Free will and fMRI

I don't believe in free will. I believe that the decision I will make today about whether to cook pasta or rice for dinner will be a result of millions upon millions of factors, none of which I have control over. These factors might include: the amount of potassium in my blood stream, whether I ate pasta or rice 18 days ago, whether I know a person named Rick and whether I've kept in contact with him over the past 10 years, my entire child hood experience with pasta and rice, whether I stub my toe on the office chair near my desk in a few minutes, whether there is any pre-made pasta sauce in my cupboard, and whether I dreamt about running a marathon last night. There's some room for error, or 'probability', due to the uncertainty principle of quantum physics, but it is a countably infinite probability.

Thus, because of what I know about physics and what I know about how our bodies and, more specifically, our brains appear to function, I'm very confident that when I finally make the decision to cook pasta (or rice), the very idea that I had any control over that decision will be an illusion. And it will be one of the greatest illusions mankind can experience.

However, despite my certainty that our brain operates under the simple quantum mechanical laws which physicists are currently studying, I'm also pretty confident that we are very far away from understanding even the basic functions of the brain as a system.

I'm very skeptical of brain imaging techniques such as fMRI because I have a hard time understanding how brain region activation leads to the wondrous conclusions scientists make whenever they get consistent measurements from people's brains. For example, let's say that when you look at a picture of your mother and regions X, Y, and Z light up in the visual cortex. Then you look at a live fish in a tank and regions X, A, and Z light up instead. Now let's say that this happens pretty consistantly with 9 out of 10 people that show up to participate in the experiment. We then use logic to try to deduce something about regions A, X, Y, and Z. We'll say things like "regions X and Z are all purpose regions that light up whenever you see something familiar, bur region Y lights up whenever you see something familiar and specific (or photographic/static) and region A lights up when you see something familiar but generic (or live/dynamic)". Then you go and devise another experiment to try to pull those possibilities apart.

My problem is that the brain is a pretty complicated object. Trying to figure out what's happening in there when I see a picture of my mom is like trying to figure out how it is that dreaming about running a marathon is going to influence my desire for pasta or rice. So a good hypothesis for the experiment explained above is that "A combination of activity in region X and Y is a result of recognizing the features of a familiar nose, but region Z is where the memories of my mother's voice lies. and back when I was an infant I learned to associate the sound of my mother's voice (one of the first things I learned about my mother in utero) with the sight of a prominant facial feature (my mother's nose). However, upon seeing a moving thing, regions X and Z light up simultaneously because, when combined, these regions help me predict the trajectory of a moving object. Region A is my fish visual memory region and it lights up anytime I see something recongizable to me as a fish."

Of course, occam's razor tells us that we should prefer the simpler explaination over the more complex one. That's a very useful axiom, but when it is applied to over-simplistic data like the activity cloud on an fMRI image I think we're always going to fail reconciling the big picture with the details. But science, being the way that it is, and people, being the way that they are, will combine to make all kinds of interesting and/or impotent predictions. Those are going to be interpreted by the popular press and digested by the masses. When we later find out we were wrong, the new material will propagate while the old material sticks around for a while in the consciousness of the populace who will continue to make claims like "Did you know that Fish and Mothers are actually the same thing in your brain?"

I don't want to be misunderstood, so let me just say that I am happy that people are doing brain image research. We have to start somewhere with any new technology. People should go ahead and use those MRI machines and try to figure out what's happening, preferably not stating their opinions and hypotheses as facts, and write and inform others of the results as carefully as possible.

Until recently, I always thought that the reason why I am skeptical about brain image research's more amazing claims was because I only had a basic grasp of the brain imagine methodology. But as I learn more about what the data actually looks like and what the hypotheses being concocted are I'm actually becoming more skeptical. And then I come across this article which provides me the first concrete evidence that imagining researchers are actually growing weary of the direction their field is heading, in part because of the same issues I have.

I still think that I have no free will, and that the cells in my brain are deterministically leading me on and on to a conclusion which is predictable by anyone with a complex enough computer. But I think that our ability to really figure out which parts of my brain are currently calculating my choice of pasta over rice is about 200 or more years off.