language evolution

The May issue of Discover has a transcript of a roundtable between the editor in chief, Corey Powell, and four researchers in robotics. It's an interesting conversation. I found the following quote from Rodney Brooks (founder of iRobot) illuminating:

Rodney, you've talked about four goals that robot researchers should be aiming for. What are they?

Brooks: First, the object-recognition capabilities of a 2-year-old child. You can show a 2-year-old a chair that he's never seen before, and he'll be able to say, "That's a chair." Our computer vision systems are not that good. But if our robots did have that capability, we'd be able to do a whole lot more.

Second, the language capabilities of a 4-year-old child. When you talk to a 4-year-old, you hardly have to dumb down your grammar at all. That is much better than our current speech systems can do.

Third, the manual dexterity of a 6-year-old child. A 6-year-old can tie his shoelaces. A 6-year-old can do every operation that a Chinese worker does in a factory. That level of dexterity, which would require a combination of new sorts of sensors, new sorts of actuators, and new algorithms, will let our robots do a whole lot more in the world.

Fourth, the social understanding of an 8- or 9-year-old child. Eight- or 9-year-olds understand the difference between their knowledge of the world and the knowledge of someone they are interacting with. When showing a robot how to do a task, they know to look at where the eyes of the robot are looking. They also know how to take social cues from the robot.

If we make progress in any of those four directions our robots will get a lot better than they are now.

That's a clever marketing ploy, I think. It makes things sound a lot simpler to break down the problems into easy (2-year-old) and harder (9-year-old).

But wait a minute. What's he's actually saying is, we need robots that work like 9-year-old children!

After all, a 9-year-old comes with the 2-year-old object recognition and the rest already built in.

It's not like the problems solved by younger children are any easier. The fact that children learn object recognition before mastering grammar doesn't mean that object recognition is simpler to manage. It may mean that grammatical ability evolved in primates that already could recognize objects. It certainly means that the brain develops in ways that entail learning to recognize objects first -- not at all irrational considering the requirements of 2-year-old life. Two-year-olds aren't going to be teaching much, they don't need the 9-year-old social awareness. But they do need to recognize objects.

Is the ontogenetic order of these behaviors in children necessary? Or is it an accident of evolution? The answer does impact our choice of strategies for replicating these behaviors in silico. I expect that you do have to recognize objects to be able to understand someone else's recognition of objects. But do you have to understand language in order to have human social understanding? Some scholars would say yes, others would say these are separate "mental modules" that in principle could occur independently.

Maybe the engineering problem will help us clarify the evolutionary one. It turns out that there was a school of thought devoted to the idea, "Evolutionary developmental robotics."

A new paper in Nature (Konopka et al. 2009) reports on microarray expression comparisons of human and chimpanzee-specific versions of FOXP2. The change of two amino acids in the human version has some pretty large consequences for the expression of other genes.

An accompanying essay by Martin Dominguez and Pasko Radic (2009) sums up the study in a paragraph:

To further understand what FOXP2 does on a molecular level, two articles have revealed some of its probable targets, but neither study compared the regulatory effects of human and ancestral FOXP2. This is precisely what Konopka and colleagues have done, using whole-genome arrays to detect differences in gene expression in human neuronal cell lines expressing either human FOXP2 (FOXP2^human) or the ancestral protein, FOXP2^chimp. The authors find that a substantial number of FOXP2 target genes are differentially regulated by FOXP2^human and FOXP2^chimp. Many of these genes met the criteria for positive selection during human evolution (although the authors had no way of assessing their statistical significance). This places their findings in harmony with previous results that show FOXP2-related genes as evolutionary arbiters. Because the authors examine human-specific gene regulation by FOXP2, their work may provide our first window on the co-evolution of regulatory networks that are important for human-specific features such as language, which probably require a number of genetic changes working in concert.

The "FOXP2" is not italicized here, because the passage refers to the protein product. I point that out to remind everybody that many important insights about gene function can only come from biochemical analysis of the resulting gene products. Most of us in paleoanthropology, even in the broadest sense encompassing genetics, don't

What I really like about the result is that it shows FOXP2 is not some "magic gene" that suddenly triggered a cognitive revolution. It's a transcription factor that affects cell proliferation, with effects that cascade in many tissues. And it's highly conserved -- which means it's not like you could just switch it to a different form and expect everything to go right. The kind of genetic comparison that I can do shows the possibility of coevolution:

Previously, we identified ChIP-chip targets of FOXP2 that themselves were also under positive selection6. We hypothesized that networks of genes important for language circuitry had been positively selected through selective pressure on human brain evolution. Thus, we also examined whether any differential FOXP2 targets were themselves under positive selection. Five genes (AMT, C6orf48, MAGEA10, PHACTR2 and SH3PXD2B) met the standard criteria of Ka/Ks > 1.0 for positive selection on the human lineage (where Ka indicates the rate of non-synonymous substitutions and Ks indicates the rate of synonymous substitutions; Supplementary Table 9). These data, along with the haCNS and expression data mentioned above, suggest that a subset of differential FOXP2 targets may have co-evolved to regulate pathways involved in higher cognitive functions.

It seems to me that a cascade of genetic changes may have laid the groundwork for this regulatory shift, and that human populations may still be catching up to that shift today. Changes in these widely-interacting "hub" proteins have to be net good (or at least neutral) or they wouldn't have happened. But that doesn't mean that all their consequences are good -- they drag along a lot of bad effects with the good ones. So such changes may be followed by a series of genetic aftershocks -- changes in the "spoke" genes with functions compromised by the developmental/regulatory shift. Those changes might still be ongoing.

Nor is FOXP2 the only candidate for such a system of genetic changes. The "haCNS" observation was this:

A significant number of the differentially expressed genes [considering human- and chimp-FOXP2] are also associated with human-specific accelerated highly conserved non-coding sequences (haCNS), but not with chimpanzee highly conserved non-coding sequences....

More on FOXP2:

"How the FOXP2 transgenic mice squeak"

"FOXP2 is really recent, it really did introgress (if it's not contamination)"

"The amazing talking Neandertals"

"FOXP2 knockout mice"

References:

Dominguez MH, Rakic P. 2009. Language evolution: The importance of being human. Nature 462:169-170. doi:10.1038/462169a

Konopka G, Bomar JM, Winden K, Coppola G, Jonsson ZO, Gao F, Peng S, Preuss TM, Wohlschlegel JA, Geschwind DH. 2009. Human-specific transcriptional regulation of CNS development genes by FOXP2. Nature 462:213-217.