyone's IQ would just perpetually drop as they got older.

Exactly.

So IQs for adults are calculated statistically based on

for that age cohort, but the practical term IQ is still used.

Okay, now for the social implications.

This is huge.

The source material provides a really striking analysis in Table 8 .1, which details the mean IQs of various occupational groups.

And this data is consistent across decades and multiple studies.

It gives us a window into the societal utility of G.

At the highest level, professions requiring intense abstract thinking.

We're talking higher professionals, top civil servants, research scientists.

They show a mean IQ of 140.

Lower professionals like lawyers and engineers average around 130.

Moving down, you get school teachers, accountants, managers who sit around 120.

And the numbers keep regressing toward that mean of 100.

Foremen and clerks are around 110.

Machine operators and semi -skilled workers hover near the 100 -105 mark.

And then laborers and farm hands average around 90.

The overall observation is profound.

There's a roughly 50 -point difference in the mean IQ between the upper middle class and the semi -skilled unskilled working class.

It's vital, though, to stress the caveats.

These are averages.

Huge overlaps.

There's tremendous overlap.

The source material explains that it is far easier to have a high IQ, say 120 or 130, and fail to reach an upper middle class occupation.

Due to illness, bad luck, lack of drive.

Right.

Then it is to reach that professional level with a genuinely low IQ.

This overwhelming correlation forces us to ask that critical question about cause and effect.

Does social class determine IQ, or is it the other way around?

Does being rich make you smart, or does being smart allow you to become rich?

The evidence leans heavily toward the latter.

Studies that track children within families show clearly that IQ determines social mobility.

So brighter children tend to move up the social scale relative to their parents, regardless of the starting environment.

And duller children tend to move down.

Intelligence is the dynamic driving factor that causes that vertical movement within the social structure.

Social class itself does not generally determine the IQ score.

We also have to address the common criticism that IQ tests are just achievement tests.

That they're basically arbitrary riddles written by the middle class for the middle class.

And therefore they only measure cultural exposure.

Right.

But the empirical evidence strongly contradicts this.

First, IQ does correlate highly with educational achievement.

But that's because a certain level of intelligence is necessary to benefit from the instruction in the first place.

You need the cognitive capacity to absorb and apply the learning.

But I think the strongest counter -evidence is the cross -cultural data.

The Eskimo example.

Yes.

The sources cite the fascinating finding that Canadian Eskimos score slightly better than white Canadians on IQ tests, despite obviously not sharing a middle -class Canadian cultural background.

Which demonstrates that the tests are measuring some underlying general ability that transcends specific cultural conditioning.

Absolutely.

And you simply cannot fundamentally coach a person to a brilliant performance on an IQ test.

There's a minor practice effect, sure.

But you can't coach a genuinely dull person to score high.

Unless, of course, you literally teach the test.

Right.

Where they just memorize the exact answers to the specific items being administered.

That's no longer a test of intelligence.

Okay.

Now we arrive at the most contentious, but also empirically driven, part of Burt's definition,

innate capacity.

This is the big one.

The core assertion here is that intelligence, the biological capacity for learning, the raw potential, is largely innate and cannot be fundamentally increased by environmental means.

Although, of course, a person's knowledge base can be expanded.

This immediately leads to that famous quantitative division of total variation or variance in intelligence within a population.

The 80 -20 split.

The strong consensus among researchers, which this chapter leans on heavily, suggests that 80 % of this variance is attributed to heredity or nature.

Leaving only 20 % to the environment or nurture.

And that 20 % is that small, definite slice where environment can make a difference.

To really grasp the weight of this 80 -20 split, we need to understand what variance is.

Variance, simply put, is the degree of scatter or the extent to which scores differ from the average.

So if everyone scored 100, the variance would be zero.

Correct.

And we can visualize what happens when you remove one of the variables.

Let's look at figure 8 .2 in the text.

It illustrates this beautifully.

What's it showing us?

It contrasts the existing IQ distribution, the shaded bell curve, with a hypothetical curve where all the environmental variance, that 20 % contribution, has been removed.

And the difference between these two curves is minor.

It confirms that environment, while present, contributes minimally to the overall spread of intelligence we see.

But now look at figure 8 .3.

This is the existing IQ distribution versus the hypothetical curve if all the genetic variance, the 80 % contribution, were removed.

And the difference is marked and profound.

The distribution just collapses into a very narrow vertical spike.

It demonstrates that genetics is the overwhelming, powerful factor responsible for the diversity and range of intelligence scores across humanity.

