PS201 - Intelligence - Lecture 1

So, this lecture is the first in a series of four on intelligence, and it's all about theories — basically, what do we even mean when we say someone is intelligent? Before we get into the meat of it, it's worth knowing that the full series will also cover how we measure intelligence, what intelligence tests actually tell us, and finally what determines intelligence in the first place (things like genes and environment). But for now, we're focusing on the theory side of things, and by the end of this lecture you should be able to give a rough history of intelligence research, talk about what everyday people think intelligence is, and describe and compare the main academic theories.

Let's start with why any of this matters. Intelligence research doesn't just live in a university lab — it genuinely affects society. At the policy level, it informs decisions in areas like health, law, and education. Organisations like schools and workplaces use it to make decisions. And even at the individual level, it can influence lifestyle choices. So it's a pretty big deal.

Now, where did all of this start? Well, you can trace thinking about intelligence all the way back to around 350 BC, with Plato and Aristotle. Their basic idea was that we gain information through our senses — a kind of passive intellect — and then we have an active intellect that makes sense of that information and allows us to use it. That's actually not a million miles away from how we still think about it today. Fast forward to 1869, and we get Francis Galton, who was one of the first people to try and study intelligence scientifically. Then in 1904, Alfred Binet arrived on the scene, and he is the person responsible for what became the first proper intelligence test.

Binet, along with his colleague Simon, created the Binet-Simon Scale in 1905. It contained 30 tasks, and they were designed to measure different mental abilities — things like recalling a sequence of numbers, rearranging words into a sentence, or completing word associations (like "cat goes with... mat"). The key idea Binet introduced was the concept of mental age. So imagine two children: one is 7 years old and performing exactly as you'd expect a 7-year-old to perform — their mental age matches their actual age. Now imagine a second child who is only 6 years old but performing at the level you'd expect from an 8-year-old — their mental age is ahead of their chronological age. It's a bit like the concept of a "reading age" — it tells you where someone is functioning relative to what's typical, not just how old they are.

In 1912, a German psychologist called William Stern took Binet's mental age idea and developed it into something even more useful: the Intelligence Quotient, or IQ. The formula is pretty simple — you take the mental age, divide it by the chronological age, and multiply by 100. So if both ages are the same, you get an IQ of 100. If your mental age is higher than your actual age, your IQ goes above 100. This single number became the standard way of talking about intelligence scores. Then, in 1916, Lewis Terman at Stanford University revised and expanded Binet's test into what became known as the Stanford-Binet Test. Terman added 40 new items, ranging from simple tasks for younger children (like copying a square or identifying which of two lines is longer) to more demanding tasks for older children (like mental arithmetic or arranging weights from heaviest to lightest). What made Terman's contribution particularly important — beyond just adding new questions — was that he tested this updated version on a large, standardised sample of 1,000 children. This matters a lot, because standardisation is what allows you to make meaningful comparisons between people. If everyone takes the same test under the same conditions, their scores actually mean something relative to each other. Terman also applied Stern's IQ formula to the Stanford-Binet scale, which is how the use of IQ scores became widespread.

Then along came Robert Yerkes in 1917, and he made a huge leap forward in terms of scale. During World War One, the American army wanted a way to classify its soldiers, and Yerkes developed two tests to do this: the Army Alpha for soldiers who could read, and the Army Beta for those who couldn't. In total, these tests were administered to an extraordinary 1.75 million people. The big advances here were practical ones — Yerkes introduced time-limited testing and group testing, meaning you didn't need a one-on-one session with a psychologist. This massively raised the public profile of IQ testing, and by 1919 there was even a National Intelligence Test being used more broadly. The takeaway from this whole historical section is that early intelligence research was largely driven by practical needs rather than theoretical curiosity — armies needed to classify soldiers, schools needed to identify struggling children — and along the way, researchers developed the key building blocks of intelligence testing: what to measure, who to measure, how to make it practical, and how to make scores comparable across people.

Now let's shift to what non-experts — ordinary people — think intelligence is. This is called implicit theory research, and it's interesting because it reveals how intelligence is understood in everyday life, not just in academic settings. Sternberg and colleagues in 1981 asked everyday people what they thought intelligence involved, and three main themes came up: practical problem solving, verbal ability, and social competence. But Sternberg also noted in 1985 that different studies produced different dimensions — and he listed things like goal orientation, contextual intelligence (learning from experience and understanding your environment), and fluid thought (thinking quickly). The point is that even among laypeople, there isn't one agreed-upon definition.

What's more, what counts as "intelligent" varies across cultures. In Western cultures, speed of mental processing tends to be strongly associated with intelligence — if you can think fast, you're seen as smart. But in Eastern cultures, the picture is different. Awareness of yourself in relation to society, an understanding of history, and even spirituality are associated with being intelligent. Yang and Sternberg in 1997 asked Taiwanese Chinese participants to define intelligence and found five themes: a general cognitive factor (things like practical problem solving and fast learning), interpersonal intelligence (relating well to others and showing empathy), intrapersonal intelligence (knowing yourself), intellectual self-assertion (confidence in your intellect), and intellectual self-effacement (interestingly, modesty about your intellect). This last one probably wouldn't appear in a Western definition.

