Picture a Norwegian conscript at 18 or 19, working through timed questions on arithmetic, word similarities and figures. Hundreds of thousands of young men sat that same test across decades. Their scores now tell one of the odd stories in the history of intelligence testing: a century of climbing, then a turn.
A test built for schoolchildren
It began in Paris. In 1904 the French Ministry of Public Instruction asked psychologist Alfred Binet to find out which pupils in the city’s public schools would struggle with the standard curriculum, early enough for them to get help. He and Theodore Simon dropped the older sensory approach and published the Binet-Simon Scale in 1905. Its questions went straight at judgment, reasoning, comprehension and word knowledge, and it introduced the idea of mental age. A child who matched the typical 8-year-old got a mental age of 8, whatever the birth certificate said.
The famous number came later. Lewis Terman at Stanford revised the scale in 1916 and adopted a formula from the German psychologist William Stern, who had proposed it in 1912. IQ meant mental age divided by chronological age, times 100. A ten-year-old who performed like a twelve-year-old therefore scored 120. In 1939 David Wechsler of Bellevue Hospital in New York dropped that ratio. His deviation IQ compared each person only with others of the same age, mapped onto a bell curve centred on 100 with a spread of 15 points.
Three points a decade
Then came the puzzle James Flynn documented across dozens of countries in the 1980s. Across most of the 20th century, raw scores climbed by about three points every ten years. A formal meta-analysis covering 1909 to 2013 pooled 271 separate samples, close to four million people in 31 countries, and found gains that varied by type of ability and faded in more recent decades. Explanations on offer include better nutrition, longer schooling and smaller families.
Norway turns the curve
Norway’s conscription records show the turn clearly. Scandinavia Standard describes a 2018 study by Bernt Bratsberg and Ole Rogeberg. Test results climbed for Norwegian men born in the 1960s, topped out for the 1975 cohort, then slid for later births. The average converted IQ was 99.5 for men born in 1962, 102.3 for 1975, 99.4 for 1989 and 99.7 for 1991. Comparing brothers within the same families rebuilt the whole curve, which points to the environment rather than a shift in genes.
A more recent study, abstracted in PubMed, followed 579,379 Norwegian men from 25 birth cohorts between 1967 and 1991. The decline appeared earliest in the sons of well-paid fathers. Families lower on the income ladder peaked later and dropped less, a pattern the researchers linked to rising education levels. After the growth in schooling leveled off, scores fell in every group, steepest at the top.
The authors raise upper-secondary completion as a possible lever for lifting performance. They also caution that pressures on test performance may have started earlier than anyone realized.
Sources
- Educational expansion and the hidden reversal of the Flynn effect-Differential trends across socioeconomic strata – PubMed
- IQ scores rose by roughly three points every decade throughout the 20th century, then reversed – Scandinavia Standard
- The History of Intelligence Testing – IQ Metrics
