The “free will” experiment

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.

95 Comments

  1. Posted July 28, 2010 at 7:22 am | Permalink

    ‘Free will’ is often defined as want free FROM cause. But why shouldn’t ‘free will’ be defined as want free TO cause?

    • Torbjörn Larsson, OM
      Posted July 28, 2010 at 9:48 am | Permalink

      ‘Free will’ is often defined as want free FROM cause.

      That doesn’t seem to be one of the consistent definitions. Software can have wants (priorities and plans) that they later use to choose (cause). But then the priorities and plans becomes part of the choosing.

      • Posted July 28, 2010 at 9:56 am | Permalink

        What I meant:

        Any measure of ‘free will’ must be based on the effectiveness and feasibility of consciousness volition opposed to the strenght of the environmental influence. We have to fathom the e…xtent of active adaption of the environment by a system opposed to passive adaption of a system by the the environment. The specific effectiveness and order of transformation by which the self-defined system (you) shapes the outside environment, in which it is embedded, must trump the environmental influence on the defined system. What is essential is that the system has to be goal-oriented and the ability to differentiate itself within the environment in which it is embedded.

        What I mean is very simple. If I could get what I want I have had free will. In retrospect the degree of freedom of want is measured by the extent to which I had to adapt my will to environmental circumstances opposed to changing the environment to suit my goals. And basically this is what I mean by ‘free will’. To extent this notion of free will you can ‘measure’ the extent to which one changed his will deliberately, that is consciously, i.e. from within (nonlinear). By nonlinear here I mean a system whose output is not proportional to its input. This is opposed to the ‘persuasion’ of a child by an adult or the contrary affection of one’s will by unwanted, non-self-regulated influence of any kind.

        Anyway, below are a few other more educated opinions on this particular topic.

        Free will according to a AI researcher: “One of the easiest hard questions, as millennia-old philosophical dilemmas go. Though this impossible question is fully and completely dissolved on Less Wrong, aspiring reductionists should try to solve it on their own.” http://wiki.lesswrong.com/wiki/Free_will

        Free will according to a neuoscientist: http://bjoern.brembs.net/comment-n629.html

        And here an economist: Determinism is tr…ue but thermostats can still control the temperature. And nobody denies that thermostats control the temperature. — Steven Landsburg paraphrasing Robert Nozick in The Big Questions

        Here’s the take of a computer scientist on the topic: “There’s no scientific reason to believe that we have free will. There’s no buffer zone that we’ve found in any of the physical laws of ho…w the universe works to make room for free will. There’s non-determinism; but there’s not choice. Choice is the introduction of something, dare I say it, supernatural: some influence that isn’t part of the physical interaction, which allows some clusters of matter and energy to decide how they’ll collapse a probabilistic waveform into a particular reality.” — Mark Chu-Carroll

        Philosopher: “A man can do what he wants, but not want what he wants.” — Arthur Schopenhauer

        Philosophy in general: http://plato.stanford.edu/entries/freewill/

        • oldfuzz
          Posted July 29, 2010 at 10:26 am | Permalink

          Good exploration of the problem regarding free will’s definition. One of the redcurrant phrases here is “depending on how you define…” which may apply more to free will than religion and species.

          If by free will we mean the ability to choose unconstrained by natural means–which may have been the case in the earliest writings on the subject–science has falsified that idea.

          If by free will we mean the ability to choose one of many possibilities, we have it. In writing a post we choose words to make our point clearly; however, that goal is missed or met and not agreed.

          As for research in neuroscience, the Charlie rose Brain Series offers excellent insights into current brain research. All sessions are at http://www.charlierose.com/view/collection/10702

          For me, the brain defies concise description. Compared to a computer network, which it seems to be more like than a computer, it is a partial mesh with trillions of nodes (neurons) each with a thousand or so inputs (dendrites) and another thousand of so outputs (synapses). We have a good idea what happens at the neuron level, but two problems that confuse the quest for precise knowledge is that no two brains are the same and no neurons in different people have the same DNA (excepting identical siblings which have different experiences, hence different brain structures).

  2. Posted July 28, 2010 at 7:32 am | Permalink

    Perhaps our magic brain fairy lives 10 seconds in the future. Did you naturalists think of that?