Intelligence also seems to be understood differently depending on your age. Yussen and Kane in 1985 studied age group differences in children aged 11 to 16. Younger children tended to see knowledge as central to intelligence and tended to think of intelligence as something innate and one-dimensional — you either have it or you don't. Older children in the study had a more nuanced view, understanding that both nature and nurture play a role, and that there are different types of intelligence — academic, social, and physical.

Experts don't fare much better in agreeing on a definition, by the way. Back in 1921, the Journal of Educational Psychology asked 14 prominent psychologists what intelligence is, and got a range of answers — Thorndike said it was about managing social situations and making good decisions based on facts, Terman focused on abstract thinking, and Thurstone had yet another angle. Sternberg and Detterman repeated this exercise in 1986 with 24 experts, and Jensen did the same in 1998 with 38 experts — and got 38 different definitions. However, Sternberg in 2000 pulled out some common threads: adaptation to the environment, basic mental processes, and higher-order thinking like reasoning, problem solving, and decision making. There's also an interesting point here from Sternberg in 1985 about how experts from different fields define the "ideal intelligent person" differently — businesspeople, physicists, philosophers, and artists all highlight different qualities like logical thinking, the ability to follow complex arguments, precise mathematical thinking, or the ability to see analogies. So even the experts don't agree — which is part of what makes intelligence such a fascinating but tricky subject.

Now let's get into the actual academic theories of intelligence, because this is where things get really interesting. The big one to start with is Charles Spearman's theory of general intelligence, or g. Spearman, in 1904, tested children on a range of tasks — memory, sensory discrimination (like distinguishing between weights and sounds), vocabulary, maths, and spatial abilities. What he noticed was something called the positive manifold: children who did well on one test tended to do well on all the others. The tests all positively correlated with each other. To analyse this, Spearman used a statistical technique called factor analysis, which essentially groups together tests that correlate with one another. Imagine doing a factor analysis on people's music taste — you might find that people who like Chemical Brothers also like Röyksopp and Calvin Harris, so those cluster together as an "electronic" factor; separately, fans of 50 Cent and Snoop Dogg cluster as a "hip hop" factor, and so on. When Spearman applied this to intelligence tests, he found that verbal fluency, mathematical ability, flexible thinking, and critical thinking all loaded onto the same single factor. His conclusion was that there is one underlying general intelligence — g — which he described as a kind of "mental energy" that drives all cognitive performance. Alongside g, he also acknowledged specific abilities (called s), like vocabulary or mathematical skill, but argued that g was the common thread.

Not everyone agreed. Louis Thurstone in 1938 pushed back against Spearman's single g and argued that instead of one general intelligence, there are seven distinct primary mental abilities: associative memory (rote learning), number (carrying out maths), perceptual speed (picking up on visual details quickly), reasoning (both inductive and deductive), spatial ability (mentally transforming shapes), verbal comprehension (reading and understanding language), and word fluency (generating and using words). For Thurstone, g was a result of these abilities rather than the underlying cause of them.

Then Raymond Cattell, working around 1966, came along and argued that g isn't one thing either, but actually two related but distinct components. The first is crystallised intelligence (Gc) — this is your accumulated knowledge, your vocabulary, your factual knowledge, things you've learned over time. The second is fluid intelligence (Gf) — this is your raw reasoning ability, your capacity to solve new problems, identify patterns, and understand new information without relying on prior knowledge. Interestingly, research by Schaie in 1985 showed that the relationship between these two changes dynamically over your lifespan — fluid intelligence tends to peak earlier and decline with age, while crystallised intelligence can continue growing into old age.

Cattell's colleague John Horn expanded this framework further, concluding there wasn't just one g but actually nine broad cognitive abilities — this became known as the Cattell-Horn Gf-Gc model around 1985. These abilities covered things like fluid reasoning, cultural knowledge, quantitative (mathematical) knowledge, short-term memory, long-term memory, visual processing, auditory processing, processing speed, and decision speed.

Building on all of this, John Carroll in 1993 developed his three-stratum model, which is a hierarchical framework based on factor analysis of an enormous 461 datasets collected between 1927 and 1987. At the bottom (Stratum I) are 69 narrow cognitive abilities. These cluster up into Stratum II, which contains 8 broad factors including fluid intelligence, crystallised intelligence, visual perception, auditory perception, memory, retrieval ability, cognitive speed, and processing speed. And at the very top (Stratum III) sits general intelligence, g. It's a neat hierarchical picture of how specific abilities relate to broader ones, which in turn relate to general intelligence.

