TOEFL Reading練習 #030 — A Century of Rising IQ Scores, Still Unexplained(日本語解説)

A Century of Rising IQ Scores, Still Unexplained

¶1 For most of the twentieth century, the average score on standardized intelligence tests rose steadily across dozens of industrialized countries, climbing at a pace far too fast to reflect any real change in human genetics. Two broad explanations have competed to account for the rise, one centered on the body and one centered on the surrounding culture, and each predicts a different pattern in which kinds of test items should improve fastest. Comparing those patterns has narrowed the debate without closing it. In the past three decades, a reversal of the very trend both explanations were built to explain has kept the question open rather than resolved.

¶2 The trend first drew serious scientific attention in 1984, when the political philosopher James Flynn combed through decades of American intelligence test records. Test publishers periodically release a new version of an exam and calibrate it so the average person taking the new version scores exactly 100, a practice that hides any long-term rise unless someone checks the older numbers underneath it. Flynn did exactly that. He gathered results from more than seventy studies that had used an older version of a test, then compared those raw scores with the raw scores of people who had taken the same older version decades earlier. The pattern was unmistakable: between 1932 and 1978 alone, Americans' raw scores climbed by nearly fourteen points, a gain of roughly three points every ten years. Researchers soon confirmed the same climb in country after country, and the pattern came to be known as the Flynn effect.

¶3 For years, the explanation favored by most researchers pointed to the body rather than the mind. Rising nutrition and public health, the argument went, had allowed each generation's brain to develop closer to its full potential, since reduced exposure to childhood infection and better prenatal care protect the nervous system during the years it develops fastest. Cross-national comparisons offered some support: in several countries, the sharpest gains in test scores occurred in the same decades that childhood stunting, impaired growth caused by chronic malnutrition, and infant mortality fell fastest. If a healthier, better-fed body was doing the work, though, the improvement should have shown up fairly evenly across every kind of question an intelligence test asks.

¶4 That is not what the test records show. When researchers broke the gains down by question type, they found far larger increases on tests of fluid intelligence, the ability to solve new and unfamiliar problems by spotting patterns without relying on stored knowledge. Increases were far smaller on tests of crystallized intelligence, the vocabulary and factual knowledge built up through schooling and everyday experience. On Raven's Progressive Matrices, a nonverbal test that asks a person to complete a sequence of abstract shapes, scores rose by roughly seven points a decade, more than double the pace of gains on vocabulary and general-knowledge questions. Flynn himself proposed a rival explanation for this uneven pattern: modern life, he argued, increasingly rewards the kind of abstract, hypothetical reasoning that puzzles like Raven's matrices demand, regardless of how much any of it improves a person's underlying health. A gap this lopsided is hard to square with a purely biological account, since a better-nourished brain should help with vocabulary questions just as much as it helps with pattern puzzles.

¶5 Neither explanation, however, anticipated what the trend would do next. Beginning in the 1990s, average scores in Norway and several other countries stopped rising and began to fall, a pattern now called the reverse Flynn effect. In 2018, the economists Bernt Bratsberg and Ole Rogeberg examined military conscription records covering more than seven hundred thousand Norwegian men born between 1962 and 1991. A simple comparison across birth years could not, by itself, tell a genetic cause from an environmental one. [A] So instead of comparing the population as a whole, they compared brothers from the same family born a few years apart, a design that holds shared genes and much of a family's environment constant. [B] Any remaining difference between an older and a younger brother's score, in other words, had to come from something that changed between one birth year and the next, not from which families were having children. [C] Younger brothers still scored measurably lower than their older siblings, even though the two shared both their genes and their childhood household. [D] Which factor, though, remains unknown. More than four decades after Flynn first documented the rise, researchers have eliminated some of the least likely explanations for both the climb in scores and its later reversal, but they have not converged on a single account that explains either one.