Cake or Death? It’s all a matter of self-control, and your vmPFC

Cake or Death? Cake, please.
Today’s article comes to you from ars technica, where a friend of the blog found this article from Matt Ford on how we pick our food. Given that Sci is in the middle of reading “How We Decide” by Scienceblogs own Jonah Lehrer, she couldn’t pass this decision-making kind of article up. It was a conscious decision, I think. Or was it?
I don’t know about you, but everyone Sci knows appears to be trying to regulate their diet somewhat. Most of my friends think the word “die” is in the word “diet” for a reason. But we all still try to eat well, though we’re not going to be subsisting on grapefruit and wheatgrass juice any time soon. We’re all hearing the messages that we need to eat lean protein, get lots of fiber, and eat as many fruits and vegetables as we can afford.
The media nowadays is jam packed with advertisements saying we need to make healthy choices, right next to other advertisements letting us know about the many many offerings available at fast food restaurants for less than a dollar. With all this coming into your brain every day, it can be really hard to make the right decisions. Would a granola bar be good right now? Or should you have some fruit instead? Should you eat the chocolate cake or the ice cream?
And these decisions become even harder when it becomes a choice of fruit vs. cake (I don’t know about you guys, but the fruit would be languishing on Sci’s shelf for months). In these cases, your decision making has to have some layers to it. Rather than just “ooooh! Cake!”, you also need to think “eh, but I already HAD chocolate today, and I haven’t had more than one serving of fruit…and I should get more fiber because apparently everyone needs more fiber…”
chocolate cake decision.jpg VS banana decision.jpg
Sorry, Banana. No contest. LOOK at the chocolate shavings on that thing…sigh…
So anyway. Our brains may want cake, and other parts of our brains want healthy. The two sides need to fight. But which areas of your brain are responsible? And how do they modulate each other? That’s what this group wanted to find out. Hare et al. “Self-control in decision-making involves modulation of the vmPFC valuation system.” Science, 2009.

The recipe for a Science paper:
1) Take dieters
2) Put dieters in a scanner
3) Mix dieters vigorously with lots of different foods of varying tastiness
4) Stir
Ok, it’s a little more complicated than that. The researchers took a group of people, all of whom said they were on a diet. They then put them in a scanner, and showed them a whole bunch of foods, which they had to rate on two scales: health and taste. Looked like this:
decision food1.png
The thing on the upper left is an apple. The other thing is a Butterfinger. So in this paradigm, most people would rank apples as high on health, and neutral to low on taste, while they would rate the Butterfinger as high on taste, but low on health. Of course, there were 50 options, for those who like neither apples nor Butterfingers.
Once the participants had rated all the items according to health and tastiness, they were shown one food (let’s say the apple), which they rated as neutral, both good for you and decent tasting. They then scanned the patients as they presented them with a set of decisions, and forced them to make a choice. Would you rather have the apple or the celery? How about the apple or the Butterfinger?
The authors suggested that dieters with higher levels of self-control (who might be more successful at maintaining a diet) would choose healthier foods in preference to non-healthy, tasty ones, and that dieters with lower levels of self-control would not. And sure enough, that’s what happened. They were able to divide the groups into high and lower levels of self-control, depending on the choices they made.
And then they scanned the brains of the dieters while they made those food decisions. They hypothesized that signals encoding “value” (yum) would be localized to the ventro-medial Prefrontal Cortex (the vmPFC). Activity in the vmPFC has been known to correlated with behavioral measures of value. You can see it here:
decision food2.png
The activity in this area when the participants saw a food they thought was tasty was pretty equal across all patients. Mmm, that looks good. But what was the difference between those who went for the cake and those who restrained themselves? That was in the dorsolateral prefrontal cortex (DLPFC). Activity in this area increased when people restrained themselves and picked the healthy food, and was lower when people couldn’t stay away from the cake.
decision food3.png
(The DLPFC, which showed more activity when participants exercised self-control)
Based on these results, the authors concluded that the basic value of an item was in the vmPFC, and that signals from this area were modulated by the DLPFC. So the more self-control you exerted, the more the DLPFC lit up, and the more you went and picked the apple because you knew it was good for you.
This study, which helps pinpoint where signals for self-control may originate, could be a big deal clinically. Not necessarily as a diet aid, but rather for problems where there’s a lack of self-control, as in addiction (where you have no self-control where the drug is concerned).
And it’s a pretty cool study, but Sci has some issues with it nonetheless. First of all, this is imaging. The most you can do (and the most the authors did, they didn’t say anything they couldn’t say) is draw a correlation between one area, another area, and a behavior. Correlation is not causation. We can’t say for certain that the DLPFC IS your self-control center. It could just be a relay area for measures of self-control coming from other places.
And I have a bit of an issue with the subjects in the study. There was no real control group. There were only dieters who had self-control, and those who didn’t. I’d want that paired up with a group of people who wasn’t dieting at all. I also want to know whether those who had more self-control were better at achieving their health goals than those with less self-control. And finally, I’d like to see a chart of people’s moods. If you make one bad decision and you KNOW you’re choosing a non-healthy option, does it make you give up? Are you more likely to just keep choosing the cake because you figure you’ve given in? Is higher activity in the DLPFC and a conscious healthy choice correlated with a positive feeling? Is it more likely to make a group show more self-control in the future? Studies which assess what the patient is experiencing as well as the choices they’re making could go further to address self-control issues in things like addiction.
The final issue is actually one with the data. the vmPFC lit up in the dieters when they assessed something of value. In dieters with higher self-control this area lit up when they perceived something they had ranked as HEALTHY. This means that healthy looked a lot better to them than it did to dieters with less self-control. And this makes you wonder whether the “self-control” dieters had more self-control after all, or whether they merely had grown to like carrot sticks. Not only that, an increase in vmPFC activity when self-control dieters saw something healthy could very well be MORE important in their choices than the modulating self-control activity from the DLPFC. I’m not at all sure how the authors might go about fixing this issue, but it’s definitely something to think about when you’re looking at people with “self-control”. Maybe, after you get hungry enough, even celery starts to look good.
Hare, T., Camerer, C., & Rangel, A. (2009). Self-Control in Decision-Making Involves Modulation of the vmPFC Valuation System Science, 324 (5927), 646-648 DOI: 10.1126/science.1168450

