This October marks the publication of my first lead author scientific article, the culmination of my thesis work thus far. Simply put, I found a previously unknown brain signal in the ventral pallidum that reports how good a food outcome was relative to the other available options. What does this signal mean, and why might it be important? In honor of Halloween, I will use trick-or-treating as an analogy to describe my results.
Imagine that you live on a street where you have gone trick-or-treating for years. You know that the first house you plan on visiting will give you either Snickers or Reese’s peanut butter cups. Let’s say, like me, you like both candies a lot, but Reese’s is your favorite. If, when you visit this first house, you end up getting Reese’s, ventral pallidum will have really strong activity when you receive your candy. But, if you end up getting Snickers, ventral pallidum will have reduced activity.
Now, let’s say the second house you visit always gives out either Snickers or saltine crackers. Saltines are obviously a much worse treat than candy, so now Snickers is by far the favorite outcome. When you go up to the second house, ventral pallidum will now have really strong activity if you get Snickers (even though it used to have its activity suppressed for Snickers at the first house) and, instead, ventral pallidum will have really low activity if you get saltines.
Finally, if you go to a third house that typically gives out all three options, ventral pallidum’s level of activity for each outcome will be according to the relative value of each: strongest for Reese’s, slightly less activity for Snickers and very low for saltines.
Why might the brain be signaling this kind of information? For one, it may help us decide whether or not to go back to that house next year. And it could help explain why we can feel very excited or very disappointed to get the same reward (Snickers, in this case) depending on what else we knew we could get.
So how did the experiments actually work? Instead of studying trick-or-treaters, I studied rats. They learned that they could receive multiple deliveries of sucrose or maltodextrin dissolved in water for about an hour each day. Sucrose and maltodextrin are both carbohydrates that taste sweet and, like Snickers and Reese’s, they are highly palatable rewards for the rats, although rats prefer sucrose. After recording the neural activity in ventral pallidum while the rats consumed either sucrose or maltodextrin — and finding a very strong signal related to their preference for sucrose — I switched out sucrose for water (like the second house in the example) and found that the response to maltodextrin became much stronger than when it had been contrasted with sucrose. Finally, like the third house, I provided all three outcomes and saw that the signal scaled with the relative value of the three outcomes.
One reason why this signal in ventral pallidum was so remarkable was that I did not find nearly as strong a signal in the region that neuroscientists have traditionally believed is the input to ventral pallidum — nucleus accumbens. This finding means that the way we think the brain processes information about rewarding outcomes needs to be revised. I hope that future work will continue to investigate how ventral pallidum is involved in our responses to food rewards.