What Does IQ Actually Measure?

The term IQ is widely present in popular culture, with discussions about IQ tests and their implications. Achieving a score of around 120 on an IQ test is often seen as a benchmark. Many individuals are unaware of their own IQ scores. The concept of IQ is frequently mentioned, emphasizing intelligence as a quantifiable measure of cognitive capability. However, the effectiveness of IQ tests remains a question. This video aims to explore the origins of IQ, its actual scope of measurement, its predictive capacities in life, and even the narrator's personal IQ assessment. Despite having never taken an official IQ test, the narrator believes their intellect to be slightly above average. Humorous science banter about exothermic and endothermic reactions ensues. While various online IQ tests are available, their accuracy is doubted. Regardless, the narrator intends to use them as preparation for a genuine IQ test they plan to take tomorrow. The narrator is optimistic about the potential for IQ improvement through practice, and this will be tested in the upcoming session.

The concept of evaluating intelligence through testing dates back many centuries, but a significant breakthrough occurred in 1904. Charles Spearman, an English psychologist, made this breakthrough when he investigated the performance of students in various subjects. He was curious about the relationship between their achievements in different subjects, such as English and Math.

One possibility was that high performance in one subject, like Math, might lead to poorer performance in another, such as English, due to time allocation constraints. This could result in a negative correlation between performance in different subjects. Another option was that performance in one subject might be completely unrelated to performance in another, given the distinct skill sets required for different subjects.

The third option proposed a positive correlation: better performance in Math might correspond to better performance in English. The correlation coefficient, which can range from -1 to +1, quantifies these relationships. A value of -1 represents a perfect negative correlation, where an increase in one variable exactly predicts a decrease in the other. Similarly, a value of +1 represents a perfect positive correlation, and 0 indicates no correlation.

The square of the correlation coefficient reveals the extent to which variations in one variable can be explained by variations in the other. For instance, a correlation coefficient of 0.5 implies that 25% of one variable's variation can be explained by the other.

Upon analyzing his data, Spearman found a clear positive correlation. Students excelling in Math tended to excel in English as well, with a correlation coefficient of 0.64. This pattern extended beyond Math and English to other subjects like Classics and French. The same positive correlation existed across all subjects, suggesting that strong performance in one subject was indicative of strong performance in others.

The question arises: How can this observation be explained?

Spearman put forth the notion that every individual possesses a level of general intelligence, which he termed the g-factor. This construct aimed to encompass an individual's capacity to quickly grasp new information, recognize patterns, and engage in critical thinking across various subjects, thereby explaining the correlated scores across different subjects. Those with a high g-factor tend to excel in all subjects, while those with a low g-factor perform poorly in all subjects. Spearman detailed these findings in his paper "General Intelligence" Objectively Determined and Measured. Despite not achieving perfect correlations, he introduced the concept of subject-specific factors, known as s-factors, to complement the g-factor. For instance, a student's math performance relied on both their general intelligence and a math-specific factor.

Spearman believed that specific factors were malleable, whereas general intelligence remained constant. He sought a reliable method to measure this overarching intelligence. Concurrently, in France, Alfred Binet endeavored to identify children requiring additional educational support. Alongside Theodore Simon, he created the Binet-Simon test. This assessment involved tasks such as identifying missing elements in drawings, defining abstract terms, and repeating sentences. One question even inquired about the most appealing face. After comparing students' performance to peers of varying ages, they were assigned a "mental age." This value was divided by their actual age, then multiplied by 100 to yield the intelligence quotient, or IQ.

The Binet-Simon test, later translated and adopted by Goddard in the U.S., became the world's inaugural IQ test. Lewis Terman standardized it at Stanford University, transforming it into the Stanford-Binet test with adjustments. For years, it remained a widely employed evaluation in the United States. However, this marked only the beginning, as numerous other IQ tests emerged, all aiming to gauge the g-factor. They achieved this by assessing a spectrum of cognitive abilities encompassing memory, verbal, spatial, and numerical skills.

Though these areas could exhibit subject-specific variations, the objective was to average them out, mitigating the influence of subject-specific biases and approximating the g-factor. While some degree of error persisted, psychologists designed IQ tests with numerous distinct sections, each containing various tasks, to minimize subject-specific deviations. Diverse IQ tests differed in question count and complexity. To standardize scoring, each test was administered to a sizable population sample. Raw scores were normalized, usually resulting in a mean of 100 and a standard deviation of 15—this practice remains standard today.

