IQ tests 'do not reflect intelligence'

IQ tests are misleading because they do not accurately reflect intelligence, according to a study which found that a minimum of three different exams are needed to measure someone's brainpower.

Human skull and brain
Are we more intelligent than our grandparents were? Credit: Photo: ALAMY

For more than a century our intelligence quotient (IQ) has been used to measure how clever people are and Mensa, the society for the intellectual elite, has even used the test to weed out sub-par applicants.

But now the scale has been dismissed as a "myth" by scientists who found that our intelligence can only be predicted by combining results from at least three tests of our mental agility.

Different circuits within the brain are used for different thought processes, the researchers showed, meaning separate tests of short-term memory, reasoning and verbal skills are needed to measure someone's overall intelligence.

Their landmark study was based on the results of an online intelligence test which was launched by the Daily Telegraph and New Scientist two years ago, and attracted more than 110,000 responses.

Dr Roger Highfield, the Telegraph columnist and one of the authors of the paper, said: "When you come to the most complex known object, the human brain, the idea that there is only one measure of intelligence had to be wrong.

"We can all think of people that have poor reasoning and brilliant memories, or fantastic language skills but aren't so hot at reasoning, and so on. Now once and for all we can say there is not a single measure such as IQ which captures all the intelligence that you see in people."

The online test, which took about 30 minutes to complete, featured 12 cognitive tests of volunteers' memory, reasoning, attention and planning as well as recording details about their lifestyle and background.

Taking into account the full range of cognitive abilities tested, they found that people's varying success rates could only be explained by combining at least three types of intelligence, and not by any single measure such as IQ.

"When you look at cognitive ability you can't boil it down to fewer than three components – short-term memory, reasoning and a verbal component," Dr Highfield explained. "There isn't one component that explains all the variations we saw in all the tests."

Following up their findings, the scientists scanned the brains of 16 volunteers while they completed the same tests and found that the three key types of intelligence relied on different circuits within the brain.

Writing in the Neuron journal, the researchers also observed that regularly playing "brain training" games appeared to have no effect on people's overall performance.

But people who regularly played computer games scored significantly higher in reasoning and short-term memory tests, while smokers and anxiety sufferers had weaker short-term memory scores.

An updated version of the test has been released as part of the ongoing research project.