The following two paragraphs are from the article: "Price and her colleagues used brain scans to confirm that these big fluctuations in performance were not ra

The following two paragraphs are from the article: "Price and her colleagues used brain scans to confirm that these big fluctuations in performance were not random — or just a fluke. They evaluated the structure of the teens' brain in the early teen years and again in the late teenage years. "We were able to see that the degree to which their IQ had changed was proportional to the degree to which different parts of their brain had changed," explains Price. For instance, an increase in verbal IQ score correlated with a structural change in the left motor cortex of the brain that is activated when we speak." So, if there were structural changes in the brain that correlated with change in IQ score, then IQ likely tells us something other than "this person does this well or badly at IQ tests." Many people are very skeptical about IQ tests. As depressing as it is, the basic science behind psychometrics has been mostly stable for decades. Look up the work of Linda Gottfredson (sp?). As for this particular study, psychometricians already knew that generalized intelligence (the so-called g-factor) crystallized in adolescence or early adulthood. The surprising thing about this study is how much change some of the subjects showed. Going from the fiftieth percentile to the ninety-something or, worse, going in reverse, is a big deal. The researchers behind this study should get more funding to continue their work.