This concept is referred to as IQ, serving as an indicator of an individual's g-factor compared to the broader population. The measurement is calibrated in such a way that around 68% of individuals possess an IQ ranging from 85 to 115, while only approximately 2% achieve scores above 130 or below 70.

Now, let's consider this math puzzle:

Eleven lions, four cats, and seven crows have a combined total of...

Oh dear.

While preparing for my IQ test, I dedicated time to practicing the various question types commonly found in modern IQ assessments. One section that invariably appears is related to vocabulary. In this segment, a single word is provided, such as "sanguine," and candidates must select the option from multiple choices that best mirrors its meaning. Choices might include "gloomy," "asinine," "recalcitrant," "optimistic," or "reflective." Conversely, the opposite meaning could also be queried. For instance, what is the antonym of "perspicacious"? Could it be "canny," "obsequious," "dull," "fanciful," or "sagacious"?

Another test segment evaluates one's aptitude for recognizing numerical patterns. Participants are tasked with selecting the number that effectively completes a given pattern. Consider this sequence: 3, 5, 8, 12. What number comes next? At first, I sought complex patterns, but as I acquainted myself with online tests, I realized that the patterns were generally quite straightforward. An effective strategy involves identifying the difference between consecutive terms. In this instance, the initial two terms are separated by two, followed by three, then four. Consequently, the logical next term should be five more than 12, equating to 17. The answer is option C. Occasionally, sequences exhibit rapid growth, as seen in the progression: three, 15, 60, 180. The question arises: What comes next? In situations like this, I focus on the ratio between a number and its preceding counterpart.

In this particular scenario, the second number exhibits a fivefold increase compared to the first. The subsequent number is four times greater than the preceding one, followed by another that is three times larger. Consequently, the solution would be twice the value of the fourth term, which is 360, as found in option B.

Among the most renowned IQ test question types are Raven's progressive matrices. These involve a three by three grid containing symbols in each cell, and the challenge is to identify the ninth cell that follows a certain pattern. I discovered that the majority of these puzzles adhere to a small set of logical rules. One is translational motion, where symbols move predictably from one cell to the next. Another is rotational motion, where objects rotate from one cell to another. The third involves missing symbols, where each symbol appears once in each row or column, requiring one to identify the absent symbols. The fourth rule is based on addition, where the sum of the first and second cells equals the third cell. This rule considers overlapping lines canceling out, and a line added to nothing equals a line.

In most contemporary IQ tests, all questions must be answered under time constraints, often allowing only about 10 to 30 seconds per question. Today, I'm scheduled to take an official IQ test, and I admit to feeling a fair amount of nervousness. I consistently strive for strong test performance, a point of personal pride. However, the outcome remains uncertain. Unfortunately, I won't be able to bring you along, as test content is confidential, and videos of the testing environment are restricted. I'll embark on the test, provide an update afterward, and share my experience.

One remarkable aspect of IQ tests is their ability to predict a significant amount about an individual's life based on an hour or two of questions covering vocabulary, numbers, and abstract shapes. Notably, a higher IQ tends to correlate with a larger brain size. A comprehensive meta-analysis in 2005 indicated a correlation coefficient of 0.33 between IQ and brain size.

Having a high IQ is essentially akin to possessing a larger brain. Moreover, IQ holds predictive power in relation to academic achievement. In 2007, Ian Deary, a Scottish psychiatrist, gauged the IQs of approximately 13,000 11-year-olds. After five years, when these same students undertook national school examinations, Deary compared their exam scores with their initial IQ scores. The correlation between their IQ scores at age 11 and their performance on GCSEs was strikingly high, around 0.8, indicating that about two-thirds of the variation in national school exam scores could be forecasted by IQ test results taken five years prior. Although this study's correlation coefficient lies toward the upper end of the 0.2 to 0.8 range observed in similar studies, it firmly supports the notion that IQ effectively predicts school success. Moreover, it also forecasts the extent of an individual's educational attainment.

