Wednesday, August 27, 2008

Decoding the ASPM puzzle

Remember the kerfuffle over ASPM two years ago? ASPM is a gene that regulates brain growth. It evolved considerably in the primate lineage leading to humans and continued to evolve even after the emergence of modern humans, with the latest variant arising about 6000 years ago somewhere in the Middle East. The new variant then proliferated within and outside this region, reaching higher incidences in the Middle East (37–52%) and in Europe (38–50%) than in East Asia (0–25%).

Interest died down when it was found that this variant, despite its apparent selective advantage, does not seem to improve cognitive performance, at least not on standard IQ tests (Mekel-Bobroy et al., 2007; Rushton et al., 2007). Nor do ASPM variants correlate with human brain size variability (Rushton et al., 2007).

Now, new light has been shed on this puzzle by a paper on ASPM in other primates. This gene was initially linked to overall brain size because non-functioning variants cause microcephaly in humans. A comparative study of primate species, however, has shown that evolution of ASPM does not correlate with major changes in whole brain or cerebellum size:


Particularly striking is the result that only major changes of cerebral cortex size and not major changes in whole brain or cerebellum size are associated with positive selection in ASPM. This is consistent with an expression report indicating that ASPM’s expression is limited to the cerebral cortex of the brain (Bond et al. 2002). Our findings stand in contrast to recent null findings correlating ASPM genotypes with human brain size variation. Those studies used the relatively imprecise phenotypic trait of whole brain instead of cerebral cortex size (Rushton, Vernon, and Bons 2006; Woods et al. 2006; Thimpson et al. 2007). Although previous studies have shown that parts of the brain scale strongly with one another and especially with whole brain (e.g., Finlay and Darlington 1995), evidence here suggests that different brain parts still have their own evolutionary and functional differentiation with unique genetic bases. (Ali & Meier, 2008)

This is a point I raised a year ago. If we look at how the new ASPM variant spread geographically and temporally, it seems to match a very specific mental ability, and not general intelligence:

At present, we can only say that it [the new variant] probably assists performance on a task that exhibited the same geographic expansion from a Middle Eastern origin roughly 6000 years ago. The closest match seems to be the invention of alphabetical writing, specifically the task of transcribing speech and copying texts into alphabetical script. Though more easily learned than ideographs, alphabetical characters place higher demands on mental processing, especially under premodern conditions (continuous text with little or no punctuation, real-time stenography, absence of automated assistance for publishing or copying, etc.).

…How well are these tasks evaluated by standard IQ tests? Although most tests involve reading, transcribing, and taking dictation, these abilities are not evaluated over long, uninterrupted time periods. If we look at the two studies that discounted a cognitive advantage for the new ASPM variant, neither tested its participants for longer than 82 min and the tests themselves involved a mix of written and verbal tasks.

It seems premature to conclude that the new ASPM variant is unrelated to cognitive functioning. Current IQ tests do not adequately evaluate mental processing of alphabetical writing, particularly under premodern conditions. Yet this is the cognitive task whose origin and spread most closely coincide with those of the new ASPM variant in human populations. It is also a demanding task that only a fraction
of the population could perform in antiquity, in exchange for privileged status and probably superior reproductive opportunities. (Frost, 2007)

Is this cognitive task localized in a specific part of the brain? There seems to be evidence for such localization … which I will review in my next post.

References

Ali, F. and Meier, R. (2008).
Positive selection in ASPM is correlated with cerebral cortex evolution across primates but not with whole brain size. Molecular Biology & Evolution. Advance access

Frost, P. 2007. "The spread of alphabetical writing may have favored the latest variant of the ASPM gene", Medical Hypotheses, 70, 17-20.

Mekel-Bobrov, N., Posthuma D., Gilbert S.L., et al. (2007). The ongoing adaptive evolution of ASPM and Microcephalin is not explained by increased intelligence. Hum Mole Genet, 16, 600–8.

Rushton, J.P., Vernon, PA.., Bons, T.A. (2007). No evidence that polymorphisms of brain regulator genes Microcephalin and ASPM are associated with general mental ability, head circumference or altruism. Biology Letters-UK, 3, 157–60.

6 comments:

Tod said...

C.Coon remarked somewhere that during the Roman Empire Syrian slaves came to be unavailable on the open market because their skills were so valuable. Armenians are another middle eastern people who have long been noted for intellectual abilities (not always an an unalloyed benifit for a people of course).

The Middle East being ASPM central would explain a lot. ASPM could retain its selective value into modern times, with tight linkage of reproductive success to special demands on mental processing akin to that required from premodern scribes. A possible example of this is the special form of expression used in talmudic scholarship, this is said to be a "complex argot" which it takes many years to master let alone conduct disputes in. Moreover reproductive success was linked to such attainment because of the prestige attached to it. [4comments at Nat. Selection in Proto Industrial Eur.]

