Free Will and Neuroscience

Neuroscientific experiments seem to provide evidence that questions long held beliefs in free will. Dale Beyerstein argues against this interpretation of the experimental results .

Neuroscientific experiments seem to provide evidence that questions long held beliefs in free will. Dale Beyerstein argues against this interpretation of the experimental results .


There is a belief amongst those who are aware of contemporary neuroscience that its findings have undercut any rationally founded belief in human free will. In particular, neuroscientists have, allegedly,  disproven the idea that we are free to consciously decide to do move a muscle, and then simply do it. And since this is an action that is at the basis of much more complex actions, these findings call into question our commonsense view of free action.

At the heart of this commonsense view is a model of rational deliberation, which philosophers for millennia – since Aristotle in Book III of the Nicomachean Ethics – have taken to be the foundation of free will. In that account, deliberation is the ranking, then choosing between alternatives, and the result of this effort is action. And such deliberation must be conscious – otherwise how could we aware of those alternatives?

Also, we also expect our muscle movements to be under our conscious supervision when we perform a free action. Think of bumping into someone and apologizing: “Sorry, it was an accident”. This is your way of saying you didn’t do it freely. Why? Because you weren’t paying attention. In addition, we would expect our free action to be initiated by a conscious ‘push’, because if we didn’t consciously initiate the movement, our responsibility for it is limited to that of someone riding along who did nothing to stop it.

This would lead us to expect the following ordering of events in a free action:

a) conscious deliberation free choice muscle movement free action

But this ordering seems to have been turned upside down by experimental results in neuroscience that are so solid they have stood the test for almost forty years. The story of this upheaval begins almost sixty years ago with the discovery that the above sequence needs another link in the chain. Coming before the signal from your brain to move the muscle is a stage in your brain called the readiness potential – as the name implies, a kind of ‘gearing up’ of your brain to send the signal to your muscle to make the movement. So far, not a problem for our notion of free will, merely a refinement. We merely have to add to the above ordering thus:

b) conscious deliberation free choice READINESS POTENTIAL muscle movement free action

But, in an ingenious experiment in the early 1980s by Benjamin Libet, rigorously analysed by the philosopher Daniel Dennett (in his 2003 book, Freedom Evolves, which is the origin of many of my ideas on the subject), designed to show exactly when in the sequence conscious awareness of the decision to move a muscle actually takes place, Libet found the readiness potential occurs before the conscious awareness of the decision. At the beginning of the experiment Libet asked subjects to move their wrist any time they wanted to – the paradigm of a free choice – and to indicate the instant they decided to move their wrist by reporting the position of a dot on a clock face. The dot would be moving around the face several times per second. This report would be correlated with brain activity recorded by sensors on the scalp. Libet found that the readiness potential occurs on average 300 milliseconds before we are aware of deciding to move our wrist. Let’s look at the ordering now:

c) READINESS POTENTIAL free choice muscle movement free action

Let’s be clear on what turns the standard model (b) on its head. The readiness potential is not conscious – that is why it took neuroscientists working from the ‘outside’ to discover it, rather than each of us discovering it for ourselves by introspection. And there it is, occurring before our decision to act. It would appear that we are not acting; some uncontrolled brain function is. It seems that neuroscientists have demonstrated that we are no more in control of our actions than we are when the doctor taps our knee with the percussion hammer and elicits a reflex kick of our leg. Well, maybe a little more, at least on some occasions. For example, once we begin to move our wrist, we might be able to stop it mid-move. As the neuroscientist Vilayanur Ramachandran put it, “This suggests that our conscious minds may not have free will but rather ‘free won’t’!” (Dennett 2003 p. 231). But wait a minute. I can think of three reasons why this conclusion might be premature.

Neuroscience has a great deal to teach us, so long as we pay attention to exactly what it is teaching us, and what it is not teaching us.