Now, it's crucial to understand that this conclusion, that 80 % of the variance is genetic, it's not based on one study.

It's based on a massive convergence of evidence.

The source material presents 10 different indicators.

We have to explore these to really justify that 80 % claim.

Let's do it.

So, proof number one.

Let's start with proof number one.

Identical twins reared apart.

These 122 pairs that were studied are genetic clones sharing 100 % of their heredity, but they grew up in entirely different unrelated environments.

And the critical baseline here is the test -retest reliability.

That's the error you get when you test the same person twice.

That error is about 4 .5 IQ points.

Right.

Identical twins reared apart only differed by an average of 6 .6 points.

So, if you subtract the standard measurement error, their actual difference due to environment is just 2 .1 points.

That level of similarity, despite completely different upbringings, is a devastating argument for the genetic influence.

It's huge.

And proof number two builds right on this.

Identical versus fraternal twins.

So, identical twins, 100 % share genes, show an IQ correlation of 0 .86.

But fraternal twins, who share only 50 % of their segregating genes, just like normal siblings, they correlate at a much lower 0 .55.

That substantial difference between 0 .86 and neutral 0 .55 strongly supports the genetic model.

Proof number three involves kinship correlations.

This is where genetic theory can actually predict the expected correlation between any two relatives based on the percentage of genes they share.

So, for instance, second cousins are predicted to correlate at 0 .14.

And the observed value is 0 .16.

Uncle Sans and nephew's nieces are predicted at 0 .32.

And the observed value is 0 .34.

The fact that the observed correlations consistently and so closely mirror the genetically predicted values provides massive support for the hereditary hypothesis.

It does.

And number four is a consequence of inbreeding effects.

Studies show that marriages between close relatives, like cousins, lead to a predictable lowering of IQ in the offspring.

Why is that?

It's due to increased genetic homozygosity and the expression of recessive deleterious traits.

It's another clear indicator of the genetic determination.

Proof number five is conceptually one of the most powerful.

Regression to the mean.

This is the foundational law of genetics.

Extreme traits in parents, like extreme height or extreme intelligence, are statistically pulled back toward the population mean in their offspring.

So let's look at Bird's study cited in figure 8 .4.

Fathers who were professionals with an average IQ of 140 had children who, on average, regressed downward toward the mean, scoring around 120 plus daily.

And conversely, fathers in the unskilled laborer group, averaging an IQ of 85, had children who regressed upward toward the mean, averaging 90 plus.

This is so counterintuitive and difficult to explain purely through environment.

If the parents with an IQ of 140 provide the absolute best environment, the most resources, the most nurturing,

their children should maintain that 140 score or even exceed it.

But they regress down.

And if the parents at IQ 85 provide the worst environment, their children should score even lower, but they regress up.

This upward and downward statistical pull is the signature of genetic inheritance overriding environmental influence.

It's a humbling point for high achieving parents, right?

Even with the best resources, the genetic tide pulls back toward the average.

It is.

Now, proof number six addresses orphanage studies.

This is a classic environmental test.

Children raised in orphanages experience a highly uniform environment, similar food, education, staffing, books.

So environmentalists predicted that this equalization would dramatically reduce the variance in IQ scores.

But the result was a letdown for them.

The reduction in IQ variance was less than 10 percent.

Which aligns perfectly with the 80 -20 split.

Removing the 20 percent environmental difference makes very little difference to the overall scatter of scores.

Exactly.

Proof number seven involves foster children's studies.

If environment were primary, a foster child's IQ should correlate strongly with the IQ of their foster mother, who provided the home and the learning environment.

But instead, studies consistently show the foster children's IQs correlate strongly with their biological mother's IQs.

While correlations with their foster mothers are generally low or statistically non -existent.

Number eight reinforces this.

Home feature correlations.

Researchers tried combining multiple measurable environmental factors in foster homes.

The parent's socioeconomic status, the number of books, the mother's dedicated time with the child.

All of these environmental factors combined could only account for about 17 percent of the total IQ variance.

Again, fits neatly within the 20 percent environmental allowance.

Okay, proof number nine looks at direct manipulation attempts.

There have been endless claims of educational programs raising IQ through tuition.

But the source material introduces a critical methodological warning here.

The failure to eliminate the experimental error of teaching the test.

Right.

This happens when experimenters, often inadvertently, teach the students the specific contents of the final test during the tuition phase.

The score goes up, but the general cognitive ability, G, has not increased.