Anonymous said...


Though more easily learned than ideographs, alphabetical characters place higher demands on mental processing, especially under premodern conditions (continuous text with little or no punctuation, real-time stenography, absence of automated assistance for publishing or copying, etc.).


What is this view based on?

As a student of Chinese, I can only think that someone who has never studied Chinese can think that it places less cognitive demands on people than an alphabetic script does.

For one things, Chinese characters have structure to them and you can make good guesses as to the pronunciation of a character and its meaning when seeing a new character (after you know a reasonable number of characters).

Moreover, ancient Chinese was in much the same boat as older versions of other scripts from other areas, so great cognitive demands were required there as well.

Can you be more specific?

Peter Frost said...

Tod,

Yes, I've come across many comments in ancient literature about the high social status given to scribes. I'll check out your other comments ...

Anon,

To save time, I'll just cut and paste the relevant paragraphs from my article:

"Writing began about the end of the 4th millennium BC in the Middle East, initially with pictorial symbols that represented ideas rather than sounds. Phonetic alphabets then emerged in the 3rd to 2nd millennia, subsequently spreading to Europe, North Africa, and the Indian subcontinent. East Asia independently developed a pictorial, ‘‘ideographic’’ system, perhaps in the 2nd millennium BC, but never developed a predominantly alphabetical system, with the exception of Korea some 600 years ago.

Alphabetical writing uses fewer characters and is easier to learn. It does, however, impose greater demands on short-term memory because longer strings of characters have to be stored and more transformations have to be performed on them. In contrast, Chinese ideographic writing seems to evoke meaning much faster, apparently because the mind encodes the characters visually and maps them onto meanings directly [7–10]."

[7] Biederman I, Tsao YC. On processing Chinese ideographs and English words: some implications from stroop-test results. Cognitive Psychol 1979;11:125–32.
[8] Hayes EB. Encoding strategies used by native and nonnative readers of Chinese Mandarin. Mod Lang J 1988;72:188–95.
[9] Schmitt BH, Pan Y, Tavassoli NT. Language and consumer memory: the impact of linguistic differences between Chinese and English. J Consum Res 1994;21:419–31.
10] Treiman R, Baron J, Luk K. Speech recoding in silent reading: a comparison of Chinese and English. J Chinese Linguist 1981;9:116–25.

Anonymous said...


Alphabetical writing uses fewer characters and is easier to learn. It does, however, impose greater demands on short-term memory because longer strings of characters have to be stored and more transformations have to be performed on them. In contrast, Chinese ideographic writing seems to evoke meaning much faster, apparently because the mind encodes the characters visually and maps them onto meanings directly [7–10].


This seems to be the fault in your thinking.

Since you are from Canada, you are no doubt aware of those chain of stores called FCUK.

I seem to recall recent research showing that people recognize words more by their shape than their actual spelling, and that you can interchange letters and they will still be recognized.

In addition, as I have already said, Chinese characters are not just arbitrary shapes that have to be remembered. They have internal structure, and they largely encode syllables.

Moreover, there are other dimensions to the language which make it unlike European languages (that I am familiar with), one of which is that all adjectival material precedes a noun or noun phrase (and the adjectival phrase can be very long). I would think that this places similar demand on short-term memory ...

Of course, Chinese people do not seem to have problems learning English or other languages as well, so if the claim is that ASPM some how provides neural support for learning Alphabetic languages ... what are East Asians using?

Peter Frost said...

Anon,

If there is a flaw in my thinking, it also exists in the studies I cite. There is credible evidence that Chinese ideographs evoke meaning faster than alphabetical letters (see the studies cited above). If this evidence is flawed, please point out the flaw.

My working hypothesis is that the new ASPM variant provides neural support for reading and writing under premodern conditions, specifically those of scribes. Those conditions no longer exist.

In short, everyone can learn to read and write with or without the new ASPM variant (I will discuss this point further in my next post). The selective advantage becomes apparent only when high mental throughput has to be maintained for long periods of time.

I've never done real-time stenography, but I have done real-time translation. Believe me, after an hour the mental fatigue becomes overwhelming. Yet those were the conditions that ancient scribes had to work under.

Anonymous said...

Chinese characters are not ideographs. And they are far harder to read, to recognize, and to distinguish one from another. The very fact that you call them ideographs shows that you have no knowledge of them. In addition, many other alphabetic and syllabic systems were independently invented, mostly by people who saw Chinese and said, "Are you kidding me? There's got to be an easier way to do this."

I, too, read Chinese. These arguments are absurd, especially as a huge percentage of the populations who were subjected to this increase in cerebral cortex were wholly or largely illiterate until the modern era. You can't just make up arguments from the vague preconceptions that you're carrying around. That's not how this works.