  3. Martin
    Posted July 28, 2010 at 7:46 am | Permalink

    The idea that factors outside of conscious thought affect our decisions shouldn’t be all that surprising. We know certain mental illnesses, like schizophrenia, depression, social anxiety, OCD, etc., are to a lesser or greater extent heritable, as is temperament, and these factors influence our decisions. Even if we we make decisions through conscious, deliberative thought, the nature of that thought is influenced by the structure of our brain. Humans display species-typical cognitive heuristics and biases. Although it’s difficult to predict individual decisions, the behavior or large groups of people is statistically quite predictable. Individual brains are still too complex and unknown to be highly predictable, but that’s a statement about our knowledge of human decision making, not the ontological indeterminacy of human decision making itself.

    • Martin
      Posted July 28, 2010 at 7:51 am | Permalink

      What I mean is that if you suffer from something like depression, your conscious thought patterns are going to be quite different from a normo-cognitive person, even your normo-cognitive self, so it should be obvious that our conscious, deliberative thoughts are not nondeterministic and free.

    • Patrick Julius
      Posted July 28, 2010 at 8:58 am | Permalink

      Yes! Subconscious brain activity affects our decisions! That really doesn’t undermine free will, which only requires that conscious brain activity ALSO affects our decisions.

  4. Jack
    Posted July 28, 2010 at 7:51 am | Permalink

    When I have to repeatedly do boring tasks like choosing between buttons that have no effect, I let my co-pilot handle it as well. Breathe now or later? You deal with it…

  5. Jon
    Posted July 28, 2010 at 7:59 am | Permalink

    Why stop at free will? You could say *consciousness* doesn’t exist, it’s just an aggregate of material processes. But it does exist empirically, because we experience it empirically. I’d argue we also experience free will empirically.

    Now free will is a vague term, and it could be flexible how we define it (we could get all Aristotelian, categorize its different aspects, etc.) But that’s as much a matter of personal experience and inherited *culture* as it is science. Terms are useful, or they are not, and they are revised. Free will is has been a useful term… it refers to something socially useful, useful enough to deploy in our legal system, for instance. (We say “of your own free will.”)

    I think it’s a good discussion to ask whether *culture*–which is a result of accumulated experience (ordinary people can be empirical too, not just scientists)–has a value and needs to be weighed versus whatever naturalistic models that scientists manage to construct through experiment and hypothesis. After all, something like the experience of consciousness is directly empirical, while the naturalistic models attempting to explain it are based on inference, deduction, the limited conditions under which an experiment is conducted, etc. It is an indirect picture.

    • Dan L.
      Posted July 28, 2010 at 9:33 am | Permalink

      Jon,

      Have you ever empirically observed a ringing sound that no one else could hear? Now imagine a room full of people, all with tinnitus — and let’s suppose the perceived sounds are slightly different frequencies. If they start talking to each other, it will seem like the sound is real — shared between multiple observers. But when they investigate, they find that they’re all hearing different sounds — the most likely conclusion is that none of them are actually hearing anything, even though they all have the experience of hearing something.

      You have direct, empirical experience of plenty of things that aren’t real. And that’s not even taking into consideration the fact that our empirical sensations are often misleading with regard to the phenomenon that caused those sensations even when they do correspond to something real. I suspect both of these caveats to empiricism are in effect when it comes to free will and consciousness.

    • Torbjörn Larsson, OM
      Posted July 28, 2010 at 9:53 am | Permalink

      We experience movies too, but not then they are shot in real time.

      The simplest model of consciousness that can be predicted from the basis of such research as in the post is that we model what we experience, including our own actions.

      Meanwhile experiments _are_ direct or they would not be experiments. They can be many times removed and use historical observations, but they are direct on the system studied.

      • Jon
        Posted July 28, 2010 at 12:37 pm | Permalink

        No doubt experiments have value, but there are lots of ways to observe and arrive at conclusions, and not all involve physical science.

  6. NewEnglandBob
    Posted July 28, 2010 at 8:34 am | Permalink

    It’s also very short and should be comprehensible to the non-scientist.

    That may be true, but is still requires a payment of $32 to read, even though it is two years old.

    • JeffC
      Posted July 28, 2010 at 9:53 pm | Permalink

      Students who might desire to read this and other articles simply don’t have the funds (eg., $32.00) to purchase them. This has become a real problem in many disciplines

  7. Posted July 28, 2010 at 8:38 am | Permalink

    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.

