Never go grocery shopping hungry: the fMRI study

About a month ago now (yikes, it’s been a while!), Sci handed over her diet to the judgment of the goddess. The goddess Isis. Since then, and since finding out that Sci ate roughly 2500 calories a day (or more, and running 5 miles a day doesn’t help THAT much), some major changes have been taking place. This little muffin dumped a lot of her carbs, and increased her protein, and more than both of those, she has upped her fruit and vegetable intake. This is less expensive than it sounds, thanks to the abundant loveliness of my local farmer’s market, but it still hasn’t been easy. Carbs taste GOOD. Cooking takes TIME. I never HAVE any time…and I like scones. A lot. Sigh.
Anyway, I have current reduced calorie intake to around 2000 calories a day (sometimes), and try not to eat too much pizza (most of the time). I am currently training for some ridiculous running distances, so exercise isn’t a problem, and I give myself weekends off to eat whatever I like. And I write it ALL down. Every day. Sci is in this for the long haul. I’m not looking to lose too much weight, rather, I am looking to improve my diet, eat more veggies and fruit and make these changes LAST.
And I have my good days and bad days. There was the Day of the Entire Pepperoni Pizza for dinner. Followed shortly thereafter by the Day of More Pizza and those Pretzel Chip Things that are So Delicious. But there are good days, too. Days when it doesn’t seem so bad to have fruit instead of fro-yo. And you know, marinated grilled chicken breasts are quite tasty!
But it’s all made Sci think a lot about appetite. Why we eat when we eat, and what we’re really eating for. Why that bowl of candy in advisor’s office is SO tempting even though I just had lunch. You know, that sort of thing.
And then I saw this study:
Piech, et al. “Neural correlates of appetite and hunger-related evaluative judgments” PLoS ONE, 2009.
Which raised far more questions than it answered.


Basically, this was an MRI study. The recipie:
Take 8 grad students
Stick them in an MRI when they’re not hungry
Show then a restaurant menu and make them think about the food (Participants were questioned beforehand to find out which foods they liked or disliked, and were presented with either neutral or tasty food items during the task.)
Then stick them in an MRI when they ARE hungry.
Repeat.
What’s different?
Simple experiment, simple answers. Basically, the scientists wanted to know which brain structures respond to foods based on attractiveness (how good you think the food is), which brain structures show the interaction between hunger and the attractiveness of the food, and which brain structures change response as the motivational state (whether you’re hungry or full) is changed. For this study, they use fMRI, fuctional Magentic Resonance Imaging, a technique which can measure the changes in blood flow in your brain. It’s thought that changes in blood flow to a discrete area (such as those to optical areas when your eyes are focused on something) correlate with activity in the region. And so they used this technique to find correlations between certain brain areas, the “attractiveness” of food, and measures of hunger.
And they got some pretty pictures:
appetitemri1.png
You can see here one of the areas (the one glowing) that they narrowed down as being activated for foods of “high attractiveness”. This area is the amygdala, an area of the brain associated with several things, such as emotional learning and the formation of memory. This includes things like the “value” of certain stimuli, such as tasty food, which is a real source for emotional learning when you’re hungry. In this case, activation in the amygdala correlated with highly attractive foods (with higher activation for more attractive foods). This isn’t too surprising, as other studies have shown that the amygdala lights up in response to sweet things, as well as in response to things like attractive faces, things which neuroscientists like to call “appetitive stimuli”.
The second area which lit up nicely in response to the tasty items was the orbitofrontal cortex, an area which has been receiving a lot more attention recently for its role in things like decision-making, which is often dysregulated in psychiatric disorders or in addictive states. And this area not only showed differences in blood flow in response to attractive or less-attractive foods, it ALSO varied in blood flow depending on whether the participants were hungry or not. So when people were hungry, those appetizing foods looks much better than they did when they had just eaten.
Finally, they got to the good stuff. The scientists looked for a correlation between items which received a higher attractiveness rating when the subjects were hungry and brain area activation. Stimuli that were of higher “value” when the subjects were hungry not only lit up the amygdala and the orbito-frontal cortex, they ALSO lit up the thalamus, the insula (an area now being recognized for it’s role in the salience of stimuli, or how much something is worth paying attention to), the prefrontal cortex…a lot of places. Unfortunately, they didn’t really have a very good hypothesis for WHY this might be the case, but I might speculate that foods that are more attractive when you’re hungry are probably those that are highest in caloric value (carbs, maybe protein and sugars). And these foods are worth paying a lot of a attention to and going after, and so you’re going to get more brain activation associated with them.
The conclusions? The amygdala appears to be correlated more with the attractiveness of the food, regardless of whether you’re hungry. In contrast, the orbitofrontal cortex varies in activation depending on how hungry you are AND how attractive the food is. And other brain regions follow along in assessing the foods that look the best when you’re the most hungry.
Overall, Sci has no problem with this study. It’s a simple experiment with simple answers. But it’s still fMRI, which means it’s just a correlation. Just because a brain area lights up doesn’t mean there’s DEFINITELY activity going on that is DIRECTLY related to the effect you’re studying. You can draw correlations from MRI, but you cannot determine causation.
And Sci feels like they could have done more with this study. They could have looked at which TYPES of food increase in attractiveness when you’re hungry. Is there a higher percent increase in the response to, say, pasta when you’re hungry (high calorie, filling, etc), vs something like a salad, which, though it’s still food and may be delicious, is not going to be as filling? Could you come up with a food ranking of foods that you’re most likely to REALLY crave when hungry, and could we then sequester them all in a single area of the supermarket so that Sci could go shopping before dinner without angst over walking by the cookies which are ALWAYS right next to the produce?
And I would love to go further with this. How much does the attractiveness of a given food, particularly a high calorie vs a low calorie food change when the subject FEELS food deprived? I’m not talking hungry for lunch, I’m talking dieting. When you feel like you CAN’T have a given food, does that make you want high calorie foods more, or all foods in general? And if it is just high calorie foods, are there reward systems activated? If this study has already been done, Sci would love to see it. Drop me a line and let me know!

