Is Free Will an Illusion?

11 April 2013

Interview with

Professor Gabriel Kreiman, Harvard Medical School, Boston

How is 'free will' is implemented in the brain? Kate Lamble spoke to Professor Gabriel Kreiman from the Harvard Medical School about his research and started by asking how you nail down such a philosophical question in science...

Gabriel -   The type of experiments that we have done follow on from the steps of several other previous scientists who try to reduce freewill to very simple questions.  Let's say, I ask you to tap your finger and you can choose exactly when you do that.  So, whenever you feel the urge to do that, you can tap on a given key.  So, we're asking a very simple question, what, when, how neurons get activated to dictate or determine that particular action?

Kate -   How do you measure those neurons?  Are you using a skull cap of electrodes or are you using a more invasive procedure?

Deep brain stimulationGabriel -   The traditional way to investigate this question has been through scalp electrodes.  These are about 2mm contacts that are around the scalp.  What we took advantage of is that rare opportunity to peek inside the brain in particular cases of patients that have epilepsy and therefore need to be implanted with electrodes, very thin microwires that can capture and listen to the activity of single neurons and very, very small ensembles of neurons.

Kate -   So, if you're measuring these neurons as they make the decision whether to click the mouse or not, what are you finding?  Are people very conscious of when they make the decisions they want to?

Gabriel -   Well, of course eventually, people are conscious about when they make that decision and their subjective report is what we take as the ground truth about what they're conscious of or not.  But what we're finding is that neurons in particular parts of the frontal lobe, more specifically an area called the supplementary motor area and the pre-supplementary motor area get activated and are making those decisions for the subject way in advance, way before subjects become conscious of that decision.  So, several hundred milliseconds and in some cases, even seconds before people know that they are going to execute that action.  The neurons are firing like crazy saying, "Now, we're going to be orchestrating this action."

Kate -   You're saying it happening way before, but we're talking hundreds of milliseconds.  How long is hundreds of milliseconds in neural terms?

Gabriel -   This is sufficiently long that we can ascertain that neurons are making those decisions before people become aware of them.  To some extent, all of our decisions, unless we believe in magic, have to be orchestrated by neurons and the particular situation in which we could record and monitor the activity of some of these neurons helps us to tap into some of these populations of neurons that are making these decisions for the subject before subjects actually realise what they're doing.

Kate -   So, if our neurons are firing and making the decisions before we're conscious of them, how does that make implications for whether we have free will or not?

Gabriel -   The simplest interpretation is that basic physical laws dictate when neurons get activated, when they cross a given threshold that's what ultimately leads to a decision.  After the fact, we believe that we are the owners of that decision.  We create this illusion of free will, but there's nothing really free in free will.  These are neurons that are activated and we can describe with equations what they're doing and when they're activated and so on.

Kate -   What's the next step in research?  How do we build up to working towards a more complicated action rather than just clicking a finger?

Gabriel -   That's very difficult.  So, one step that may not sound extremely exciting is to actually ask subjects to choose which hand they want to use.  So, not just which finger or when, but they now have a little bit more freedom.  Ultimately, we do not think that we can make predictions about what you're going to have for lunch tomorrow or who you're going to marry 5 years from now.  So, we're making very simple decisions over very, very small scales with our capability in terms of recordings.  So, to some extent, I'd like to come back to a discussion of determinism as well as chaotic systems.  We cannot really predict very well when we flip a coin whether it's going land on head or tails, but no one would claim that the coin actually wanted to land one way or the other.  It's just a problem that's very sensitive to initial decisions.  So similarly, we don't envision anytime soon being able to make predictions on scales of hours or days that are accurate in any way.  But that may just simply be problem with technology, how many neurons we can tap into.  Ultimately, what you're going to do 5 hours from now may very well depend on a lot of external factors, not just your own brain.  But the main point we're trying to make here is that our illusion of free will, may be nothing more or nothing less than ensembles of neurons and circuits firing together in specific patterns.

Kate -   It's quite intimidating as a person who thinks that they go around deciding what they do every day.  How much does the fact that you're having to work on epilepsy patients and where you put the electrodes are based on clinical opinion, how much does that limit to the next steps of the research that you can do?

Gabriel -   We're very limited in many ways.  We're limited in which locations we can investigate, we're limited in terms of how many neurons we can record from.  On the other hand, one of the only ways in which we can investigate the human brain at this level of resolution.

Kate -   Does it have any implications in treatment if we can understand how people make decisions and at what point they make decisions?  Can we intervene at any point?

Gabriel -   Yeah, that's a fascinating question and we're only taking very preliminary steps in that direction.  There's a couple of neurological conditions that involve severe disorders in volition.  Parkinson's is the case where people have a lot of difficulty initiating movement.  There are many problems in addiction research where again, these are basic decisions that people need to make.  So, there has been handful of studies, one by the neurosurgeon that I work with, Itzhak Fried, where they stimulated electrically this particular area and the pre-supplementary motor area.  That is they inject that current into the brain, again for clinical reasons in these patients, and people reported that they felt the urge to make a movement.  So, in principle, one can bias the activity of neurons and enhance people's abilities or people's intentions to make specific decisions.

Add a comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.