And when these studies are properly controlled to eliminate that error, true IQ increases remain well within that 20 percent environmental boundary.

Okay, finally, proof number 10.

Two kinds of mental defect.

This is a crucial clinical distinction.

It is.

Children with IQs between 50 and 70 are typically the low end of the normal population distribution curve.

They often come from families where parental IQ is also low.

But the second group, those with IQs below 50, often categorized as imbeciles, they're different.

Very different.

The majority suffer from a singular severe gene disorder that effectively overrides all other influences.

These children come from all family types and parental IQ is often average.

And this group creates a small hump at the very bottom of the distribution curve, highlighting that fundamental difference between multigenic normal low IQ and single gene pathological low IQ.

So given all this evidence, the chapter heavily emphasizes that attempts to fundamentally shift innate capacity by purely environmental means are extremely difficult.

We have two classic case studies to illustrate the reality of these manipulation attempts.

The first is the highly publicized failure of Hayne's abdominal decompression.

The claim was sensational.

Using suction decompression on pregnant mothers to increase oxygen to the developing fetus's cortex.

And supposedly causing a 30 % advancement in mental age in the children by age two.

But the research just collapsed under scrutiny.

First, as we already established, IQ tests at age two are terrible predictors of adult IQ.

And second, there was severe design error.

The study relied on volunteers who were overwhelmingly highly educated, high IQ, middle -class mothers already predisposed to having bright children.

So when properly controlled tests were run, eliminating that volunteer bias, no difference in IQ was observed at age three.

It was a bust.

The second case study offers a hint that biological pathways might be more promising even if it's limited.

Glutamic acid.

It was once optimistically thought to raise IQ by 10 points.

Right.

And while initial results were confusing, the modern conclusion is nuanced and specific.

Glutamic acid has virtually no effect on subjects of average or above average intelligence.

However, it does raise the performance level of dull children and those classified as mental defectives.

And this effect is also observed in rats performing maze tests.

So this suggests that thorough experimental research into optimal doses and identifying the precise physiological mechanisms of action might offer a more viable, albeit small, path forward than purely educational manipulation.

It's a biological problem.

A biological solution seems more plausible.

If intelligence is indeed 80 % innate, we should be able to find direct, measurable physiological evidence of it in the brain.

The source material leads us into neuroscience with evoked potentials, or EPs.

EPs are brain waves recorded by an EEG in response to sudden, sharp stimuli like a tone or a flash of light.

Figure 8 .5 in the text illustrates the typical wave components, the positive P and negative N deflections relative to the stimulus onset, S.

And these brain wave reactions are highly consistent within an individual.

More importantly, when you compare genetic relations, the similarity of EPs is much greater in identical twins than in fraternal twins.

Which confirms their strong genetic component.

So what's the direct link to intelligence?

EPs correlate with IQ scores.

Brighter subjects show measurable physiological differences compared to dull subjects.

Specifically, they exhibit smaller latencies, meaning shorter waves, and greater amplitude, meaning bigger waves.

Wait, okay, why do shorter and bigger waves indicate more intelligence?

Shorter latency reflects a faster signal processing time.

It takes less time for the neural signal to reach its destination and generate the response waves.

So that gives us the first hard physiological evidence for the speed element of G.

Exactly.

It echoes the findings of the reaction time experiments we discussed earlier, and the greater amplitude likely indicates a more robust or synchronous firing of neurons.

Both point toward greater neural efficiency underlying the psychological score.

This biological determination brings us naturally to the question of sex differences.

Historically, it was a common politically charged myth that men were intellectually superior.

A myth that was used to justify limiting women's education and social freedoms.

But the development of objective intelligence tests just dismantled that myth.

Standardized tests showed, right from the start, that girls and women were, on average, equal or even slightly superior, especially in childhood, to boys and men in overall IQ scores.

The source material does note an important subtlety here, though.

This exact equality in overall average IQ might be considered a statistical artifact.

Well, test makers, observing this overall balance, had to balance the test items.

They had to ensure that the few items where men excelled were offset by items where women excelled.

But the fact remains that this balancing was only possible because the sexes were approximately equal to begin with.

Okay.

But this average equality hides two major differences, and the first one is spread and variation.

Figure 8 .6 illustrates this dramatically by comparing the male and female distribution curves.

The figure shows that the female curve is narrower, more peaked, women are more uniform, with a surplus falling in that middle range, the 9110 IQ range.

And the male curve, conversely, is flatter and wider.

It shows greater dispersion.

