Rethinking FOXP2

Earlier studies have indicated that a gene called FOXP2, possibly involved in brain development, is extremely conserved in vertebrates, except for two notable mutations in humans. This finding suggested that this gene may in some way be involved in the evolution of language, and was thus dubbed by the popular press “the language gene”. See, for instance, this and this for some recent research on the geographic variation of this gene (and related genes) and its relation to types of languages humans use (e.g., tonal vs. non-tonal). Furthermore, a mutation in this gene in humans results in inability to form grammatically correct sentences.
This week, a new study shows that this gene is highly diverse in one group of mammals – the bats:

A new study, undertaken by a joint of team of British and Chinese scientists, has found that this gene shows unparalleled variation in echolocating bats. The results, appearing in a study published in the online, open-access journal PLoS ONE on September 19, report that FOXP2 sequence differences among bat lineages correspond well to contrasting forms of echolocation.

As Anne-Marie notes, this puts a monkey-wrench in the idea that FOXP2 is exclusively involved in language, but may be involved in vocalizations in general:

Said gene might have a new function (sensorimotor) besides the one originally attributed to it (verbal language).

Jonah Lehrer notes that the same mutation that in humans eliminates ability to use or comprehend correct grammar is also found in songbirds and the gene is expressed at high levels during the periods of intense song-learning. The story is obviously getting very interesting – does this gene have something to do with vocalizations? Or with communication? Or something totally third?
Looking forward to further responses by other blogs, hopefully Afarensis, John Hawks and Language Log?
The article on FOXP2 in bats was published yesterday on PLoS ONE so you can access it for free, read, download, use, reuse, rate, annotate and comment on.
Update: Mark Liberman explains more (and takes me to task for a mistake I made in haste last night) in this post on Language Log.
Update 2: John Hawks explains.

4 responses to “Rethinking FOXP2

  1. “… this puts a monkey-wrench in the idea that FOXP2 is exclusively involved in language”
    Quite right. Also, the patterns of sequence variation in bats point to no obvious links between sequence, selection and function. And the paper says that echolocating whales aren’t especially bat-like in their FoxP2, so there doesn’t seem to be a smoking gun for echolocation (or any other function) in the sequence.
    As far as I can tell (I might be wrong), we don’t know what genes FoxP2 regulates. That seems like a good thing to go looking for to me.

  2. Totally agree. I hope this triggers a frenzy of active research on this gene in a variety of organisms (has anyone looked in invertebrates yet?).

  3. Well, I took the bait — see
    http://itre.cis.upenn.edu/~myl/languagelog/archives/004936.html
    But you’ve gotten rather confused about what Dediu and Ladd did — they correlated the geographic distribution of lexical tone with the geographic distribution of ASPM and Microcephalin, neither of which is FOXP2 or is especially closely related to it. See the end of my Language Log post for details.

  4. Dear Bora,
    Here is what Babel’s dawn has to say on FoxP2; thought might interest you.
    Guru