What is Intelligence? The Three Main Theories of Intelligence – Two Good, One Bad
December 30, 2017
When people talk about a person’s ‘intelligence’ it is not generally clear what underlying ability or abilities this term refers to. This article is intended to clarify in layman’s terms what psychologists and brain scientists can mean by intelligence. Basically, there are two good theories–and scientists are divided on which is the best theory–and one bad one which all scientists I know of reject. A good theory is one that is supported by the evidence; a bad theory is one that is not.
Official IQ tests such as the WAIS-IV claim to measure individual differences in an underlying ‘ level of cognitive ability given by a single number–your IQ or intelligence quotient. But is it true that there is a single underlying mental ability that we differ in and that explains what makes us different in our cognitive abilities? If someone is good at maths, are they also likely to be good at language comprehension, reasoning, thinking analogically, learning languages and general knowledge, due to their underlying ‘intelligence level’, as this theory implies?
Or are there ‘multiple intelligences’ underlying out abilities–perhaps dozens or even hundreds of them–each independent from each other, and measured by different types of test. If you have an ability in mathematics, is this ability completely unrelated to your ability in learning languages or play general knowledge games like trivial pursuit? If this is the case, is the idea of having a single IQ score quite meaningless? Or alternatively, are there a small number of underlying cognitive abilities (perhaps two or three) that we differ in, which are relatively independent from each other–and which together explain most the differences in our cognitive abilities?
1. The theory of general intelligence (g)–a good theory
A long standing an influential theory for our cognitive abilities states that underlying all our cognitive abilities (math, language comprehension, general knowledge) is a single factor–called general intelligence (also known as unitary intelligence, general cognitive ability, or simply ‘g’ ) that individuals differ on and which explains those differences.
Spearman (1923) proposed that underlying all cognitive abilities a ‘general ability’ factor (g) that all the abilities draw on. Individuals differ in g according to a bell curve distribution on this theory. g can be thought of in terms of information processing power. Some people –those with higher g–can process more information, more efficiently than others. Using a computer analogy, they have more RAM. The more RAM a computer has, the more complex and information-intensive the programs that can be run on it. If you have an IQ of 160 like Quentin Tarantino has, you have lots of RAM, large ‘bandwidth’ for processing information. If you have an IQ of 78 like Muhammad Ali as a young man (whose IQ was measured by the army), then you have less RAM. Muhammad Ali had many talents, but according to the unitary intelligence theory, intelligence wasn’t one of them.
The evidence for this theory is the same evidence that allows us to reject the theory of multiple intelligences. All standardized tests of cognitive ability (and there are dozens of them, measuring a wide range of different abilities) are positively correlated–not perfectly, but to a large degree. This means that if someone scores higher than average on one of those tests, they are likely to score higher than average on all the other tests–even ones that appear totally unrelated. Scoring higher in an arithmetic test means you will probably also score higher in a vocabulary test. This remains true, even when you take other factors like educational background, or family socioeconomic status into account. This is compelling evidence that there is a single underlying level of cognitive ability that is applied to each of the tests and that performance on one test is not independent from performance on another as the multiple intelligence theory claims.
Spearman (1904)–the psychologist who first proposed the g theory–argued that the variance (the person to person variation) of performance between individuals on ANY cognitive task can be attributed to just two underlying factors: g (general intelligence) and s –the skill unique to that particular task. A person could invest relatively more time into developing a specific skill such as arithmetic, and this will raise their score on an arithmetic test relative to another test such as vocabulary that they didn’t train or practice on, but their general intelligence g will still account for most of their performance on the arithmetic test. G is still the most important factor in explaining levels of performance, whatever the test.
2. The theory of multiple intelligences-a bad theory Spearman’s ‘g’ theory is the opposite of the theory of multiple intelligences. The theory of multiple intelligence is an appealing one because it gives some room for everyone to have their own unique strengths in ‘intelligence’. But as we have seen it turns out that our cognitive strengths and weaknesses are best explained by how much time and effort is we have invested into particular skills or types of knowledge. If I take up a technical trade and become good at it, and find that I am struggling with reading fiction, this doesn’t necessarily mean that I have a special ‘intelligence’ for technical thinking and have no ability for reading or language. The fact I struggle with fiction is better explained by the fact that I have invested my intelligence into building up this particular type of expertise and thus see more of a return on that investment in technical modes of cognition. If I had spent as much time reading fiction as I have applying myself to technical problems, chances are I’d be good at that.
3. The theory of fluid intelligence (gF) and crystallized intelligence (gC)–another good theory
This theory builds on the general intelligence theory, and was originally proposed by the psychologist Raymond Cattell back in 1943. It holds that g is meaningful–that we each have a different general intelligence level– but contributing to g are two different types of intelligence: fluid intelligence (gF) and crystallized intelligence (gC ). Fluid g is the ability to reason and problem solve with novel tasks or in unfamiliar contexts (measured reasoning tasks), while crystallized g is defined as acquired knowledge and is measured using tests of general knowledge, mathematics, and vocabulary. This dual way of understanding intelligence allows for knowledge that you have built up in particular areas to compensate for limitations in overall reasoning and problem solving ability– our ‘raw intelligence’. You may succeed due to knowledge about a task or domain (crystallized g), or due to sheer mental ‘horsepower’ (fluid g).
Where the idea of ‘multiple intelligences’ makes sense: as crystallized intelligence that we invest in
Our crystallized intelligence allows for ‘multiple intelligences’. You could have a high level of crystallized intelligence in graphic design, for example, while having only an average level of fluid intelligence. But you will only be able to use your crystallized intelligence for graphic design in situations in which you are familiar and have built up expertise. Unless you have a high level of fluid intelligence when you are confronted with an unfamiliar problem in graphic design–something ‘out of context’, requiring some difficult figuring out-then you are likely to have difficulties. On the flip side, if you have a high level of fluid intelligence, it will take you less time to pick up graphic design (or whatever) skills as you learn your basic skill set. Your learning will be more efficient, and you will find it easier. In general, the more fluid intelligence you have the more you will be able to ‘invest’ it into crystallized intelligence skills and knowledge–the more ‘multiple intelligences’ you will be able to develop if you so wish. In the context of work, the more gF you have the more quickly and efficiently you can be trained. One study showed that it took people in the 110 to 130 IQ range about 1 to 2 years to catch up with the super-charged performance of those with IQs of 130+ who had only 3 months’ experience on the job.
Looking at all the evidence, both the general intelligence (g) theory, and the fluid intelligence (gF) and crystallized intelligence (gC) are well supported and useful in explaining how we differ in our cognitive abilities. In my view, the fluid and crystallized theory is the more insightful and useful. It helps me understand intelligence-and how we can improve it-better. For instance, research shows that you can do a specific type of ‘working memory’ brain training to increase your fluid intelligence level substantially–but this training does not directly affect your crystallized intelligence.