The crosshatched areas at the extremes below 80 and above 120 show a surplus of males.

So the conclusion is that men dominate both the mentally defective category and the genius category.

There are more very dull men and more very bright men, while women are clustered more tightly around the average.

And while this variability is an attractive explanation for the social paradox of why men dominate certain highly competitive fields, the source injects a sociological caution.

Right.

The certification of mental defect is often a social act, not purely a psychological one.

What do you mean by that?

Well, for example, high IQ psychopathic males are sometimes certified because society doesn't know what else to do with them.

And furthermore, because society traditionally requires men to earn a living, low IQ males are more visible as failures.

Whereas women with similar scores might remain in the home until married, making their lower ability less socially salient.

Exactly.

And the same complexity applies to genius.

The source is very clear.

The lack of female genius dominating certain fields is not due to a lack of high IQ women.

Instead, the potential causes lie in non -cognitive factors.

Primarily social disability, the challenge of balancing a highly demanding career with family expectations and certain personality traits.

The personality traits necessary for outstanding success in these competitive, often male -dominated fields, aggressiveness, dominance, fierce assertiveness, are often closely linked to male hormones.

So those traits, rather than IQ itself, may be the limiting factor for high IQ women reaching the very highest peaks of competitive success.

Okay, so that's the first difference.

The second major difference is the distinct patterns of specific abilities that make up the overall IQ score.

Both sexes have areas of

Starting with elementary sensory differences, which relate back to that law of apprehension.

Women typically show lower thresholds for touch and pain, better auditory discrimination, and greater sensitivity to olfactory stimuli.

And these differences are observed from birth, so they may be genetically linked to sex hormones.

Right.

Men, conversely, show better visual discrimination, and are more responsive to visual erotic stimuli.

And when we turn to the higher order cognitive functions, the division is sharp.

Men excel in visual spatial ability.

This is the skill required to organize and relate visual stimuli in complex spatial contexts, and critically, the ability to imaginatively manipulate or mentally rotate visual material.

Think of those complex figure matrix examples we described earlier, or mentally tracking which gear turns in which direction in a machine diagram.

Exactly.

And this male superiority is evolutionarily ancient.

It's observed in other mammals, including rats and chimpanzees, which strongly suggests a deep biological and genetic link.

The evolutionary suggestion is that males required this ability for success in hunting, defense, and long -range territorial roaming.

And conversely, women demonstrate clear superiority in verbal skills.

Girls learn to talk earlier, develop a better vocabulary faster, and show generally superior language usage across the board writing, spelling, grammar, sentence construction.

Early verbal ability in girls is actually a better predictor of adult IQ than it is for boys whose development is more erratic.

It's essential, though, to specify the nature of this superiority.

Yes.

This female advantage pertains primarily to the executive aspects of language.

The source states women do not hold an advantage over men in verbal reasoning problems.

So those complex analogies are logical problems where the solution requires higher order relational reasoning, even if the problem is stated verbally.

Correct.

Other measured differences include women excelling in rote memory, remembering unrelated or personally irrelevant facts.

Men tend to excel in numerical and mathematical abilities,

specifically complex mathematical reasoning and abstract manipulation, not simple arithmetic.

And finally, men are superior in mechanical tasks, which is linked to their visuospatial edge, while women excel at manual dexterity.

This primary ability, asymmetry visuospatial on one side, linguistic skills on the other, is linked by the source material to cerebral asymmetry.

So the brain itself.

The left hemisphere is generally dominant for language, and the right is concerned with spatial and non -linguistic skills.

And there's evidence that myelination and dendrite growth is more advanced in the left hemisphere for young girls and the right hemisphere for young boys, suggesting a deep biological foundation for these observed pattern differences.

Okay.

Despite the convergence of evidence supporting the robustness of G and its genetic basis,

intelligence testing has faced valid and persistent criticisms that we have to address.

The first one is the difficulty in separating heredity and environment in the measurement process itself.

Right.

Since IQ tests use acquired skills and information, words, numbers, symbols, it seems impossible to truly measure innate capacity outside of an acquired environmental context.

It's like Galileo trying to measure the laws of motion outside a vacuum.

Friction or environment is always present.

The psychological solution, and this is a key distinction, is the construction of tests that rely less on cultural knowledge, what are known as culture fair tests.

These tests, like abstract figure matrices, depend primarily on the ability to perceive and manipulate nonverbal relations, minimizing the learned component compared to, say, verbal analogies.

