Samstag, 4. April 2015

Does the Regression to the Mean in IQ reflect a Regression to the Mean in g?

According to Cochran, regression to the mean in IQ isn't just a purely statistical phenomenon, but also a real phenomenon: The children of people with high/extreme IQ-scores usually get, on average, lower/less extreme IQ-scores. (To 'exclude' the possibility that the regression to the mean is just caused by measurement errors, both parents and their children should be tested several times.)
So, if the regression to the mean in IQ doesn't reflect a regression to the mean in g, the following should be found: The less g-loaded an item of the IQ-test, the more regression to the mean; the more g-loaded an item of the IQ-test, the less regression to the mean.


  1. Sounds about right. Look forward to this finding.

  2. Parents of very high IQ may have a propensity to have children later and as they already have increased mutational load that increases their intelligences (and not necessarily want to display reduced fitness or health), this excess of mutations caused by advancing age and the higher mutational load (than compared to people of average intelligence) can increase the chances of the children do not inherit the complex combination that makes your smarter parents.

    Another possibility, in addition to this I set up, is that a great technical and quantitative intelligence, may be more epigenetic in nature or an event so rare and that is not being selected, the more likely it is that children do not inherit the talents of parents. But there is a multitude of cases and for the moment, we can say that the average regression is not a universal phenomenon. I hope you are reconsidering this possibilities.

    Another problem is that people too (technically) intelligent tend to have few children. How to know the real heredity if most of them do not have the optimal number of children ***

    Just a group of couples who marked very high in IQ tests (and choose the phenotypes, only the very high verbal IQ ** ** or math skills or 'g' factor '** ....), a similar racial background and had many children at a young age or ideal, so we can have an even more accurate judgment of this phenomenon.

    My immediate family is a case where there was a real inheritance of intelligence as my parents and my brother are above average in intelligence (and my father family, a lot of technical smarts). What is interesting to think is that families where only men are born, the effect testosterone or even randomization vulnerabilities that can bring, can have an effect, especially if it is true that excess testosterone (necessarily exposure to testosterone, not to be excessive, if there is already a vulnerability to intrauterine mutation, if it really happens) can alter brain morphology.

    I have many doubts about the supposed epigenetic effects and believe that the combination of fraternal and maternal genes in the design, they can lead to uterine sensitivities. I also can not understand how cigarette smoke could affect the uterus, whether it is just or especially the lungs receiving the load of toxins that are present in the fuc*ed cigarette.

    Women who smoke during pregnancy may be more likely to

    have some kind of mental disorder

    and either

    they could be less intelligent or conscientious.

    It makes sense that bad habits can affect the child who is in the belly. But the uterus may have some defense mechanism against external agents. Finally, when only seeing simulation (or actual), toxins affecting the development of the fetus can we finally tell if this can actually happen for most cases.