13 Responses

  1. “If you make one bad decision and you KNOW you’re choosing a non-healthy option, does it make you give up? Are you more likely to just keep choosing the cake because you figure you’ve given in?”
    I wonder how much the fact that it’s a hypothetical choice modifies this.
    I seem to remember a story way back in the day that men were more likely to react to a diet “failure” (like eating chocolate cake) by pledging to do more situps/compensate for it, whereas women were more prone to the “well I’ve already eaten on cake, what’s another one” type of failure.

  2. I’d like to take issue on how they categorise the foods as ‘healthy’ and ‘nonhealthy’. Chocolate cake, butter fingers and other yummy, sweet fatty treats can be a valuable part of a healthy diet. You just have to make sure you have them in moderation. If you really love chocolate and want to keep to your diet, it is healthier to incorporate a small serving of chocolate as part of the diet so you don’t have to miss out and pine for what you are ‘not allowed’ to eat. Of all the nutrients we need calories is the biggest. Why not substitute for a serving of pasta a small slice of low fat chocolate cake occasionally? The cake has more nutrients (eg. fibre, VitE and a few B’s from the wholemeal flour, egg protein and oils, milk proteins and oils, magnesium and theobromine from the cocoa) and is a good source of calories, which you do need, especially if you are exercising as well as dieting. This labeling of foods with moral values stuffs with peoples minds.

  3. I know when I was last losing weight, I actually trained myself to like eating healthy stuff. I would eat veg in preference to biscuits, because I actually enjoyed it more.
    I also found that for me, the simplest technique was to simply not buy cakes and biscuits – if I had it in the house, it was a strong temptation and I ate it, but if I didn’t see it I didn’t eat it. Unfortunately, this is no longer working because I’m living with my partner and my brother, who both fill the cupboards with junk.

  4. This is an excellent review of the paper, but I actually have an issue in addition to the ones you listed.
    They are arguing for the purposes of this paper that vmPFC represents food value. Yet in a variety of other fMRI papers — including, I suspect, other published by Rangel et al. — other brain areas have been posited as the location of reward-related mapping. I have seriously heard no less than 5 including OFC, ventral striatum, ACC, etc.
    It would lend a great deal more credibility to studies of this nature is if there were ANY cross-study consistency about what is contained in these reward-related maps. It isn’t that I deny that there are distributed reward representations in the brain. Just ask Nestler. It is that you can’t have it both ways. You can’t say that a brain region does something in one study and then something totally different in another just because it makes your data seem cooler.
    I am really getting down on all of these fMRI decision-making papers because I find their reasoning on WHAT is encoded in the studied brain region extraordinarily facile. There seems to be no ongoing effort to reconcile findings across studies.