    I’m not sure that is the right way to say it. We could be looking at the retrieval of the memory content of the brain that weight the decision and not the decision itself. I wouldn’t expect the contents of someone’s brain to radically change during a laboratory experiment. It doesn’t rule out consciousness as the controlling factor; it could be showing that consciousness needs a whole bunch of time to gather up information, or that consciousness “trusts” this particular unconscious brain process to be right and goes along with whatever is dredged up from a person’s memory.

    • Jack
      Posted July 28, 2010 at 8:41 am | Permalink

      “I think that this problem can be solved as long as we do not think of free will as a momentary act. Once we understand that deliberation and decision are processes that are spread out over time, even, in some cases, very short amounts of time, then there is plenty of room for conscious components that are more than accessories after the fact.” – Shaun Gallagher

      • Patrick Julius
        Posted July 28, 2010 at 8:59 am | Permalink

        Indeed!

      • Posted July 28, 2010 at 9:04 am | Permalink

        Thank you. Freewill as a momentary act points back to the very wrongheaded Christian idea of absolute freewill that there is some kind of willpower inside of us that is unbounded by reality and thus not dependent on mechanisms in the brain, otherwise humans would all be molecular puppets of God (O, the horror!) just like dogs.

        I found Shaun Gallagher’s full interview (link) with Michael Gazzaniga, conducted in 1998, where that quote appears in the concluding section.

      • Torbjörn Larsson, OM
        Posted July 28, 2010 at 10:10 am | Permalink

        Possibly, but that still means our experience of consciousness is seriously distorted and deluded.

        The proposal would mean a fantastically complicated mechanism that collected observations of the mind and transformed it to a singular (in time and “space”), coherent consciousness in “real time”.

        Since we already have models of self, it is what happens. But I would argue that it is a only a difference in resolution. “Conscious components” assume a 1-1 map between “conscious components” of mind and components of consciousness, “model of self” assumes minimal (well, roughly, it is biological after all) map to a simple simulacrum.

  8. Kyle Marquis
    Posted July 28, 2010 at 8:39 am | Permalink

    Free will always struck me as an idea not clearly thought through. Seriously, what did people who believe in free will expect researchers to find when they started studying volition? Some kind of…essential…volition oil? A glowing “will orb” that looks like a power-up in a video game? What would “free will” look like, how would we notice it, what effects would it have that we could predict? Even if we found some sort of non-material “brain fairy,” (heheh) how does that do anything more than push the question back one step? What happens when we crack open the brain fairy?

    I can get a more consistent and testable definition of “God” than I can of “free will.”

    • Patrick Julius
      Posted July 28, 2010 at 9:00 am | Permalink

      Well, it’s not well thought-through by incompatibilists and libertarians.

      We compatibilists have thought it through quite extensively.

      Free will is the capacity to direct behavior through rational decision-making.

      This is all one needs for moral responsibility, and it seems pretty obvious to me that we have it.

      • Gordon B
        Posted July 28, 2010 at 9:51 am | Permalink

        If free will is simply the capacity to direct behavior through rational decision-making, wouldn’t a sufficiently advanced AI program or robot have free will?

        I would think so, and while it does not effect whether or not this compatiblist version of free will exists, it does have significant implications with regard to how we view moral responsibility and apply the concept of “free will” in society.

        Would you say that a company that manufactures, programs, and otherwise trains a robot is not at all liable for damage that robot perpetrates if the robot has the right “free will” cognitive modules? If you think this company is liable, wouldn’t you also want to say that parents bear some moral responsibility for damages their children perpetrate (even their grown children with “free will”)?

        • Posted July 28, 2010 at 11:41 am | Permalink

          Yes, another compatibilist! I thought I was the only one.

          Compatibilists are the only people whose definition of “free will” is not tied up with unpredictability, and thus will survive advances in neuroscience that demonstrate that the brain is deterministic.

          By our definition of “free will” (we prefer some states over some other states, and thus choose actions that we think are more likely to lead to the states we prefer), yes, dogs and computer-controlled videogame characters and autonomous UAVs have free will too. They all act so as to optimize some parameters.

