<P> Duration of average schooling has increased steadily . One problem with this explanation is that if in the US comparing older and more recent subjects with similar educational levels, then the IQ gains appear almost undiminished in each such group considered individually . </P> <P> Many studies find that children who do not attend school score lower on the tests than their regularly attending peers . During the 1960s, when some Virginia counties closed their public schools to avoid racial integration, compensatory private schooling was available only for Caucasian children . On average, the scores of African - American children who did not receive formal education during that period decreased at a rate of about six IQ points per year . </P> <P> Another explanation is an increased familiarity of the general population with tests and testing . For example, children who take the very same IQ test a second time usually gain five or six points . However, this seems to set an upper limit on the effects of test sophistication . One problem with this explanation and others related to schooling is that in the US, the groups with greater test familiarity show smaller IQ increases . </P> <P> Early intervention programs have shown mixed results . Some preschool (ages 3--4) intervention programs like "Head Start" do not produce lasting changes of IQ, although they may confer other benefits . The "Abecedarian Early Intervention Project", an all - day program that provided various forms of environmental enrichment to children from infancy onward, showed IQ gains that did not diminish over time . The IQ difference between the groups, although only five points, was still present at age 12 . Not all such projects have been successful . Also, such IQ gains can diminish until age 18 . </P>

Which of the following has been proposed as a possible explanation for the flynn effect