This might not be surprising, given that some standardized school tests essentially resemble IQ tests. Notably, assessments like the SAT, ACT, and GRE exhibit a correlation of about 0.8 with standard IQ tests. Personally, I obtained an SAT score of 1,330, approximately corresponding to an IQ of 130. It will be intriguing to ascertain if my official IQ score aligns with this or if my familiarity with IQ-style questions influenced my results. However, IQ's predictive capacity extends beyond academia.

One of the most robust findings is that IQ can anticipate success in the job sphere, particularly in technical or intricate roles. Measuring occupational success typically involves evaluating performance through ratings from supervisors, income levels, and productivity metrics used by economists, with correlations generally ranging from 0.2 to 0.6. The effect is most pronounced in complex positions, which is logical. Military training exhibits the most substantial impact. In fact, the U.S. military does not accept candidates with IQs below 80, and they cap the number of recruits with IQs between 81 and 92 at 20%. During the Vietnam War, this requirement was eased to expand the applicant pool, but those below the threshold were 1.5 to three times more likely to fail recruit training, necessitating three to nine times more remedial training. This burden eventually rendered the military more effective without these additional recruits. Ultimately, 5,478 recruits from this initiative died, facing a fatality rate three times higher than regular recruits. The military restored its requirements, resulting in the ineligibility of anyone with an IQ below 80—approximately 30 million Americans. Even beyond military contexts, IQ seems to play a role in longevity. A Scottish study tracked individuals from the age of 11 to 76, revealing that on average, a 15-point increase in IQ corresponded to a 27% higher likelihood of reaching age 76. This relationship is confirmed by a comprehensive meta-analysis, indicating that individuals with higher IQs experience a lower risk of mortality within the timeframes examined.

Lastly, IQ is also predictive of income levels.

This study provides clear evidence of a tendency for income to rise alongside IQ, yielding a correlation coefficient of 0.3. However, the variance is considerable; remarkably, the top three earners in this study all possessed IQs below 100. A comprehensive meta-analysis encompassing 31 studies unveiled a correlation of 0.21 between IQ and income, which, though statistically significant, is relatively modest. This signifies that merely 4.4% of the variation in income can be accounted for by IQ.

The reason for a less robust correlation between IQ and income could potentially be attributed to the fact that intelligence is not universally rewarded to the same extent within economic contexts. For instance, some vocations, such as certain real estate ventures, may not demand an exceedingly high level of intelligence, yet can yield substantial earnings. Simultaneously, professions that often attract highly intelligent individuals, such as academia, might not offer corresponding high financial remuneration. Many individuals with exceptional IQ scores might not be driven by the same desire to accumulate wealth.

The connection with net worth is even more tenuous. Despite higher IQ individuals generally earning more annually, this doesn't necessarily translate into greater wealth accumulation. This disconnect raises a pertinent question: If IQ is correlated with academic achievement, job performance, income, and longevity, why isn't it more widely discussed and tested?

One reason can be traced back to IQ's complex history. When Henry Goddard introduced Binet's test to the United States, its purpose and interpretation underwent significant changes. Binet had designed his test with the belief that intelligence could be enhanced through education, intending to aid struggling students in catching up. However, in the U.S., the modified test was used to rank adults based on their intelligence. Spearman and other researchers suggested that the general intelligence factor (g) was unalterable, a trait established at birth and unchanging throughout life. Additionally, many believed that g was hereditary, passed down from parents to their children—nowadays, this concept is attributed to a genetic foundation.

There exists evidence supporting these notions, as IQ tends to remain relatively stable across a person's life. In a study, individuals were tested at age 11 and retested at age 90, with their scores correlated at around 0.5 to 0.6. The genetic basis for IQ is also apparent; identical twins show a strong correlation in their IQ scores, akin to retesting the same person a few weeks apart.

Henry Goddard utilized the assertions that intelligence was hereditary and unalterable to position IQ at the forefront of the American eugenics movement. Eugenicists aimed to hinder individuals with undesirable traits from procreating. In numerous states, laws were enacted that permitted the compulsory sterilization of individuals who fell below a certain threshold on IQ tests. The constitutionality of these laws was validated by the Supreme Court in 1927. Surprisingly, terms such as "moron," "idiot," and "imbecile" were employed as scientific terminology at that time. In his ruling, Justice Oliver Wendell Holmes articulated, "Society can prevent those who are manifestly unfit from continuing their kind. Three generations of imbeciles are enough." Consequently, more than 60,000 people were involuntarily sterilized due to these laws, which later became a model for Nazi Germany. Hitler himself acknowledged being influenced by American eugenicists, and some Nazis even quoted the American Supreme Court decision during the Nuremberg trials.