          The difference between them (dogs, NPCs, UAVs) and us is that we have an “I”, a sense of self. Illusory and fictitious as it may be, it allows us to create an imaginary locus at which the responsibility buck stops. It’s like a center of gravity. It doesn’t really exist, but it encapsulates a lot of really complicated stuff into an elegant and useful (albeit invented) element.

          To answer your very good question;

          If a person trains a dog, if a hacker programs an AI, if an air force programs a UAV, than the state-preferences of those creatures (dog, AI, UAV) are really just a reflection of the state-preferences of the programmers. The programmers have an “I” (where the responsibility buck stops), the creatures do not, so responsibility trickles up to the programmers.

          Do children have an “I”? If not (very young children arguably do not have an “I”, they can’t recognize themselves in mirrors, etc), then the responsibility does trickle up to the parents. If children do have an “I”, then they could be blamed for doing wrong things out of being selfish (just like a psychopath), but the parents take the responsibility due to their decision (and any consequences thereof) of unleashing their little psychopath out into the world. (And I mean this with no resentment. I was a little psychopath once. We all were. It takes us a few years to work out that our actions can have consequences that hurt others, and that this hurts us as well and ought to be minimized).

          • Gordon B
            Posted July 28, 2010 at 1:10 pm | Permalink

            Well, I would consider myself a hard incompatibilist (though you may have been refering to Patrick), because my preferred answer to the question of whether NPCs, UAVs, dogs, and humans have “free will” is no. I find it quite puzzling that compatibilists would admit to assigning “free will” to systems like NPCs and UAVs because it renders the concept meaningless.

            If every reactive system has “free will”, but only a certain select few instances of these systems can be morally accountable, then we’ve just demonstrated that “free will” effectively has no bearing on the issue of moral responsibility. Why not just drop it? It would certainly do away with the libertarian baggage!

            With regard to the “I” and sense of self, I don’t like the center of gravity analogy because, as far as I know, all physical objects (or collections of physical objects) have a center of gravity, but not all physical objects have a sense of “I”. What are the necessary conditions for a sense of self? When in the developmental history of an individual and when in the evolutionary history of our species did this sense of self emerge? Would a robot that is functionally equivalent to a human have a “sense of self”?
            If we’re going to base our system of ethics around a “sense of self”, we’re going to need to understand it thoroughly! Otherwise, we’re just acting upon our gut reactions as to who or what has moral accountability, which can lead to a lot of inconsistencies and suboptimal results.

            • Posted July 28, 2010 at 1:28 pm | Permalink

              Yes, I guess I was really replying to Patrick when I said; another compatibilist.

              “Then we’ve just demonstrated that free will effectively has no bearing on the issue of moral responsibility”.

              I agree.

              “Why not just drop it?”

              Indeed. Why not? It’s a pretty useless idea. We can all make decisions based on our interests (which often include not being detrimental to others), sometimes we’re just and sometimes not. That’s all we really need to know. The question of free will just makes things messy without adding anything of value.

              “If we’re going to base our system of ethics around a sense of self, we’re going to need to understand it thoroughly!”

              You mean, in the same way/extent that we base it around free will? This comments thread shows that we don’t understand THAT thoroughly! (Incompatibilist determinists don’t even think it exists!). So no, I don’t think we’d have to understand the “I” thoroughly before we say that it, and not free will, is the responsibility buck stopper. Besides, I think the “I” can be understood more meaningfully than free will could.

              “What are the necessary conditions for a sense of self? When in the developmental history of an individual and when in the evolutionary history of our species did this sense of self emerge?”

              All very good questions. Lots of people have been making progress on them from a variety of angles. Do you really want me to summarize their results? This comment is long enough as it is… Anyways: The answers to those questions are, I believe, closely related with the answer to: What does it mean to say that someone is ‘morally responsible’ for something?

              “Would a robot that is functionally equivalent to a human have a sense of self?”

              Yes. Zombies are impossible. If it talks like a self-aware being, it IS like a self-aware being.

              If it’s acting to fulfill its own wants (and not the direct commands of someone else), if it knows what it’s doing and why (and it says it wants to do it), if it can explain “I want… I feel good when…”, then it can take responsibility. A thermostat, dog, UAV, or NPC can’t do that. But a robot someday probably will.

