February 26, 2012

Intelligence genes more elusive than Higgins' boson

Havard University has issued a press release where Christopher F. Chabris, author of a yet unpublished paper on the matter, ponders the elusiveness of genes that could define intelligence. 

Twin studies have suggested that there is at least some truth to an association of intelligence (measured by IQ) with inherited genes, however now it seems clear that no individual gene is likely responsible of any notable influence on expressed intelligence. 

Chabris ponders that the genetic influence is probably the work of many genes acting collectively and not any single one of them, and also of the interactions of genes and environment.

What our results show is that the way researchers have been looking for genes that may be related to intelligence — the candidate gene method — is fairly likely to result in false positives, so other methods should be used.

Update (Oct 18): the paper is available here (PDF).


  1. The missing heritability issue with high IQ really is quite fascinating. Efforts to find genetic causes of mental retardation have been much more successful, suggesting perhaps that perhaps looking for an absence of stupid genes that pushes up the Bell Curve for the remaining black box of genetic IQ sources, rather than an excess of smart genes might be a more fruitful model to pinpoint at least some meaningful part of the heritability.

    I've also seem some pretty convincing heuristic arguments that the genetics of major personality traits are more likely to have a few big genes behind them than IQ, so looking for those genes may be a more fruitful enterprise.

  2. "... absence of stupid genes"...

    I like the idea. Maybe it's wrong after all but it's the kind of thinking outside the box that often works.

    However I tend to agree with the tone of Chabri in the sense that it is probably not any single gene or group of very influential genes but the combined action of many different genes (plus, of course, the environment).

    One possibility that has not been considered here is that epigenetic changes can be inherited. If intelligence (or other traits) are largely epigenetic but inherited, it'd be difficult to discern, specially as we don't make much IQ research with fruit flies... Maybe it's about time to start studying long-range inheritance of intelligence in octopuses?

  3. There is something to be said for the epigenetic model, but the twin studies are such a decent fit for plain old additive shared genetic sources that I'm a bit doubtful that a more complicated model is really needed. The biggest arguments for epigenetics are the known exceptionally high substance abuse rates where epigenetic factors are well documented, and the fact that environmental effects are a lot stronger among the poor than the middle class (although genetics as a "peak potential IQ" could also fit that data). The Flynn effect is also a pretty good fit to an epigenetic model, since the Flynn effect time lines are pretty similar to the time that it takes an epigenetic signal to work its way out of a lineage.

    Dominant retardation genes are predominantly first generation mutations at varied loci and a lot of them have been identified. IQ variation in the normal range, in contrast, is pretty fine grained, but it takes remarkably few genes in a polygenetic trait to get that result. I'm inclined to think that factors like overall mutational load, as opposed to SNPs, for example, may be fruitful to look at for associations - schitzophrenia and bipolar searches have found associations there, however and you'd think that if there was also an IQ link to that it would show up in the data.

  4. Twin studies would not be able to discern inherited epigenetics, something that was discovered in the last years and that I imagine that most geneticists do not yet really get to accept in full, because of semiconscious resistances.

    Inheritable epigenetics are a game changer but also very real.

    "Dominant retardation genes are predominantly first generation mutations"...

    But a good deal of mental issues have epigenetic causes (or rather environmental ones effected by epigenetics, see also this). Not only inheritable epigenetics challenge the idea of twin studies being able to detect genetic causes on their own but environmental epigenetics will be almost the same for twins in pregnancy and, normally, for much of their early life.

    Identical twins differ more and more with age and that is probably because they are accumulating their own distinct epigenetic variation caused by the environment. But at birth and later they are very similar and not all is attributable to genes alone.

    It's a tricky issue that requires of new approaches.

    See also:



  5. Clarification:

    When I said: "Twin studies would not be able to discern inherited epigenetics"...

    I meant that: "Twin studies would not be able to discern inherited epigenetics"... from inherited classical genetics. The inheritance factors would be detected but whether they are classical genes (complex or simple) or inherited "epigenes" or a mixture of both, it'd be impossible to know.

    On the other hand, when I, further down, say that novel environmental epigenetics could also be mistaken by genes, I'm probably wrong because twin studies should be able to pick that up at least to a great degree.

    But the issue of inherited epigenetics stands as a serious issue.

  6. This comment has been removed by the author.

  7. This comment has been removed by the author.

  8. Updated with link to the paper. Also I will comment now (separately) on a new paper on evidence of epigenetics in IQ differences in identical twins.


Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).