Given this disturbing history, it's unsurprising that many people currently dismiss IQ entirely. Regarding the science of intelligence, numerous misconceptions emerged from the early researchers' work. One misconception is that genetics solely determine IQ. However, the effects of genetics versus environment cannot be definitively quantified for ethical reasons. Twin studies indicate an approximate 50/50 split between heritability and environment over a lifetime. Though a precise estimate for humans remains elusive, based on the available literature, the range likely lies between 40% and 70%.

Moreover, IQ isn't entirely static, as education can enhance it over time. Furthermore, intelligence might not be a single monolithic trait, as initially believed. Contemporary science recognizes two intelligence forms: fluid and crystallized. Fluid intelligence pertains to learning, information processing, and problem-solving, while crystallized intelligence encompasses accumulated knowledge. Both types of intelligence increase during childhood, but fluid intelligence peaks in early adulthood before gradually declining.

On the contrary, crystallized intelligence maintains a higher degree of stability. However, IQ has been subjected to further misuse by promoting the notion of racial disparities in intelligence. An observed discrepancy exists between the average IQ scores of black and white Americans. Additionally, articles have been published regarding the IQ scores of various nations across the globe, with many purportedly displaying average IQs below 70, the threshold for intellectual disability. This has led some to postulate genetic differences in intelligence between races or nations. Yet, I contend that this interpretation of the data is a severe distortion.

The predicament, in my view, is that IQ tests don't necessarily gauge what they're assumed to measure. A telling example lies in a representative sample of white Americans whose average IQ is 70, consisting of ordinary individuals who lived around a century ago. Researcher James Flynn analyzed the average IQ test results spanning the past century. These tests are periodically updated and re-normalized to maintain an average score of 100. Flynn observed that with each re-normalization, the scores needed to be slightly adjusted downward by approximately two to three IQ points per decade. If this adjustment wasn't applied, the result would be a consistent increase in the average IQ of the entire population, accumulating to around a 30-point boost. This phenomenon is termed the Flynn Effect.

The question arises: were our recent ancestors on the brink of intellectual impairment, considering 70 is commonly associated with mental retardation? Alternatively, are we approaching a state of universal giftedness, as 130 is the threshold for such categorization? Genetics haven't experienced significant changes over the past century. Thus, what brought about this increase? Although the true causes are debated, improving childhood nutrition and health, better education, and a shift from manual labor to more abstract thinking are suggested factors. This shift in labor may have enhanced our aptitude for answering IQ test questions.

Despite their purported objectivity in measuring intelligence, IQ tests are not as definitive as they seem. Even within a single country, separated only by time, cultural shifts can influence the average IQ scores. Given this, why shouldn't we anticipate similar cultural disparities to exert an effect between different groups concurrently? Some tests are marketed as "culture fair," implying their questions are equally applicable to all cultures. However, constructing such a test is impossible. Culture fair tests evaluate visual relations, geometric shapes, and patterns, yet disregard the fact that cultures diverge in their linguistic resources for shapes and spatial relationships. These discrepancies influence how categories are conceptualized and utilized. Moreover, cultural variances might impact the perception of printed materials. Culture fair tests overlook areas like ethnobotanical knowledge, dog training for hunting, or survival skills in rainforests—potentially more critical forms of intelligence for survival than predicting the next sequence of numbers.

However, these abilities are less prevalent in our culture, and our means of measuring them are limited. Consequently, IQ puzzles are perceived as the definitive method for quantifying intelligence, and even the creators of these tests concur. Rigorous standards are enforced before a test validated for one population can be applied to another vastly different population. Within the limited scope of intelligence that IQ aims to assess, factors beyond the g-factor can influence the final IQ score, including motivation. The extent to which an individual is incentivized to complete the test can significantly impact their score. Various studies have explored the effect of offering payment to subjects for completing an IQ test, and a large meta-analysis demonstrated that such motivation increased IQ scores. The larger the monetary incentive, the greater the average improvement, with some cases witnessing an increase of up to 20 points, particularly among those with below-average IQs. Therefore, IQ tests not only measure general intelligence (g) but also motivation.