All three of these frameworks — Cattell's Gc/Gf, Horn's extended model, and Carroll's three-stratum hierarchy — were eventually combined in 1999 into what is now called the Cattell-Horn-Carroll (CHC) theory of cognitive abilities. This is currently one of the most influential theories in the field, particularly when it comes to developing IQ tests. The full CHC model includes up to 16 broad-stratum intelligences — including things like fluid and crystallised intelligence, visual-spatial ability, auditory ability, processing speed, memory, and even more niche ones like olfactory, tactile, and kinaesthetic abilities. Each of these broad intelligences then breaks down into narrower sub-abilities. For example, fluid intelligence (Gf) includes things like deductive reasoning and speed of reasoning, while crystallised intelligence (Gc) includes language development, lexical knowledge, general verbal information, and even foreign language proficiency. It's a comprehensive and dynamic theory that continues to evolve.

Moving beyond these cognitive ability theories, the lecture also covers two "alternative" approaches. The first is Howard Gardner's theory of multiple intelligences, introduced in 1983. Gardner was an educational psychologist who felt that traditional intelligence testing was too narrow — particularly that Western education systems were over-focused on logical-mathematical and linguistic intelligence, at the expense of other abilities. He identified seven intelligences originally: linguistic, logical-mathematical, and spatial (the more "conventional" ones), plus musical, bodily-kinaesthetic (using your body with skill), interpersonal (relating to others), and intrapersonal (understanding yourself). Later, in 1998, Gardner added two more: naturalist intelligence (the ability to interact with and understand the natural world — think biology or meteorology) and existentialist intelligence (the ability to understand your place in the grand scheme of things). Gardner's framework is particularly popular in educational settings because it validates forms of ability that traditional IQ tests don't capture.

The second alternative is emotional intelligence (EI), which is all about understanding and managing emotions — both your own and other people's. There are three main models covered here. The first is Salovey and Mayer's four-branch model from 1990. In their model, emotional intelligence involves four progressively more sophisticated abilities: perceiving emotions (recognising them in yourself and others), facilitating thought with emotions (using emotional information to make judgements and prioritise), understanding emotions and how they relate to each other, and finally managing emotions through detached evaluation and regulation. Each branch builds on the one before it, making this an "ability" model of EI — it treats emotional intelligence as a genuine cognitive skill, not a personality trait.

The second model is Goleman's, from 1995 and expanded in 2001 and 2002. Goleman's view is that emotional intelligence is fundamentally about learning to manage our basic emotional responses — like the fight-or-flight instinct. His is a mixed model, meaning it blends emotional intelligence with personality and behavioural traits like conscientiousness and adaptability. He outlines four emotional competencies: self-awareness (knowing your own emotional states), self-management (regulating your own emotions), social awareness (understanding and influencing others' emotions), and social skills or relationship management (maintaining good interpersonal relationships). In total, these are built from 25 different underlying abilities. Goleman's model is particularly popular in workplace and leadership contexts.

The third model is Bar-On's emotional-social intelligence model, introduced in 1997 and updated in 2005. Bar-On's view is that emotionally and socially intelligent behaviour is adaptive — in other words, it helps us survive and thrive in our environments. Like Goleman, his is a mixed model. He organises emotional and social competencies into five major domains and 15 aspects. The domains are: intrapersonal skills (recognising your own emotions — things like self-awareness, assertiveness, and self-regard), interpersonal skills (understanding others' emotions — empathy, social responsibility), adaptability (managing emotions — problem solving, flexibility, reality testing), stress management (managing change — stress tolerance and impulse control), and general mood (generating positive affect — happiness and optimism). The key takeaway is that for Bar-On, emotional expression is not a weakness but a core part of being well-adapted to life.

When comparing the Salovey-Mayer model with the Goleman and Bar-On models, the big difference is that Salovey and Mayer focus purely on emotional intelligence as a distinct, measurable ability. This is useful because it lets researchers cleanly study how EI relates to other things like mental health or school performance. Goleman and Bar-On's mixed models, by contrast, blur the lines between intelligence and personality — which makes them harder to study scientifically, but arguably more useful in applied settings like the workplace, where understanding the full range of emotionally intelligent behaviours (not just the core ability) is important. This blurring is actually one of the main criticisms of EI theory — Eysenck in 2000 argued that mixing intelligence and personality factors together is "unscientific." There is also a broader critique that there is no clear benchmark for establishing whether emotional intelligence scores are actually meaningful.

To wrap up, the big messages from this lecture are that there is no single agreed-upon definition of intelligence — it means different things to different cultures, different age groups, and different academic disciplines. The major formal theories broadly agree that there is some form of general intelligence (Spearman's g), but they disagree on whether that single g is the whole story, or whether intelligence is better understood as a collection of multiple, related abilities. And then there are alternative frameworks like Gardner's multiple intelligences and the various models of emotional intelligence that challenge the traditional focus on cognitive ability altogether. It's a rich and genuinely contested field — and that's what makes it worth studying.