The research - published in Science magazine - found that a part of the brain called the frontal lateral cortex was the only area where blood flow increased when volunteers tackled complicated puzzles involving sequences of symbols and letters.
Their findings seem to support the 1904 theory of psychologist Charles Spearman, who argued that people used a particular part of the brain when performing complex tasks.
Other theories contend that intelligent thinking requires various portions of the brain working together like different parts of an engine.
The researchers, from the Medical Research Council's Cognition and Brain Sciences Unit at Cambridge University and from Dusseldorf University, used scanning techniques to assess the blood flow in subjects' brains when they were performing the tests.
Science magazine also carries an article attacking the scientists' approach to the subject. It argues that human intelligence cannot be defined in such a specific way.
'G-factor'
The debate over the nature of intelligence has continued since Charles Spearman set his mind to the problem early in the last century.
He proposed that the same bright people tended to do well at different tasks because they employed a generally useful component of their brains which he called the "g" factor.
However, many questions about the mystery of intelligence remain unanswered.
The research, led by John Duncan from the MRC, asked volunteers to carry out a number of "high-g" and "low-g" tasks which involved identifying mismatched symbols or sets of letters.
High-g tests were designed to employ a wider range of cognitive functions and be more challenging than low-g tests.
Brain scans
While taking the tests, the volunteers' brains were scanned using a technique called positron emission tomography.
The scientists found that high-g tasks did not require numerous regions spread around the brain to be used together.
Instead, activity was concentrated in the lateral frontal cortex, in one or both brain hemispheres.
The researchers wrote: "The results suggest that "general intelligence" derives from a specific frontal system important in the control of diverse forms of behaviour."
But they acknowledged that this apparent intelligence centre might itself be divided into finer components.