The impact of training and coaching is also noteworthy, as it can enhance IQ scores by up to eight points. Personal test-taking strategies, adeptness at managing time pressure, and even anxiety levels contribute to performance. A small amount of anxiety can be beneficial, but excessive anxiety can hinder performance.

Moving on to my personal experience with the test, I believe I did well, and the training I received was certainly beneficial. After completing the test, I experienced a bit of difficulty speaking due to its demanding nature. The math section, in particular, felt easy, and I'm confident in my performance there. The time limits posed the most challenging aspect, especially in discerning patterns within sequences of shapes, where I felt I couldn't finish some questions. Skipping questions, eliminating incorrect choices, and knowing when to guess were also crucial strategies.

Anxiety, too, played a role, with a small level being advantageous, but excessive anxiety proving detrimental to performance. In conclusion, I'm satisfied with my performance and believe the training was instrumental. Now, let's fast-forward to the future and assess my actual results.

Upon receiving my results from the author of the IQ test, it's evident that I displayed strengths and weaknesses across different areas. My performance in the math and numerical sections was strong, resulting in a quantitative index of 143, placing me higher than 98.8% of the population. For crystallized intelligence, I scored 132, and for fluid intelligence, I scored 118, which still places me higher than 88.5% of the population. This variance isn't unusual, as individuals tend to excel in some areas while being less proficient in others.

The concept of g, or general intelligence, accounts for this variation, and strengths and weaknesses are natural within this framework. My best estimated g score for this set of tests is 134, which is higher than 98.8% of the population. Overall, I'm content with the results and my performance.

Turning to the broader implications of IQ, it has practical applications in various fields. In my clinical practice focused on forensic neuroscience, IQ is pertinent in cases involving the death penalty. An example is the Atkins defense, named after the Supreme Court case that eliminated the death penalty for individuals with intellectual disabilities. It's worth noting that attempts to intentionally fake poor performance on an IQ test are detectable using mathematical algorithms, with over 95% accuracy in identifying such behavior.

Looking ahead, there's potential interest in enhancing cognitive abilities early in life to delay cognitive decline. This delay could extend the period before individuals face functional difficulties in daily tasks as they age, such as handling finances or reading labels, providing greater independence in later stages of cognitive decline.

If a method to permanently enhance people's intelligence could be uncovered, it would undoubtedly offer significant benefits. One of its potential applications lies in identifying individuals with remarkable intellectual capabilities that may not have been apparent through other means. Traditionally, teachers might recommend a student for a gifted and talented program based on their classroom performance. However, when an IQ test, a standardized objective measure, is introduced, a higher proportion of students from less privileged backgrounds or minority ethnic backgrounds can be identified. This approach aims to counteract inherent biases and offers a more impartial assessment.

Historically, gaining admission to prestigious schools often depended on social connections, parental influence, or financial means, overshadowing individual performance. The concept of developing a somewhat objective measure, like an IQ test, aimed to level the playing field by minimizing these social biases.

The relationship between IQ and society, as well as the perception of IQ, is complex. The public and psychologists share a love-hate relationship with intelligence and IQ test scores. While psychologists may be hesitant to discuss intelligence and test scores, the general public's attitude toward IQ is diverse. Extremes exist on both ends of the spectrum, with some emphasizing its importance to an extreme degree, while others dismiss it as irrelevant or even harmful due to associations with racism and prejudice. The more moderate middle-ground perspectives tend to be overshadowed in this polarized discourse.

It's crucial to avoid the mistake of equating IQ with an individual's worth. The true value lies in how one interacts with and supports those around them. As Stephen Hawking aptly put it, "People who brag about their IQ are losers." While IQ can offer insights, it doesn't determine life outcomes. Improving one's knowledge and analytical skills can profoundly impact personal achievements. An example of such a resource is Brilliant.org, which provides opportunities to master various subjects, from data science and math to programming and technology. Brilliant offers customized learning paths to help individuals achieve their goals, emphasizing that building skills and knowledge is a more valuable pursuit than fixating solely on IQ scores.