            • Torbjörn Larsson, OM
              Posted July 28, 2010 at 2:39 pm | Permalink

              Hmm. I can see how the compabilist definition racks up the ante another notch (rational choices, not just any choices), but it is essentially the same idea I have used.

              But – it makes it unobservable what I can see!? We can’t define or observe “rational” (reasoned) as such, or others rationality (reasoning).

              I wouldn’t be able to use it.

            • Gordon B
              Posted July 28, 2010 at 4:53 pm | Permalink

              @AtheistSpy: It seems we have run out of room! Though, if you want to summarize the research on cognitive development, feel free to throw it in a parallel thread like this.

              I think we agree on many points. The only other thing I would like to add is that often people have a rather myopic view of where the responsibility buck stops.

              When an airplane crashes, the NTSB spends months investigating the incident, establishing the web of causality that lead up to the disaster, and identifying various parties whose actions contributed to the tragedy. It also makes recommendations as to how to prevent the chain of events from happening again in the future. Society rightfully demands these steps.

              Yet, when a person commits murder or some other criminal act, society usually just writes the individual off as being a monster who chose to commit evil and that’s that. The buck stops there!

              But should it? Or rather, should all of it? I think it is a false choice to suppose that we must put all the blame on the first agent down the chain of causality or else we’re faced with an infinite regress. We could dole out progressively smaller amounts of moral responsibility to agents down the chain of causality. For instance, can’t we stop about $0.50 of the buck on the guilty individual, and distribute the remaining responsibility by figuring out what other morally responsible agents (which in most cases would be parents and other parental figures) could have acted differently to prevent the disaster? This would be akin to assigning the immediate cause of an airplane crash to “pilot error”, but also laying blame on the airline’s policy of overworking their pilots or failing to provide appropriate training.

              People might be repulsed by the idea that they could possibly be thought to have contributed to some heinous crime commited by someone else, but the flip side of it is we also have the ability to be positive influences on others and move them away from destructive paths.

          • Posted July 29, 2010 at 1:25 am | Permalink

            quote:
            Compatibilists are the only people whose definition of “free will” is not tied up with unpredictability, and thus will survive advances in neuroscience that demonstrate that the brain is deterministic.

            It has already been shown numerous times that the brain, like the universe, is indeterministic:

            http://www.nature.com/nature/journal/v466/n7302/full/nature09086.html

            http://www.plosone.org/doi/pone.0000443

            http://arjournals.annualreviews.org/doi/abs/10.1146/annurev.psych.55.090902.141429

            Determinism has been dead since Heisenberg’s uncertainty principle and even deader since chaos theory. I don’t think determinism can get any deader than that…

  9. poke
    Posted July 28, 2010 at 8:39 am | Permalink

    The biggest problem with this, besides the confused talk of decisions “originating” in the brain and so forth, is simply that the notion of being consciously aware of making a decision, in the sense used here, is confused. I can become conscious of having made a decision years after the fact. I could, for example, realise I married the wrong person and become conscious of my decision to do so for reasons other than love years or even decades after having done so. The point being that being conscious of having made a decision is not a necessary criteria of having made a decision at all, let alone being conscious of it at the time of making the decision.

    The concept they have appears to be similar to Libet’s, that a voluntary action is one that coincides with a feeling of acting voluntarily, but that’s simply not true. We act voluntarily all the time without a corresponding feeling of having acted voluntarily. The closest thing to acting on a feeling would be to have a sudden urge or an irresistible impulse to do something, the voluntary nature of which is questionable. Bennett and Hacker’s “Philosophical Foundation of Neuroscience” contains an excellent discussion of this topic.

    • Posted July 28, 2010 at 9:18 am | Permalink

      Poke, what is it with you and your denial that the brain is the originator of decisions?

      • Jon
        Posted July 28, 2010 at 9:31 am | Permalink

        It’s funny. Sometimes my wife chides me for not making decisions with my brain, then the next minute for making decisions too much *with* my brain.

        It can be very confusing.

        • Kevin
          Posted July 28, 2010 at 9:36 am | Permalink

          I often make the observation “nature gave men two heads, but only enough blood to run one at a time”…

          Perhaps this is what your wife is complaining of?

