Is Free Will an Illusion?

By Brandon Keim

Long before you’re consciously aware of making a decision, your mind has already made it.

If that’s the case, do people actually make decisions? Or is every choice — even the choice to prepare for future choices — an unthinking, mechanistic procedure over which an illusory self-awareness is laid?


Those questions are raised by a study conducted by Max Planck Institute neuroscientists and published Sunday in Nature Neuroscience. Test subjects chose whether to push a button with their right or left hand; seven seconds before they experienced making the choice, their brain activity already predicted their final decisions.

For more on the experiment, see my Wired.com story, for which I had the privilege of speaking to Martha Farah, director of the University of Pennsylvania’s Center for Cognitive Neuroscience and a prominent neuroethicist. As is so often the case in journalism, we had a fascinating (email) conversation that didn’t fit into the article itself, and I decided — ha — to publish it here.


Me: The big question is how much people should feel comfortable extrapolating these results to other, more seemingly complex decisions about which we feel a deep personal connection — do I rent an apartment, get involved in a relationship, leave my job in search of another, and so on.

Martha Farah: The authors have taken an important first step toward understanding how we make decisions, and toward revealing the apparently prolonged cascade of unconscious processes that precede the conscious decisions we make with what seems like "free will." But of course there is always a trade-off in science between making a process scientifically tractable and making it realistic. Remember, Galileo rolled balls down inclines and theorized about infinite frictionless planes; he didn’t set about trying to understand the fluttering, zig-zagging motion of a falling leaf! The authors started with a very simple decision-making task, and their results now form the basis for some good working hypotheses to be tested with more complex decisions.


Me: How do these results square with our notion of free will? Do they obviate free will, which in that light is an illusion; or might there still be a balance between free will and unconscious decisions; or is free will still paramount, but operating at some other level?


MF: Let me start with a very general observation. Neuroscience is changing the way we think about ourselves. One of the hardest changes for people to assimilate is the idea that our intentional, voluntary behavior is the product of a physical system, the brain. If physical processes in the brain cause our actions, then how can there be free will? How can we be held responsible for our behavior? Can’t we just all plead "my brain made me do it"?

The Soon et al paper jumps right into the middle of these issues. It shows us how limited, even misleading, our introspections are. According to the authors, many seconds before we are aware that we have made a decision, we have — or at least, our brain has! All of the data of cognitive neuroscience are pushing us to replace the idea of mind-body duality, which is so intuitive, with the idea that mental processes are brain processes. But these results on the neural processes underlying free decisions rub our noses in it! One can assimilate findings about color vision or motor control being brain functions a lot more easily than findings about consciously experienced "free will" being a brain function, and hence physically determined and not free at all!

I don’t think "free will" is a very sensible concept, and you don’t need neuroscience to reject it — any mechanistic view of the world is good enough, and indeed you could even argue on purely conceptual grounds that the opposite of determinism is randomness, not free will! Most thoughtful neuroscientists I know have replaced the concept of free will with the concept of rationality — that we select our actions based on a kind of practical reasoning. And there is no conflict between rationality and the mind as a physical system — After all, computers are rational physical systems!

Me: As I’m sure you often hear, this makes my head spin a bit. One follow-up: re: the replacement of free will with the concept of rationality, selecting actions based on practical reasoning — I can see how rationality and the mind as physical system don’t conflict, but doesn’t the very concept of selection (and, arguably, reasoning) imply an agency that is rendered illusory by findings like these?

MF: Depends what you mean by agency… If you think of a computer selecting certain actions based on a combination of inputs and stored information about goals etc, then there is a (not too head-spinny) sense in which the computer is the agent selecting the actions. (Of course, what makes the computer that kind of agent that it is, making the selections that it does, is its whole history — how it was designed, what kind of goals and knowledge have been programmed in, etc. — But it is the computer, in its current state, that is selecting and so it seems reasonable to say it is the locus of the rational decision.)

Going back to the Nature Neuroscience findings, the parts of the brain whose activity are correlated with the decision and precede the person’s conscious awareness of having decided — as well as potentially other parts — are the analogs of the computer described above… And this happens well before the conscious experience of "free will" making the decision.

One advantage of focusing on rationality rather than free will is that it enables us to retain the concept of moral and legal responsibility. If someone is rational and is not under coercion (eg someone holds a gun to your head and says you’ll be shot if you don’t do X) then it is reasonable to hold him or her responsible…

Me: Still struggling a bit. But not because of any deficiencies or illogic in your own excellent explanation — I suspect my response to all this is skewed by some instinctive (subconscious — ha) need to cling to the idea of free will. Perhaps because my sense of free will is tied in some inexplicable way to my sense of self and …
authenticity? That’s not the right word. Maybe I should come right out and say (non-religious) soul, of which free will is a manifestation.

Somehow the computer doesn’t seem satisfactory, in the sense that a rational program would make the same decision again and again again. Somehow that doesn’t seem alive. I’m always unsettled by computational analogies to the soul, in the sense that the appearance of self-awareness is not the same thing as self-awareness.

That the Turing test was devised by someone for whom disguising or transcending the body was so (sadly) necessary, and Norbert Wiener himself such a person of mind rather than body, adds to my unsettlement. But that’s my own superstition, rather than a real critique.

Clearly I have enormous difficulty accepting that that my conscious experience of choice is false. The very possibility threatens my sense of self; more fundamentally, it doesn’t make sense to me. Thinking about this produces a sensation I usually equate with self-contradictory instruction sets — "try not to think of a white elephant; what existed before the universe began?" — or certain optical illusions. It just doesn’t compute.

I’m quite aware that my perspective is based on feeling, not rational arguments, but I just can’t shake it. Maybe I’m simply programmed to conflate choice with self, and would think differently if I’d been raised differently. Maybe my kids will someday be just as frustrated with my anachronistic self-conception as I am with my dad’s seemingly hard-wired antipathy to computers. ("The program’s still open, but the window is hidden! How many times do I have to explain this?!") I hope they’re more patient than I am.