And this distinction led Raymond Cattell to formalize two powerful concepts, fluid ability versus crystallized ability.

Okay.

What's the difference?

Fluid ability is the uncommitted potential energy, the biological raw capacity, best measured by those culture fair tests.

It's the genotype of intelligence, the raw power to reason.

And crystallized ability.

Crystallized ability is the kinetic energy.

It is the result of applying that fluid potential to specific items of knowledge gained through education and culture.

It's measured best by culture -dependent tests, like vocabulary or general knowledge quizzes.

So while the two correlate closely,

fluid ability testing is essential for properly

handicapped children or making accurate group comparisons.

As demonstrated by the Eskimo example, they score well on the culture fair tests, but less well on the culture -bound tests due to their different knowledge bases.

The second major criticism is aimed at the format of the test items themselves.

Standard IQ items are typically convergent.

They demand a unique, prescribed, single correct solution.

And critics argue that this format neglects crucial mental attributes, suggesting that divergent tests might better capture things like originality and creativity.

So these divergent tests are scored not on correctness, but on the quantity of solutions offered.

Like, think of as many uses as possible for a brick.

Exactly.

But here is where the empirical findings get really interesting and unexpected.

What does the research show?

Research shows that the correlation between convergent, so IQ tests, and divergent, so creativity tests, is actually quite high.

Often surprisingly so.

And there's no solid evidence that genuinely creative people, famous artists or inventors, perform demonstrably better on divergent tests than on standard IQ tests.

So what variable does correlate with high divergent scores?

Personality.

Specifically, extraversion.

Extroverts tend to score better because they are more willing to rapidly generate and list a high quantity of solutions, including the sillier or less conventional ones.

Whereas introverts tend to be more self -critical and suppress ideas they deem unworthy.

Precisely.

And since divergent tests score quantity over quality, this highlights a personality bias rather than a distinct measure of creativity separate from G.

A third criticism suggests that the selection of test items is just arbitrary.

That the tests are just random riddles that happen to be correlated, but they lack any fundamental scientific meaning.

The rebuttal here is definitive.

Item selection is not arbitrary.

Items survive the rigorous development process only if they meet objective psychometric principles.

Namely, if they show a consistent and high pattern of inter -correlations within the population.

And the items that correlate most highly with G are those that best incorporate Spearman's three cognitive laws.

Which suggests that the procedures are lawful, scientific and meaningfully measuring an abstract cognitive reality, not just random generalization.

Finally, we reach the most philosophical criticism and one that must be addressed carefully.

Intelligence is not a measure of a man's worth.

And this is the one criticism where psychologists fully concur.

Science, by necessity, involves abstraction.

Intelligence, defined as the capacity for abstract thinking and problem solving, is one abstract virtue.

It is distinct from compassion, altruism, integrity, or bravery.

A highly intelligent person can still be utterly lacking in ethical character, a self -seeker, or a criminal.

IQ tests have absolutely nothing to say about these other essential dimensions of personality and humanity.

Nothing at all.

So what does this all mean for us?

Despite the test's limits,

intelligence remains a profoundly useful variable, socially and practically.

The ability to measure this variable with some precision allows nations to manage the complexity required for survival, especially those that lack great natural riches.

And society must support every individual in reaching the highest level of achievement their innate endowment permits.

So we've completed a pretty comprehensive deep dive into the source material.

We established that foundational definition of intelligence as a robust general cognitive ability G, measured via Spearman's laws.

We demonstrated that IQ is a highly stable measure after age 8, and that the variation we see in the population is overwhelmingly determined by heredity, that 80 % genetic variance versus 20 % environmental.

We track the journey from ancient philosophy all the way to modern biological findings like evoked potentials, which confirm a measurable speed and efficiency in the neural structure underlying G.

And we examine the nuanced differences between the sexes, finding equality in overall IQ, but profound likely biological differences in the patterns of specific abilities, with men showing greater variability in visuospatial superiority and women showing better verbal skills.

The information is clear.

Intelligence is critical for societal survival and progress.

Given the strong evidence presented here regarding the innate biological determination of IQ and knowing that it is an abstract virtue distinct from moral character, here is a final provocative thought for you to consider.

How should modern society balance the imperative to identify and maximize the use of that high innate ability for collective scientific and organizational progress, while simultaneously fostering and rewarding those other equally essential virtues like altruism, compassion, or ethical leadership, which IQ tests specifically do not and cannot measure?

A deep and enduring challenge that extends far beyond the textbook data.

Thank you for joining us on the Deep Dive.