Mirror neurons: language, actions and intentions
This morning I listened to part of the Brain Science podcast, an episode discussing mirror neurons with Dr. Michael Arbib. In his 1998 paper, “Language within Our Grasp”, Arbib and his co-author, Giacomo Rizolatti argue that Broca’s area (implicated in human language capabilities) contain mirror neuron systems which are used in gesture recognition. A similar mirror nueron system in monkeys (in what’s called area F5) functions in the same way.
Arbib and Rizolatti’s claim, when it comes to language, is that “such an observation/execution matching system provides a necessary bridge from ‘doing’ to ‘communicating’,as the link between actor and observer becomes a link between the sender and the receiver of each message.”
Thus recognizing actions and imitation (and only later on, communication) are the primary building blocks for language, not an innate Chomskian Universal language.
* Note - as always, I write these posts with the caveat that I’m interested in other fields because of their connection to philosophy; I have no specialized training in neuroscience or the sciences in general. Please always check primary sources and do not take my summaries as authoritative.
In Arbib’s theory, mirror neurons “light up” when a monkey or human recognizes that they are observing someone perform an action, as opposed to merely making movements. If I lift an ice cream cone to my mouth, certain neurons will light up as I make the movement. However, if you lift an ice cream cone to your mouth, mirror neurons are those neurons which will light up in that instance as well as when I lift the ice cream cone. These neurons in humans are located in Broca’s area.
What’s interesting is that the monkey area F5 has been suggested by some to be a structure that is an evolutionary precursor of Broca’s area in humans. Thus, perhaps it was a “neural prerequisite for the development of interindividual communication and finally of speech.”
How would this work? A theoretical reconstruction is below:
1. Actor A performs an action and Observer B’s mirror neurons are activated. The brain is basically “ready” to mimic the action, but the action is blocked.
2. Sometimes a “prefix”–perhaps a muscle twitch of some kind–occurs, which signals to Actor A and Observer B that they’re directed to the same action/object.
3. This association would be recognized and selection pressures would benefit animals who are able to use this proto-communication. As Arbib put it in the podcast, changes in the brain influence culture which then influences the brain and so on.
4. In more detail, area F5 and Broca’s area are involved in controlling oro-facial, brachio-manual and oro-laryngial movements. So, the monkeys could move their mouths, make gestures and emit sounds. Gestures aren’t usually the de facto method of communication, except among certain groups, so the authors speculate that lip-smacks and grunts became paired with gestures later on. The benefit would be seen in expanding the audience from one observer close by to one’s facial expressions to a wider group.
That’s the basic idea, at least, although much more detailed work is being done in fields like neuropsychology, chemistry, cognitive neuroscience, etc.
In terms of philosophy, I wonder - and this is pure speculation without having the texts in front of me, or much understanding of mirror neurons - if this is any kind of solution to the problem of the indeterminacy of translation. Quine’s worry always seemed odd to me: that in my pointing at a rabbit and saying “Gavagai!”, you cannot really know whether I am saying “There’s a rabbit!” or “Look at the temporal slice of rabbitness over there!” or “Movement in the field!” While those translations cannot be fully discounted–because any “test” I make to try to uncover “Gavagai” has to take other sentences into account–ideas like “temporal slice” or generalizing about movement don’t seem as adaptive as identifying potential food sources. Instead, those other concepts are parasitic on basic actions.
That doesn’t solve the philosophical problem in any way, but it simply makes me wonder about the strength of the worry. If meaning is built, evolutionarily, upon my mirror neurons firing and recognizing action, then, aren’t we starting off with more similar mental lives than disparate? In a world in which humans’ brains worked differently and we quantified objects differently–like temporal slices, perhaps, counting my pencil-today as different than pencil-yesterday and pencil-tomorrow–then those other translations would be more feasible.
Source: Arbib, M., Rizolatti, G. (1998). Language Within Our Grasp. Trends in Neuroscience, 21(5), 188-194.
July 10th, 2008 at 1:25 pm
I think Quine’s concern goes beyond the ‘temporal slice of rabbitness’. If I remember Paul Roth’s take on Quine correctly, the mere act of indexically pointing has no substantial verbal logic. Maybe this was more Roth than Quine, but the concern would be mine as well. My youngest (2 years old) just hit her communicative stride and as my wife and I ask every day, “How did she know to say that?”