          • Jon
            Posted July 28, 2010 at 9:43 am | Permalink

            Actually, it applies to a lot of things… We do have a number of glands. The brain is just one organ.

      • Dan L.
        Posted July 28, 2010 at 9:38 am | Permalink

        I didn’t read this as denying that decisions are made in the brain. I read it as pointing out that one can choose a course of action without understanding the reasons one chose that course of action until much later (or ever in many cases). In which case, the course of action was, in some sense, not voluntary — some mysterious, shadowy part of your brain pushed you towards it without you even being aware.

        I think it’s a great observation, and rings very true to my subjective experience of decision making.

        • Posted July 28, 2010 at 10:01 am | Permalink

          I was responding to the first line:

          The biggest problem with this, besides the confused talk of decisions “originating” in the brain…

          Maybe the quotation marks around the word “originating” did not mean what I thought they meant. Nevertheless, the unconscious pressures that guide our decisions still originate in the brain.

          • Dan L.
            Posted July 28, 2010 at 10:09 am | Permalink

            Adrenal glands, testicles, ovaries, liver function, pancreas, stomach contents, blood sugar, and thyroid function all influence decision making and none of them are parts of the brain — though I’m not sure whether this is what poke was driving at. Poke might also have been trying to say that talking about decisions “originating” is what’s problematic in the first place; “originating” might just seem to poke to be a misleading way to talk about decisions the same way that talking about electromagnetism as a fluid in the 18th century prevented researchers from hitting on the idea of charge. Poke might just be worried that the notion of decisions “originating” commits us to certain, possibly false, notions about decisions.

            At any rate, I’m sure we’d all agree that without the brain, there wouldn’t be any decisions.

            • ritebrother
              Posted July 28, 2010 at 10:33 am | Permalink

              But the point is that the brain is the locus where the signaling molecules produced by those peripheral glands and organs are have their effects in driving CNS-mediated responses. The causal stimulus and signaling agent may be peripheral, but the proximate response (i.e. “decision”, by whatever definition) is made. Although I agree that “decision” is too anthropomorphic to describe many of these processes (e.g. regulation of feeding behavior by peripheral satiety signals via the hypothalamus).

            • ritebrother
              Posted July 28, 2010 at 10:35 am | Permalink

              Missing words: …proximate response is made IN THE BRAIN.

            • Posted July 28, 2010 at 10:38 am | Permalink

              At any rate, I’m sure we’d all agree that without the brain, there wouldn’t be any decisions.

              This is what I’m not sure poke believes to be true.

              talking about decisions “originating” is what’s problematic in the first place…

              If that is what poke meant, then I have to say I really don’t think that is an issue at all. One could always go that route with just about any process. Look at the life of a single human. We could start counting how old she is from the day of her birth or say that she originated from conception after her parents mated, or we could extend what we mean by her origination to the evolution of humans 2.4 million years ago or say she originated 3.5 billion years ago with the start of life on Earth or push it back even further to the formation of the elements she is made from.

            • poke
              Posted July 28, 2010 at 11:29 am | Permalink

              Yes, it’s odd to talk about decisions originating anywhere. The best sense I could make of “the origin of a decision” would be the reason for the decision. “I took the bus because the train was delayed” might give the origin of my decision to take the bus, for example. I don’t think people usually talk that way though.

              But, yes, I absolutely do agree that you need a brain to make decisions.

      • poke
        Posted July 28, 2010 at 11:24 am | Permalink

        It’s exactly the same as I was saying in the other comments. People, and not brains, make decisions. That doesn’t mean you can make decisions without a brain. Clocks tell the time but you can’t locate the ability of clocks to tell the time in one part of the clock. The brain is part of a person who makes decisions but it’s erroneous to say the brain makes decisions.

        Saying decisions “originate” in the brain is likewise confusing. You need a brain to make decisions. Certain events in the brain are causally correlated with making a decision. Without those brain events there can be no decisions. But it doesn’t follow that the brain event is therefore the decision itself or the origin of the decision.

        • Dan L.
          Posted July 28, 2010 at 11:42 am | Permalink

          Replying to ritebrother here because I can’t reply directly and what poke says here is relevant:

          I’m with you in that putting the focus on non-cortical contributions to decision making elides the importance of the brain in decision making (it’s the locus, as you say). But I think that’s what both poke and myself are driving at, that the focus might be too much on the brain as things stand. This is one of my big problems with substance dualism, for example: a disembodied mind that doesn’t feel hunger, lust, anger, sorrow, etc. isn’t really a human mind, and all of those sensations causally implicate non-cortical tissues. Like poke says, it’s persons, not brains, that make decisions.

