ps418

"Complex and emergent" with respect to gene x gene interactions (epistasis)? That is almost certainly correct with respect to musical or mathematical ability (but not general cognitive ability), and for some personality traits. For example, the genetic variance component for subjective well-being is largely epistatic, because while MZs correlate strongly whether reared together or apart, DZs do not correlate signficantly whether reared together or apart (Lykken and Tellegren, 1996).

General cognitive ability does have very signficant narrow-sense heritability, and therefore a significant additive genetic variance component. There is every reason to believe, therefore, that there are many loci making consistent independent contributions to variance in cognitive ability, so-called quantitative trait loci or QTLs.

I take you to be saying, and correct me if I have misunderstood you, is that there is a large epistatic variance component for cognitive ability, but a trivial additive variance component, which is incorrect. Even Devlin et al's (1997) critical meta-analysis deduced broad and narrow heritabilities of 'IQ' of approximately 0.5 and 0.34, respectively.

You specified traits whose expression is controlled by "a few genes." This is a good point, for practical reasons: in order to genetically engineer a quantitaitve trait like general cognitive ability, you'd have to first identify a large number of QTLs. There is reason to belive that in this case there are a large number of QTLs, each explaining only a very small portion of the population variance (less than a few percent). As research with experimental organisms has shown, it is incredibly difficult to pin down QTLs with such small effects. Further, even if you could pin down enough QTLs, you'd have an incredibly difficult time trying to bring them all together into one embryo. The technological barriers to such gene by gene manipulation will probably be prohibitive for decades to come, at least. On this, I basically agree with Pinker.

Again, I'm not sure what you mean exactly. The noticable differences between MZAs and MZTs do not extend to cognitive abilility or most personality dimensions. For instance, there are no significant group differences in MZA and MZT correlations for tests of cognitive ability (Bouchard et al., 1990), when both groups are tested as adults, and MZTs are only slightly more similar with respect to Big Five personality dimensions, or MMPI or MPQ or TPQ personality assessment tools than are MZAs. That's not to say that there are no differences, becasue there are: its just that the shared rearing environment is not a terribly significant source of variance for these two domains (see Bouchard and McGue, 2003 for a review of recent research).

I certainly agree that cloned embryos would allow more refined testing of prenatal environmental contributions to twin similarity. However, there are several other ways to examine the contributions of such prenatal environmental influences to trait similarities without using cloned embryos, and these suggest that the shared prenatal environment does not account for much of the similarity of twins with respect to general cognitive ability or most personality measures. From a previous thread discussing this issue:

Bouchard, at al., 1990. Source of human psychological differences: The Minnesota study of twins reared apart. Science 250, pp. 223 - 228.

Bouchard, T.J., and McGue, M., 2003. Genetic and environmental influences on human psychological differences.
Journal of Neurobiology 54, pp. 4-45.

Devlin et al., 1997. The heritability of IQ. Nature 388, pp. 468-471.

Jacobs et al., 2001. Heritability Estimates of Intelligence in Twins: Effect of Chorion Type. Behavior Genetics 31(2), pp. 209-217.

Lykken, D., and Tellegen, A., 1996. Happiness Is a Stochastic Phenomenon. Psychological Science 7(3).

Riese, M.L., 1999. Effects of Chorion Type on Neonatal Temperament Differences in Monozygotic Twin Pairs. Behavior Genetics 29 (2), pp. 87-94.

Wichers et al., 2002. Chorion Type and Twin Similarity for Child Psychiatric Symptoms. Archives of General Psychiatry 59(6), pp. 562-564.

Patrick

jayh

The interactions of genotypes would make prediction very difficult, because the actions are in themselves non linear. There is no intelligence gene, high intelligence the fortuitous interaction of a number of different genotypes, prediction of the outcomes before the fact may always be out reach.

Jesse

jayh:
The interactions of genotypes would make prediction very difficult, because the actions are in themselves non linear. There is no intelligence gene, high intelligence the fortuitous interaction of a number of different genotypes, prediction of the outcomes before the fact may always be out reach.

I agree that the interaction of genes during development which determine the phenotype is highly nonlinear, so you can't usually identify specific genes for specific traits. I just don't think it's chaotic, since there is no sensitive dependence on initial conditions--one might say that the developmental process has a single point attractor rather than a Lorentz attractor (although it would probably be hard to make this rigorous). But I agree that because of the nonlinearity a prediction of the effects of changing the genotype would be very difficult, and might require nothing less than an extremely detailed simulation involving every gene, every protein, every cell in the developing embryo, etc.

ps418

You mean, interactions between genes? There is no "intelligence" gene in just the same sense that there is no Phenylketoneuria gene, or height gene. The genes of course are not "for" these phenotypes even though they are associated with those phenotypes.

As to whether high intelligence is the result of fortuitous interactions of a number of different genes, this is an empirically tractable question. As far as so-called "general cognitive ability" or 'g' is concerned, the evidence does not support this strong genetic emergentist view. The evidence that does exist suggests that "most of the genetic variance for g is additive, that is, the effects of the individual genes seem simply to add up rather than there being interactions between the genes" (Plomin, 1999). And there are aleady polymorphisms which have been preliminarily associated with variation in 'g' (e.g. Barbaux et al, 2000; Egan et al, 2001; Malhotra et al, 2002; Comings et al, 2003; Payton et al, 2003). Time will tell is these associations replicate in new samples, but there is no empirical reason to think that it will be impossible to enhance intelligence to some extent by selecting genes.

On the other hand, as I said in my last post, it is almost certainly true that some specific cognitive abilities are the result of such complex interactions. Lykken calls such genetic traits emergenic. Mathematical genius make well be an example of such an emergenic trait (Lykken, 1998), since it does not 'run in families' (in strong contrast to 'g'), though frankly no one knows enough about this to say much.

Barbaux et al, 2000. Polymorphisms of genes controlling homocysteine/folate metabolism and cognitive function.
Neuroreport 11(5):1133-6.

Comings et al, 2003. Role of the cholinergic muscarinic 2 receptor (CHRM2) gene in cognition. Molecular Psychiatry 8(1), 10-11.

Egan et al, 2001. Effect of COMT Val108/158 Met genotype on frontal lobe function and risk for schizophrenia. Proceedings National Academy Sciences 98(12):6917-22.

Lykken, David (1998) The Genetics of Genius, in Genius and the Mind: Studies of Creativity and Temperament in the Historial Record. Oxford University Press.

Malhotra et al, 2002. A functional polymorphism in the COMT gene and performance on a test of prefrontal cognition. American Journal Psychiatry 159(4):652-4.

Payton et al, 2003. Cathepsin D exon 2 polymorphism associated with general intelligence in a healthy older population. Molecular Psychiatry 8(1), 14-19.

Plomin, R., 1999. Genetics and general cognitive ability. Nature 402, C25 - C29.

Patrick