14 Responses

  1. Causation vs correlation? Sci is retarded! There a lots of things wrong with conclusions based on functional imaging. Your criticism is just stupid!

  2. I’m trying to remember what kinds of food I usually buy when I’m hungry and my memory plays tricks with me… I recall buying lots of fruit and veggies, sometimes an odd can of beans or some funky colourful pasta… seems I don’t want to remember the candy bars bought on impulse right before the cash register (but I do!).

  3. This study has flaws with its underlying precepts. Everyone knows graduate students are always hungry.

  4. Nice summary. sounds very like Berridge’s Liking vs Wanting – we’ve found evidence that BLA in animals is involved in the ‘liking’ appetitive response. There’s a 2006 review by Killcross & Balleine in TINS that’s worth reading too

  5. ohboy: um…feel free to read the post, first. I didn’t just up and call this a great study (it’s ok, but it’s nothing stunning), and correlation vs causation is only one of the problems associated with fMRI. I did not actually go into the many other problems associated with fMRI because I did not have time. Some other post, perhaps.
    Mike: well, I don’t know that they were graduate students, per se. They were the right age, though (around 27), to be grad students and post-docs.

  6. the real question: for grad students, is how desireable a food is correlated with it’s free status?

  7. ahhahahahaha! Nevermind. Jorge has it covered (I find this amusing because immediately after reading this I caught today’s PhDcomic in my reader). Probably for much less than fMRI studies cost.

  8. Sci, I’m fairly sure that ohboy’s post is either outstanding performance art, or a computer program set on “hateful & drunk.”

  9. DrJohn: yeah, figured it was computer set to “Flyby”. Sci is afraid she doesn’t really tend to react well before her first three cups of coffee…

  10. Very interesting. I recently cut my caloric intake to 2160 calories a day. I used some site to calculate caloric needs of a sedentary 180 lb man, not that I am one, but that I some day aspire to be one. Anyway, I find it doesn’t matter if I’m hungry or not when I go to the store as long as I have a list. Just stick to the items on the list and get out of there before the Oreos come calling! Then it doesn’t matter what looks attractive to eat.

  11. Menu items? They’re testing recall. And who can recall how something really smells?
    Now, can they do the test presenting real food substances?
    Can they alternate between different foods and make the little bright spot in the brain blink off and on?

  12. I know that whenever money is looking tight and I might not be able to afford much of anything until the next payday, I get this overwhelming urge to find food, cook food, and horde whatever I get… well, whatever I don’t eat on the spot.
    Until I do something like make a loaf a bread or a pot of soup, I’m totally distracted and can’t concentrate on work.
    But sucking on a spoon of fatty peanut butter does help soothe me.

  13. There was the Day of the Entire Pepperoni Pizza for dinner.

    I tend to do this while petsitting.

  14. “…an area now being recognized for it’s role in the salience of stimuli…”
    That would be “its”. It’s disconcerting to think that a Ph.D. doesn’t know the difference.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: