Intelligence: How Psychology Has Defined and Measured It

What psychology means by intelligence

The Noba Project defines intelligence as the set of abilities a person uses to learn new things, remember them, apply them to solve problems, and adapt to situations never encountered before. It’s a deliberately broad definition, because under that single label the field has built very different theories of what intelligence actually is.

Readers coming from psychological science will find a useful anchor here: every theory covered in this Recap rests on a testable method, not on an opinion about one’s own way of thinking.

Spearman’s g factor

In 1904, the English psychologist Charles Spearman spotted a recurring pattern in his data: strong performance in one intellectual area, verbal ability, say, tended to come with strong performance in far-removed areas too, from logic to reasoning. From that correlation he drew the idea of a general factor, which he called g: a single underlying ability running through every intellectual task. Alongside g, in Spearman’s two-factor model, sits s, specific to each individual task.

Within academic psychology, the g-factor theory remains, in 2026, the dominant view of intelligence. It underlies more than 70 IQ tests in current use, including the Stanford-Binet (fifth edition, 2003) and the Wechsler Adult Intelligence Scale (fourth edition, 2008).

Practical example: a student who scores well on a language exam and, around the same time, also does well on a logical-reasoning test unrelated to reading shows the kind of pattern that led Spearman to propose the g factor: separate areas moving together.

From Binet to modern IQ

The intelligence test was not built to measure some abstract trait; it started as a practical assignment. In the early 1900s, the French government asked the psychologist Alfred Binet to identify pupils who needed extra classroom support. Working with his colleague Théodore Simon, Binet introduced the concept of mental age: how a child’s intellectual performance compares with the average of same-age peers — a different idea from the one Jean Piaget would describe years later, with cognitive development unfolding in stages.

From that first test, the Stanford professor Lewis Terman adapted the format into the Stanford-Binet, published in 1916; later the psychologist David Wechsler, critical of how heavily the Stanford-Binet leaned on verbal ability and of reducing intelligence to one single score, proposed an alternative measure.

The way IQ gets calculated has changed along with the tests.

Gardner’s multiple intelligences, still an open question

Between the late 1970s and early 1980s, the psychologist Howard Gardner put forward his theory of multiple intelligences: eight relatively independent intelligences, linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, naturalistic, interpersonal, and intrapersonal. Gardner had identified seven at first; he added naturalistic intelligence in the mid-1990s.

Gardner himself later ruled out other candidates, including a proposed existential intelligence and a moral or digital intelligence, because they didn’t meet his own criteria.

The theory has spread widely outside academia, in schools and popular science writing. Within academic psychology, its standing is still unsettled: researchers such as Waterhouse (2006) and White (2006) argue the theory lacks solid empirical research to back it up. A 2006 study by Visser and colleagues found a strong correlation between Gardner’s more reasoning-based intelligences and Spearman’s general g factor, a result the authors themselves do not treat as confirming the theory’s stronger claims. A direct experimental test of Gardner’s theory is difficult to design, which is part of why academic psychology has never granted it the same standing as the g factor.

Readers coming from cognitivism, which treats the mind as a system that processes information, will recognize in this comparison — one single factor against eight separate abilities — one of the most contested questions in the whole psychology of intelligence.

The limits of intelligence testing

No test, historical or modern, measures intelligence outside a context. These tests were built by Western psychologists and reflect Western-centric values, a limitation noted for several instruments still in use.

Negative stereotypes, meanwhile, can raise anxiety in the person being tested and lower their score: a phenomenon known as stereotype threat.

There’s also an effect tied to time rather than to any one person: average IQ test scores have climbed generation after generation, to the point that younger test-takers, given an older test, tend to score higher than the original sample that test was calibrated on. The psychologist James Flynn described this pattern in 1987, and it now carries his name: the Flynn effect.

In 1966, the psychologists Horn and Cattell distinguished two further forms of intelligence: fluid, the ability to solve problems on the spot, and crystallized, the ability to draw on accumulated language, skills, and experience. It’s a different split from the one between the g factor and multiple intelligences, but it helps show how much more layered psychology considers intelligence to be than any single number.