          @Aratina Cage:

          I understand that sometimes we need to relax our usage of language because our causal intuitions are misleading when it comes to scientific causal principles. But I believe there’s a real danger here of reifying metaphors and other constructs we use to get a handle on phenomena, as opposed to using them to understand the phenomena themselves. Out intuitions are often mistaken, and it seems only more likely that this would apply to the brain. In this, poke and I seem to agree.

          So while I see why you might take what poke said as an endorsement of substance dualism, I think it’s clear at this point that this is not the case and poke’s objection has more to do with avoiding “theory-laden observations” or whatever you would like to call them.

          • ritebrother
            Posted July 28, 2010 at 12:01 pm | Permalink

            Dan L: I see you point now, and I agree entirely – thanks for clarifying.

            • ritebrother
              Posted July 28, 2010 at 12:03 pm | Permalink

              YOUR point. (Sheesh – It’s a bad finger day.)

        • Posted July 28, 2010 at 12:44 pm | Permalink

          Well, that is more understandable and, gladly, not dualist, but I still disagree with your contention that it is erroneous to say that brains make decisions (and I don’t know where you are going with the clock analogy, sorry).

          So, why stop at the body in the decision making process? Even more disconnected things such as ambient temperature and the amount of light in the area influence decisions. The brain, though, is the device that transforms all the influences from the rest of the body, the environment, and its own memory into the outcome that we call a decision. Nothing else in the body does that.

          Your confusion over the use of the word “origination” with respect to decisions also seems to stem from taking in a view of the process that is arbitrarily too broad over time and space. I think we should focus on the process that transforms the vast array of factors into an action when we talk about the origin of decisions, and that would be a brain process.

          • poke
            Posted July 28, 2010 at 3:02 pm | Permalink

            Well, by calling everything outside the brain an “influence” you’re presupposing that it’s the brain that makes decisions. I’m not saying that the body influences decisions or that we use our bodies to make decisions. Decision making is an ability and doesn’t need to be located in a particular part of person’s body or in their brain. That’s the point of the clock analogy or the car analogy I gave before. A clock can tell us the time but we cannot locate that ability in some part of the clock. You car might be capable of driving at 100 mph but you can’t locate that ability in the engine or the wheels, even though both the engine or the wheels are required for driving at 100 mph. An engine alone cannot drive at 100 mph just as a brain alone cannot make decisions.

            Obviously the brain is very important to our ability to make decisions, just as the engine is very important to the ability of a car to drive at 100 mph. You could even say the brain is the most important part of our body when it comes to decision making. I can lose an arm or a leg or my eyesight or hearing without it affecting my ability to make decisions whereas certain types of brain damage might dramatically affect my ability to make decisions. All these things can be true without us having to locate decisions within the brain.

            • Posted July 28, 2010 at 8:46 pm | Permalink

              Decision making is an ability and doesn’t need to be located in a particular part of person’s body or in their brain.

              What is stopping you from taking the environment into account? Why stop at the body? Really, you are being ridiculous. The mechanism making the decision is in the brain, not anywhere else.

              I can lose an arm or a leg or my eyesight or hearing without it affecting my ability to make decisions whereas certain types of brain damage might dramatically affect my ability to make decisions.

              You think? Maybe it is because decisions are brain processes.

            • poke
              Posted July 29, 2010 at 6:37 am | Permalink

              Aratina, since I can’t reply to your post I replied to my own: I’m not stopping at the body. People make choices. People have bodies, yes, but their bodies do not make choices anymore than their brains do. The ability to make choices is dependent on the circumstances as much as it is the body (or the brain), so obviously it takes the environment into account. You can’t make a choice if you only have one option available to you, for example, and the choices available are given by the circumstances.

              Furthermore, since decision making isn’t a mechanism it obviously cannot be a mechanism in the brain. Decision making is an ability of the whole person and not a mechanism nor a brain process. These are essentially truisms. I decided to buy new shoes today, not my brain, for why would a brain need shoes? I am not a brain piloting a body. That is Cartesian nonsense. Popular nonsense, sure, but nonsense nonetheless.