It was not mirroring activity (if I understand the thrust of this post). Every single day there are new combinations of phrases and sentences that grip on to different aspects of reality as she sees it. Not many of these are modeled, but rather seemingly invented. Maybe your point is that the mirroring capability has to be there in order to generate some initial language skills, and surely that mirroring aids in the effort. But as you know, that is a huge linguistic leap to posit that language is fundamentally built upon mirroring (I’m not sure if that’s what you’re saying).
I’m still waiting for people to strap video cameras to toddlers all day and night so non-parents can investigate this bizarre language phenom as well. I remember Paul Roth describing language modeling for children as ‘merely priming the pump’. He also admitted to modifying his views of language genesis after having a daughter (when he was in his forties). By my forth child, I had better figured out what is mirrored (i.e. my oldest daughter saying ‘that hurts my feelings when you…’) verses tooled language (i.e. my youngest saying ‘I not know how it pink’ as an answer to a novel question). And the veritable expolsion of language is so dizzying that it causes most parents to talk insesantly about what ‘junior said today’. Just some anecdotal thoughts.
July 10th, 2008 at 4:49 pm
If I remember Paul Roth’s take on Quine correctly, the mere act of indexically pointing has no substantial verbal logic. Right - I think the “rabbit parts” was to show that pointing doesn’t tell us what is being pointed out, whether temporal slices or movement etc. (Actually, on a side note, the paper I just read on eliminativism and intuitions related to this question, specifically of folk beliefs about statues existing contra van Inwagen’s claims that they are like unicorns in some way.)
Maybe your point is that the mirroring capability has to be there in order to generate some initial language skills, and surely that mirroring aids in the effort. But as you know, that is a huge linguistic leap to posit that language is fundamentally built upon mirroring (I’m not sure if that’s what you’re saying).I’m pretty sure that’s what the paper was saying (I’m reserving judgment on whether it’s right, of course! I have no idea!) The authors did some work explaining how they thought this occurred - and there are a few follow up papers which I’ll admit I haven’t read. The paper I cite above is actually available online here
They use “case grammar” to explain “pre-linguistic” grammatical representations, dividing it into declarative and imperative proto=sentences which have “slots” into which objects would go. The transition to language would fill these slots with verbs, nouns, etc. But in terms of action, it would be something like:
Declaration: grasp-A(Luigi, raisin)
Where Luigi is the agent and raisin is the object
The problem, which I think you’re getting at above, is that we do more than mirror actions. Back to Quine, how do we know to slot ‘raisin’ into the object spot in a radical translation? (Would interpreting chimp sign language be a real life radical translation example?)
And the authors admit that other parts of the brain, like the temporal lobe, have to be involved when we make the transition to language using tense, mode, etc. I should go back and look at Stephen Pinker’s recent book, which talks a bit about this sort of thing. Anyway, in his podcast, the author of the study said that while we can never, with 100% certainty tell the story of how language evolved, comparative work between chimps, apes, humans and other animals is what he thought would fill in some of those gaps.
But, back to your daughter, he kept reminding us that even the smartest ape using sign language is stuck at the level of a two-year old. There’s something tremendously different about the way children learn language and apes and yes, I don’t think mirror neurons are the whole story, but I don’t get the sense that the researchers think that, either.
July 11th, 2008 at 12:49 am
If I recall Quine correctly (and it’s been a while since I read him), it’s inaccurate to speak of “the indeterminacy of translation” as a problem) . . . a philosophical puzzle that requires solution. Rather, he puts forward the i. of d. as a thesis, saying “This is the way things are.” His strategy is: he sets up a conceptual problem (Do we translate gavagai as “rabbit”, “temporal rabbit-slice”, or “undetached rabbit part”?) argues that the problem has no solution, hence isn’t really a problem; and when we come to see that it isn’t really a problem, certain larger philosophical problems will dissolve before our eyes.
In other words, there isn’t really a freestanding “philosophical problem of translation” (although there are all kinds of interesting questions to be raised about translating from one natural language to another, including many of philosophical interest.) Quine intended the original scenario as a way of sharpening our intuitions about fundamental issues concerning the metaphysical underpinnings of our conceptual schemes, as manifested in the languages we speak. Of course, some would say that, rather than clarifying anything, he’s just quined the problem away.
July 11th, 2008 at 1:13 pm
That’s what I get for talking off the cuff when I haven’t finished Word and Object yet! Let me revisit the text today and look again at that section on radical translation to see where I got the idea it was a philosophical puzzle.
Thanks…
July 16th, 2008 at 5:33 pm
[...] Chatter: Eversaved: I’m actually sad I missed this… Jeong: Wish I had a mac Alan: If I recall Quine correctly (and… Alan: Am I overlooking something obvious, or is… dru [...]