The first two concern two assumptions about the nature and functioning of the conscious mind. First the nature assumption: Most philosophers have, until recently, held that the mind is a unity. That is, it has no parts; it cannot be divided. Most famously Descartes (Meditations on First Philosophy, Meditation six), used this idea to argue that the mind must be distinct from the body. The physical body, like all physical things, is not a unity; it consists of parts. Add to this the identification of the mind with consciousness (which is also a unity on this view), and rationality. Furthermore, we identify our mind with just the conscious mind. And last, we view consciousness as an all-or-nothing affair: either it is there or it is not, and it makes no sense to think of it as sharing its responsibilities with any organ of the body, since it is a self-contained entity. Now we see what troubles us about the results of Libet’s experiment: We identify ourselves, the bearer of our free will and responsibility, with our consciousness, and, the latter being mental, cannot be divided. Nor can we think of ourselves as doing something freely while unconscious bodily processes are contributing to the action. So if we admit that the awareness potential occurs without consciousness and is a mere physical process, it is not you (your conscious mind) that is responsible, because you came too late on the scene. It would appear that we have painted ourselves into a corner as far as free will is concerned. So how do we get around this predicament? Well, we recognize the source of this idea of the unitary mind, totally separate from the body: it is in the religious idea of the soul. For Buddhists and Christians, the soul is only tenuously (and mysteriously) connected with the body; and pretty soon will be completely separate from it. But while we are connected with it, it is the we that is free and responsible for our actions. However, those of us who are atheists have already rejected this soul notion, and all that is left is to jettison the baggage of the unitary consciousness. What modern neuroscience teaches us is that consciousness is not indivisible, but rather is the product of a great many brain processes coalescing into what we experience as a unitary perspective on ourselves and the world. What we need to keep in mind is that each one of these brain processes is a part of us, not something that just happens to be happening beside what goes on in the conscious us.

That leads us to the second reason for being more sanguine about free will. Once we see consciousness as a product of brain processes – and see those processes as occurring in our brains – there is less room for disassociation from the awareness potential that is the beginning part of a muscle movement. Sure, we are not aware of it, but that goes for a lot of things that we consider to be free actions. I walked into the kitchen just now, and it was a free action despite me having no conscious awareness of exactly how I made my feet move. I just did. Freely. In fact, I wasn’t aware of anything from the time that I decided I needed a cup of coffee, until I found myself in front of the coffee machine, when I needed to be aware of my next move. I was concentrating on the details of this article. I ‘delegated’ the control of the mechanics of my trip to my unconscious brain processes. However, if my conscious attention was needed for something – for example, if the cat had decided to stand his ground in the narrow entrance to the kitchen – I would have come to attention to deal with the problem. But I still wouldn’t know how I make my feet stop. I just do it. Freely.

Yes, but I can hear your objection: in this last example, you made the decision consciously, then ‘delegated’. In the Libet experiment, the awareness potential occurs before your ‘delegation’.  Well, that is true, but perhaps there is something about this experiment that makes it a unique case, not one from which we should make generalizations. Libet asks his subjects to make the decision to move for the same reason Nike tells you to do things: Just do it. In other words, for no reason at all. He was telling his subjects not to deliberate before doing the action, but rather to randomize. That’s the Game Theorists’ fancy academic term for “thoughtlessness”.  So we now see the third reason for not drawing the conclusion from Libet’s experiment that we have no free will. Astute readers will have already seen this point in the series (c) above: there is no deliberation link shown there. Of course we have no free will when we do not deliberate and simply act thoughtlessly. But when we do deliberate and start off a whole set of processes in our brains that results in a series of actions, then those actions are free. But wasn’t that deliberation started off by a whole set of processes in our brain and in the environment? Sure, but another way of putting that point is that we see a situation in the environment (no coffee), and we use our past experience stored in our brain (that the coffee machine in the kitchen can make coffee) to come up with a solution to our problem (go to the kitchen and use the machine to make coffee).

Given what I have said so far, it would seem that we can’t expect to learn very much about free will from neuroscience. But that is far from the truth. Neuroscience has a great deal to teach us, so long as we pay attention to exactly what it is teaching us, and what it is not teaching us. Perhaps a little parable might be useful to demonstrate the difference.

A lawyer gives her defense of her client:

My client didn’t do it. What happened was this: the gun was in my client’s hand, it happened to be pointed at the victim, certain signals formed in my client’s brain, the signals proceeded along her nerves to her finger, the finger crooked, the trigger engaged, and a bullet was expelled from the gun, unfortunately striking the victim.

Of course, no jury – with the possible exception of O.J. Simpson’s – would be taken in by such a ploy. It’s obvious what the lawyer has done here: she has switched topics from who is responsible for the crime — i.e., who did it – to how it was done. Every aspect of her description is entirely accurate, and is in fact describing what is necessary for any action to take place. But what she leaves out is the information that shows the client’s intentions, why the client had formed them, and what he then did. The understanding of why we do things – in the sense of whether our reasons are good or bad, and the understanding of whether we did an action freely or not – is the job of introspection and philosophical reflection. The understanding of how we do these things is the job of neuroscience. A good scientist arranges an experiment so as to remove any aspects which are unnecessary and which could confuse the results (i.e., to remove the confounding variables). Libet did just that. He removed from the subjects the opportunity to deliberate about when to raise their arm, because it is a confounding variable, given what he wanted to study. And in the process removed the free will from the subject. But only for the duration of the experiment.