Well, I hope we’ve had some fun discussing free will, but before we move on I want to talk about the Soon et al. paper to which I alluded yesterday. It’s two years old, but it’s worth describing because the results are so startling and counterintutive. (It’s also very short and should be comprehensible to the non-scientist.) I’m not claiming that this work proves the absence of free will, but it should certainly give us pause when reflecting about how we make choices.
Soon et al. built on earlier work by Benjamin Libet, who showed in 1985 that subjects asked to press a button at a time of their choosing showed brain activity in the supplementary motor area (SMA) about 400 milliseconds (0.4 seconds) before they were aware of having made a decision to press the button. Soon et al. note, however, that the SMA may not be a place where the decision to press the button actually originated, and also that Libet’s experiment, while showing that a decision could be predicted, didn’t show that there was “a free decision between more than one behavioral option.” (That is, it wasn’t a choice between two different outcomes; it was simply a choice to act.)
Here’s what Soon et al. did. First, they hooked up subjects to a functional MRI machine that recorded activity in various parts of the brain. Then the subjects were presented with a computer screen on which a letter of the alphabet was flashed; these images changed every half second. They also had two buttons, one under the index finger of each hand.
The subjects were asked to press a button with either hand, and also to remember the letter that was on the screen at the moment when they decided which button to press. (They indicated this letter by pressing another button.) Button presses took place about every 22 seconds, and left and right buttons were pressed with equal frequency. At the same time, the MRI showed the location of brain activity, which could be correlated with which button was subsequently pressed.
Here’s the surprising result: the brain activity that predicted which button would be pressed began a full seven seconds before the subject was conscious of his decision to press the left or right button. The authors note, too, that there is a delay of three seconds before the MRI records neural activity since the machine detects blood oxygenation. Taking this into account, neuronal activity predicting which button would be pressed began about ten seconds before a conscious decision was made.
The earliest brain activity occurred in the frontopolar cortex (FPC) and subsequently moved into the parietal cortex, areas different from the SMA where Libet detected activity. Curiously, the brain activity determining when the button would be pushed was detectable—5 seconds beforehand—in the SMA, but the activity reflecting which button would be pushed occurred in the FPC. As the authors note, “there appears to be a double dissociation in the very early stages between brain regions shaping the specific outcome of the motor decision and brain regions determining the timing of a motor decision.”
The authors conclude:
Taken together, the two specific regions in the frontal and parietal cortex of the human brain had considerable information that predicted the outcome of a motor decision the subject had not yet consciously made. . . Thus, a network of high-level control areas can begin to shape an upcoming decision long before it enters awareness.
This is dry scientific prose, but what it implies is that our decisions—certainly in the case of which button to push—appear to be made long before we’re conscious of making them. This is a really interesting result with wide-ranging implications, and I’m surprised it didn’t get published in Nature or Science rather than Nature Neuroscience. And I think it has to be considered when we talk about things like free will. What is making the decision, if not our own conscious selves? Could we find the same result if we hooked up somebody to an MRI and sent him to Baskin-Robbins to choose one of 31 (now 26) flavors? Could we get this predictive brain activity for even more complicated decisions? There’s a lot of exciting research to be done.
What does the study, then, say about free will? This is discussed at Wired:
Caveats remain, holding open the door for free will. For instance, the experiment may not reflect the mental dynamics of other, more complicated decisions.
“Real-life decisions — am I going to buy this house or that one, take this job or that — aren’t decisions that we can implement very well in our brain scanners,” said Haynes. [John-Dylan Haynes, an author of the study.]
Also, the predictions were not completely accurate. Maybe free will enters at the last moment, allowing a person to override an unpalatable subconscious decision.
“We can’t rule out that there’s a free will that kicks in at this late point,” said Haynes, who intends to study this phenomenon next. “But I don’t think it’s plausible.”
Soon, C. S., M. Brass, H.-J. Heinze, and J.-D. Haynes. 2008. Unconscious determinants of free decisions in the human brain. Nature Neuroscience 11:543-545.
Libet, Benjamin. 1985. Unconscious cerebral initiative and the role of conscious will in voluntary action. Behavior. Brain Science 8:529-566.