I actually had another piece in mind for this Friday’s Weird Science, but a friend of mine sent me a link to this paper, and I HAD to do it. It’s not just weird, it’s awesome!
Long, Fee. “Using temperature to analytze temporal dynamics in the songbird motor pathway.” Nature, 2008.
I will start this post with an observation. My brother and I observed, around about the time the movie “Titanic” came out, that you can do “the Macarena” to ANY song. Really. Any song you like. Pick one. Our personal favorite was “My Heart will Go On”, partially because we didn’t like the song. This paper made me think of the Macarena, and how you can do the same sequence of movements to any song written in anything like 4 or cut time (or even 6/8, but you had to get creative). The same sequence of movements can be performed at many different paces.
But what controls how fast you perform a particular series of movement? How can you sing “Happy Birthday” at a breakneck pace, and then at the pace of a funeral dirge, all without losing the tune (if you can keep the tune in the first place)? And that is what this article is about. And it’s the way they did it that makes it weird.
Motor movements are complex. Especially learned ones. It doesn’t take a lot to simply twitch a leg, but highly complex motor movements are involved even in something as simple as walking. And playing the piano or violin gets mind boggling. Complex motor movements become encoded in the brain in a hierarchical series. Take a song. Singing a song is a complex learned behavior. It’s not just knowing the notes to the song and whether your voice has to go up and down, it’s also about knowing the rhythm of the song, and how fast its going to go. The question the authors were asking in this paper is how exactly the pace of a song is controlled. What areas of the brain control how fast you are singing?
To do this they looked at zebra finches. Zebra finches are cute little guys, and a favorite of neuroscientists. You practically look at them and they breed (always a plus), and they exhibit all sorts of interesting and complex innate and learned behaviors to look at.
In this case they were looking at zebra finch song. Birdsong is a learned behavior in finches, and when they learn their little particular snippet of song, they will sing it over and over, exhibiting “bouts” of singing. Each bit of song runs about 0.5-1 sec. And it’s not a simple song. Each song has little bursts of song syllables, and the whole thing can have a very complex acoustic structure. Luckily for scientists, when a finch learns a song, the behavior becomes stereotyped; the song will always run for the same period of time. If you measure the song snippets in various finches, they vary in length by only 1{9f43b4361d9a125bc126dd2a2d1949be02545ec69880430bc4fed2272fd72da3}!
Zebra finch song is pretty well characterized, and is mediated by several well-studied groups of motor neurons. But it’s still not known which parts of the system are used for timing, syllables, or phrasing. The authors of this study focused on two different brain regions, the high vocal center (HVC), and its downstream motor nucleus, the robust nucleus of the arcopallium (RA). The HVA projects to the RA, and the RA in turn projects to vocal motor neurons to control singing.
So how do you find out if a set of neurons controls timing? You either speed them up or slow them down. And you do that using temperature. So in this case, the authors put a heating and a cooling coil on top of the finch’s head, with electrodes going into their the HVC or the RA. And then heated them, or cooled them, and looked at the effects on song speed. And I’m going to include one of their pictures because it’s REALLY cool to look at.
So what you can see if you look at the picture is the effects of heating and cooling the HVC. They heated it up a little, and it actually sped up the song a little, though I don’t think it was significant. On the other hand, when they cooled the HVC down, the song got slower! And a LOT slower. The more they cooled it, the more drawn out the song got. But the really awesome thing that you can see is that the song NEVER changed FORM. It remained the same tune, in the same pattern, sounding exactly the same, just at a much slower tempo.
They also tried this in the RA: no luck. Nothing changed about the song no matter how much they cooled the neurons. So this means that the stereotyped tempo of a zebra finch’s song can be localized to one tiny little area of the brain! This actually goes against the hypothesis of many groups, who thought that song timing had different aspects that were divided up among different areas of the brain. But what the authors have found here is a much more elegant and parsimonious explantion for how tempo works in the brain. One area, going downstream to signal another, and then right out to the motor system. Gorgeous.
Ok, for those of you who are looking at this and going “parts of my brain get cool when I slow down a song? Um, that’s stupid.” You’re right. Cooling the groups of neurons in this example was a way to get them to slow down far more than might normally be seen. Neuron firing from different areas of the brain speeds up and slows down all the time, in reaction to all sorts of stimuli. Your neurons don’t have to be put on ice.
So now they’ve localized the temporal aspects of song to one tiny area of the brain, I’m sure they will start in trying to find which other parts control which aspects of song. Humans don’t necessarily have areas of the brain that correspond perfectly to the zerbra finch HVC and RA. But there are many human studies out there of stimulation in various brain areas to look at effects on behavior and cognition. I can’t wait to see the study on stimulation while you try to hold a tune! I hope they have video.
Michael A. Long, Michale S. Fee (2008). Using temperature to analyse temporal dynamics in the songbird motor pathway Nature, 456 (7219), 189-194 DOI: 10.1038/nature07448
And with that, I’m off to the Society for Neuroscience meeting! It’s the biggest meeting of the year, and it’s kind of like 20,000 neuroscientists all swimming home to spawn. We all swim against the current, fighting traffic delays, bad flights, cancelled flights, lost luggage, and crappy hotels an hour away from the convention center, all this just to cruise for hours amongst thousands upon thousands of posters. Seriously. One morning or afternoon poster session is so big it takes you ten minutes to walk from one end up the room to the other. We’re all going to be on total science overload. Get up, think science. Learn science. Present science. Meet more scientists. Go out to dinner and talk science. DRINK and talk science. I dream of science…and hopefully it will give me some great ideas for posts! If you happen to be at the meeting, DrugMonkey is scheduling a meetup, and I’m hoping to show up and get lost in his many fans. And if you happen to be amongst the posters, find me if you can. 🙂
Edit: I finally checked my google reader, and you should REALLY look at Grrl Scientist’s post on this article! She’s got sound clips! And since she’s an ornithologist (as I am very not), she also has better explanations.
