As some of you may know by now, Sci blogs a bit about dopamine. Dopamine seems pretty simple at first look (one chemical, one transporter, five receptors, how hard can it be?), but in fact, dopamine modulates a huge number of processes, particularly those related to learning and motivation. We talk a lot about dopamine being a "pleasure" molecule, and in a way it in, but it's more complicated than that. It's not just pleasure, it's motivation and reward processes, which in a way are deeper than just the pleasure you might feel at having sex or eating a pizza. Obviously dopamine can have some pretty big effects on things like, for example, motivated movement (Parkinson's), or disregulated motivational processes (drug addiction).
But what about touch? Can dopamine levels influence how you process touch, and how well you can do on a test for it?
Pleger, et al. "Influence of dopaminergically mediated reward on somatosenory decision-making" PLoS Biology, 2009.
First of all, how do you even TEST this? I mean, there's tons of effects of dopamine all over the brain, a lot of which are very hard to differentiate. Dopamine is kind of a neuromodulator, released as a neurotransmitter, and affecting the firing rate of neurons in other regions. And how do you differentiate a task that works with reward as well as somatosensory information (somatosensory has to do with your sense of touch, but more than that, it's feelings of touch and movement internally and externally all over the body)?
In this case, the authors of this study didn't really try to distinguish between touch and reward when looking at dopaminergic effects. Rather, they combined the two, making the reward the participants received dependent on their ability to complete a touch task. This both solves some problems, and creates others, as we shall see.
So what did they do? They combined a behavioral experiment in humans (feeling touch on their right and left hands, and evaluating the frequency for money), with fMRI, which looks at blood flow in in the brain over time. And then they combined both of THOSE with dopaminergic drugs to increase and decrease levels of dopamine in the brain. You can see this will get complicated. Worry not, Sci will sum up.
The Behavioral Task:
So here's how it goes. You're in an MRI machine (we'll get to that part) with electrodes on both of your index fingers and a pedal under your feet. Each set of trails is in little mini-sets of four, each with the same possible reward value. Before each mini-set, you get a flash on the screen telling you how much you will get rewarded for each of the four trials you get right. Smiley face = you're doing this for free. Other than that, 2 coins = 20 cents, 5 coins = 50 cents, and 8 coins = 80 cents. Like this:
You also notice the arrows at the bottom of the screen. Each arrow indicates which finger you're supposed to be concentrating on for the next trials.
Then you were given four trials, and for each one you got right, you earned the specified amount of money. Each trial consisted of your index fingers being shocked twice (lightly, enough for you to feel it, no pain). One shock would be low frequency (three pulses), while the other would be high frequency (five pulses). This would occur on both fingers, but each finger would get a DIFFERENT stimulus, so you really had to concentrate on the finger being tested. Then you had 5 seconds to respond as to which stimulus (the first or the second) was stronger. You did this by pressing the pedal at your feet, once for the first, twice for the second. THEN, a few seconds later, you would get a sign on the screen saying whether or not you got it right, and what your reward was.
By giving you a few seconds between the stimulus, the response, and the reward, the authors hoped to be able to distinguish via fMRI the signals as they occurred (fMRI has a time lag, so this can be a problem).
fMRI, or funtional magnetic resonance imaging, is a way to measure in pretty good detail how blood is flowing in the brain in response to stimuli. The blood flow increasing or decreasing is thought to correspond with an increase or decrease in brain activity in the related area. There are some pretty big problems with this (changes in activity could be small enough not to bring increased or decreased blood flow, sometimes increases in blood flow could mean a decrease in activity, things could be delayed or very fast in time scale, etc), but it's the best thing we have right now to study brain activity in close to real time in humans, and to look at the entire brain and break it down by brain area. Sci has volunteered for a number of fMRI studies, and if you're interested, you should try it! It's boring as anything while you're doing it, but it's awfully cool to see pictures of your own brain afterward. And then you can send them to Sci, and Sci will tell you that you have a HUGE...caudate.
So anyway, there are a set of brain areas that always "light up" in a scanner when someone is doing a behavioral task during fMRI. These will include the prefrontal cortex, the supplementary motor area, premotor cortex (the motor areas are for the motor responses you are presumably making), posterior parietal cortex, insula, caudate, and striatum. When you are doing a task that involves somatosensory input (touching anything, so most of the time), the primary somatosensory cortex will also show activity. So in this case, they focused on the signals coming from reward related areas (like the striatum), and those related to somatosensory input (the somatosensory cortex), and saw how these changes during the task, and under the influence of dopaminergic drugs.
There are a couple of ways that drugs can increase dopamine levels in the brain, though many of them are illegal and/or addictive (like cocaine, amphetamine, etc). In this case, however, the authors used the popular Parkinson's treatment Levodopa. This is actually a compound called L-dopa, which is the precursor molecule to dopamine in the dopamine synthesis pathway.
If you increase L-Dopa, you increase dopamine levels in the brain. In the case of Parkinson's, where dopamine neurons are not present, this allows people's movement, induced by dopamine levels, to return. In the case of people without Parkinson's, it will still raise the dopamine levels in the brain.
The other drug used in this study is haloperidol, which is an antagonist at the D2 receptor, and is a typical antipsychotics (marketed as Serenese and a bunch of other names). The net effect of haloperidol is a decrease in dopamine levels.
So with these two drugs (and a placebo), the authors looked at the behavioral tasks under levels of normal dopamine, high dopamine, and low dopamine. They looked at how successful people were at the task, and how the fMRI signal changed at levels of high dopamine vs levels of low dopamine.
Not surprisingly, they found that high levels of dopamine increased activation in reward-related brain areas such as the striatum. But they ALSO found a correlation between dopamine levels and the primary somatosensory cortex. This correlation was also reward dependent, meaning that fMRI signal levels increased in the somatosensory cortex when the people got their rewards for good performance. The people with high dopamine levels ALSO performed better at the somatosensory task. These levels and the level, of performance were correspondingly low when the participants were under the influence of haloperidol.
The authors concluded that high levels of dopamine not only worked in reward regions, but also had modulatory effects on somatosensory regions. This means that, when you have a higher reward in the offing, you perform better at the somatosensory task, because you somatosensory areas become more sensitive.
Sci's going to admit that she's got some issues with this paper. For one thing, it's really hard to differentiate the effects of dopamine and their effects on reward, attention, and motivation, from direct somatosensory effects. Just because the somatosensory cortex was more highly activated doesn't necessarily mean that it's more sensitive to reward related stimuli. On the contrary, the sensitivity to reward-related stimuli could be the increased attention and sensitivity to reward caused by higher levels of dopamine. Additionally, dopamine has known effects on alertness, which definitely has a lot to do with somatosensory stimuli, and it could be through this effect that the sensory stimuli are becoming more salient.
So basically, Sci agrees that increases in dopamine are probably resulting in increased performance in the sensory task. But she's not so sure this is from direct effects on the somatosensory cortex. Though it could be. But Sci thinks you might have to do a different test (more invasive, and probably not in humans) to figure this out, and separate out the direct effects of dopamine on your somatosensory skillz.
Still, it's a very elegant study for one so complicated, and does a good job of pulling together pharmacology, imaging, and behavior into one human study. For more details (Sci didn't really have time to cover ALL the results), check it out for yourself!