He presented his findings at a seminar in Cambridge. He showed that year-on-year rises in IQ scores - known as the "Flynn Effect" - were down to improvements in abstract thinking rather than increases in overall intelligence.
"When we look at those older than us, they are not any less intelligent.
But we may see they are a little more literal and rule-bound, more utilitarian, that they look more at the concrete and less at the abstract and hypothetical," Professor Flynn told the audience at Cambridge Assessment's Psychometrics Centre.
The promotion of the national curriculum, the expansion of schooling, the increasing importance of science and logic in Western society, and the popularity of demanding leisure pursuits were all influences, he said.
"Some people think you can see this in TV," he said. "Thanks to the rise in mental ability, producers are having to provide more complex plot lines.
"Years ago you had I Love Lucy, which was a few characters and pure escapism. Now 24 has 24 characters and 10 sub-plots. Shows are rising to a new level of expectation."
But vocabulary had not significantly improved because children lived in a more visual culture, he said. Average IQs have increased by roughly half a point every year in the industrialised world since testing began. His findings were based on more than 100 years of IQ tests, challenging candidates to spot patterns and groupings.
Posing the question "What is the similarity between a dog and rabbit?"
almost a century ago, researchers were more likely to receive literal answers such as "There aren't any" or "Dogs eat rabbits".
But children today were more likely to give what was considered the best answer, which is "They are both mammals."
Professor Flynn said IQ gains were likely to level off as affluence and family size stabilised. But they would probably be replaced by an improvement in different forms of reasoning.
"It's not so much a rise in intelligence as an investment of intelligence in new problems," he said.
His findings will be published in a book, What is intelligence beyond the Flynn effect?, to be published in July.