  5. They should have also qualified what they meant by “diet”. One can diet to lose weight, and the term is more commonly used in this sense. However, once can also diet to increase their weight if they should choose to, or even to try to increase lean muscle mass specifically. How would these latter dieters have compared? Is it possible that in trying to gain weight they would go with the lean protein in chicken jerky or whey powder than the bacon cheeseburger and in doing so exert just as much self control as the other dieters?

  6. Jake: You are right, I know a lot of stuff about the ventral striatum (including NAc) being implicated in goal-directed and reward-directed behaviors. I didn’t mention it because I was assuming that they meant that vmPFC was measure of quantifiable value, while the ventral striatum was more basic. OFC I thought was more complex, but I’m not sure.
    I agree that you can’t say what is encoded. You can only say there is a correlation between activity in an area and a behavior, and drawing any conclusions as to what that activity is is a delicate matter.
    Toaster: good point. I can look in the supplementary materials if you like. Ideally, I would want them to recruit from somewhere like Weight Watchers, a program in which values of foods are very clearly quantified.

  7. Great post. I have to agree with you and several other commenters:

    In dieters with higher self-control this area lit up when they perceived something they had ranked as HEALTHY. This means that healthy looked a lot better to them than it did to dieters with less self-control. And this makes you wonder whether the “self-control” dieters had more self-control after all, or whether they merely had grown to like carrot sticks.

    If you assume the the light-up represents a Pavlovian response to (past) feelings of gratification, then maybe the people with “higher self-control” had merely experienced so much sense of gratification from “doing the right thing” that they lit up as much as anybody responding to physical gratification.
    BTW, did you know the Research Blogging Profile Page attribute all your great posts to Evil Monkey?

  8. But AK, perhaps we are ONE AND THE SAME!!!!!!!!!!!! MWAH-HA-HA-HA!!!!!

  9. Meh.
    I *heart* Eddie Izzard. That’s the extent of rational discourse I can manage at the moment. Great post, sorry I’m too mentally flogged to leave a great comment.

  10. I was not very fond of this paper, and my criticisms go well beyond the ones already voiced here. If you look at the subject selection criteria in the Supporting Online Material, Hare et al. say:
    ‘We recruited two types of subjects: 1) individuals who self-reported being on a diet to lose or maintain weight and 2) individuals who self-reported no current monitoring of their diet. All subjects reported that they enjoyed eating sweets, chocolate, and other “junk food” even though they might be restricting them from their current diet.
    . . .
    Subjects were classified as self-controllers (SC) or non-self-controllers (NSC) based on their behavior during the experiment, and not on their self-reports about diet status during the recruiting process.’
    So really (as I said in my post), the study isn’t about dieting at all, because one could be a “self-controller” in the experiment and yet report no dietary restrictions in real life. Of course, this is assuming the Supporting Material is correct and the main paper is wrong about subject selection.
    Along with Jake, I also had a problem with the assumption that the brain’s “value assigner” is located in vmPFC. The words dopamine, nucleus accumbens, and basal ganglia did not appear in the article (ignoring previous studies on the topic). The phrase “vmPFC-striatal network” appeared once but was dissed right away.
    Finally, I thought the authors were remiss in not citing the pertinent literature on cravings/appetite and PFC. Bring on the cake…

  11. i choose cake. my bananas need to be made into banana bread. though, the smell of several vegetables from the allium family make me VERY HUNGRY. i can’t explain it.
    you know, these fMRI studies just keep coming, and the correlations never get me very excited.

  12. David Kessler was on Fresh Air yesterday promoting his new book (see below). In his interview, he cited a lot of research based on brain imaging that supported his contention that fat, salt, and sugar stimulate certain parts of the brain. Having only heard the interview and not read the book, I am not sure if the kind of study discussed here is what he is talking about, but it sure sounds similar:

    Fresh Air from WHYY, May 13, 2009 · Former FDA commissioner David Kessler warns that sugar, fat and salt can hijack our brains and cause us to overeat. In his new book, The End of Overeating, he describes the way the food industry works with the advertising industry to create the food cravings that are so hard to resist.
    Kessler, who is a pediatrician, served as FDA commissioner under presidents George H.W. Bush and Bill Clinton. He has been the dean of the medical schools at Yale and the University of California, San Francisco.

  13. i like chocolate, but hate cake. hmm, i’d toss the cake and frosting. also overripe bananas or dried banana “chips” are yucky…
    cake vs a good var of tangerines, mango, muscat grapes, late (season) apple, cherry… cake loses!
    pomelo, sweetsop, good dill pickles, …

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: