My profile of Read Montague and the dopamine prediction-error hypothesis is now online. I wanted to write this article for two main reasons. First of all, I think the dopamine story is incredibly exciting and remains one of the best examples of how subtle shifts in neural firing rates can allow the brain make sense of the real world. Yes, I know there are caveats, but the prediction-error hypothesis is still a very powerful paradigm. Wolfram Schultz should win a Nobel Prize.
Secondly, there's so much crappy fMRI research out there - and it always get so much press attention - that I wanted to do some reporting on the best brain scanner experiments, even if they required a little more explanation than "The "fill-in-the-blank" brain area is responsible for sarcasm/romantic love/jealousy/etc." Not only has Montague's lab helped pioneer some very clever paradigms - he invented hyperscanning - but he insists on using very rigorous analytic techniques. He's a scientist who is skeptical of the technology and his own data. As Montague told me: "All of these [fMRI] studies with twelve, thirteen subjects are completely bogus. Where we're at with brain imaging right now is where genetics was thirty years ago. The stuff we think is signal is really noise. We've got no handle on individual variation, so we end up throwing out most of the really interesting stuff." A typical Montague fMRI experiment will involve at least 75 subjects, which is an extremely large n for the field.
Here, by the way, is Montague's latest Science paper which looks at how people with borderline personality disorder play simple economic exchange games.
Brain & Behavior
Comments (6)
More mind candy from Jonah: his article about the dopamine prediction-error hypothesis! though the term is not a grabber. Jonah describes a simple experiment done by Schultz that explains a lot of human and animal behavior. This has been used by years by kind, astute animal trainers.
Writes Jonah:"His experiment observed a simple protocol: he played a loud tone, waited a few seconds, and then squirted a few drops of juice into the mouth of a monkey. While the experiement was unfolding, Schultz was probing the dopamine-rich areas of the monkey brain with a needle that monitored the electrical activity inside individual cells. At first the dopamine neurons didn't fire until the juice was delivered; they were responding to the actual reward. However, once the animal learned that the tone preceded the arrival of the juice - this requires only a few trials - the same neurons began firing at the sound of the tone instead of the sweet reward. And then eventually, if the tone kept on predicting the juice, the cells were silent. They stopped firing altogether."
If you want an animal to repeat a desirable behavior you give them a reward at the exact time they exhibit the behavior. (In so far as is possible you ignore unwanted behavior...this is the kicker).Sometimes this is not convenient (getting a fish to a dolphin in midleap) so you sound a whistle, having earlier sounded the whistle when the dolphin is eating a fish. So the dolphin learns to expect a fish after the behavior you cue and ultimately will enjoy jumping itself. Or if the animal gets bored, you have to reinforce with something different and not everytime. Read "Don't Shoot the Dog!" by Karen Pryor or "What Shamu Taught Me About Life, Love, and Marriage" for the details of this kind and wise way to communicate with the animals in your life. The hardest part is training yourself but it works on you too!
Posted by: jb | August 13, 2008 2:28 PM