            • Posted July 29, 2010 at 7:26 am | Permalink

              poke 6:37 a.m.,

              decision making isn’t a mechanism

              Decisions are the outcome of a process in the brain. While there are external factors to the process, it doesn’t make sense to talk about decisions happening outside the brain.

              I decided to buy new shoes today, not my brain, for why would a brain need shoes?

              You are doing it again, placing a boundary on the decision-making process that is far too wide, making it almost obscure. You could just as easily ask yourself, “Why would a body need shoes?”, and come up with some seemingly deep — but wrongheaded — answer.

              I am not a brain piloting a body. That is Cartesian nonsense. Popular nonsense, sure, but nonsense nonetheless.

              Explain how that is dualist, if you don’t mind. To me, it doesn’t match the criteria for dualism at all: there is no ghost in the machine but instead a community of groups of living machines that are interdependent on each other for survival, one group of which happens to be the brain. What makes the brain different from the rest is its structure and composition, not some mysterious power or distinct essence.

            • poke
              Posted July 29, 2010 at 8:24 am | Permalink

              Aratina, It’s not dualist but it is Cartesian. The Cartesian conception of the mind locates all our mental and intellectual abilities in the mind due to Descartes’ radical skepticism. Before Descartes our various mental capacities (for thought, decision making, etc) were construed as abilities of the whole person, in keeping with our ordinary usage of psychological terms (and perfectly compatible with a naturalistic conception of the world). Modern materialism identified the capacities of the Cartesian immaterial mind with the brain and mixed in some additional Empiricist confusions. Thus, we have contemporary psychologists and neuroscientists claiming that what we see is not the world but rather a construction of the brain and identifying the brain as the seat of thought, decision making, feeling, sensation, etc. They give the brain exactly the same properties as Descartes gave his immaterial mind.

              There is, obviously, a process or mechanism within the brain that is needed in order for me to make decisions. This, I think, we both agree on. What I’m objecting to is identifying this process or mechanism with decision making. To get back to the experiment I was commenting on. We could decide, arbitrarily, to call this brain process “decision making” and its successfully completion “having decided” and so forth. If we do this then we can say, “The brain makes a decision before we are aware of having made a decision.” This sound profound and is confusing. It’s also misleading because there are two uses of the word “decision” here; the first use is as a technical term that singles out the completion of a particular brain process, the second is its usage in ordinary language.

              My comments about how brains don’t need shoes, and so forth, may seem simplistic but the idea is to show that there is an ordinary usage for the word “decision” that does not fit with your usage of the word to single out the completion of a process in the brain. I buy shoes, I don’t buy shoes for my body’s feet, as I would have to if I was a brain occupying a body. I don’t decide to buy shoes for my body’s feet as I would if it was my brain that made decisions about my purchases. All I would like here is for you to consider that the ordinary usage of the word “decision” does not fit with decisions being brain processes but rather with decision making being an ability of a whole person who has feet and wears shoes.

              Now, given that we have two potential usages of the words “decision”, “decision making”, etc, does it make sense to call the process in the brain decision making? I don’t think it does. If instead of saying “the brain makes a decision before we are aware of having made a decision” I say “the cluster of neurons necessary for decision making fires before we are aware of having made a decision” it sounds mundane, even obvious, and is not misleading at all. The same problem does not come up here. Indeed, the problem would not come up if I simply remembered that I was using “decision” in two different ways in the original sentence. But the point is: Why call this brain process “decision making” if it’s going to lead to such confusion?

              There’s a deeper issue though. Some scientists and researchers don’t simply want to label the brain process “decision making” in this misleading way, they typically go on to talk as if the brain process is the real decision making and that our ordinary talk of decision making therefore rests in an illusion. It’s really (part of) our brains that are deciding for us and we’re simply confused about making the decisions ourselves. (Just as, supposedly, what we see is a construction of our brain rather than the real world because seeing is a brain process.) But this is nonsense and an example of why we should be careful how we use words. You can’t identify brain processes with concepts, such as decision making, that it only makes sense to apply to the whole person. The result is confusion.

            • Posted July 29, 2010 at 12:07 pm | Permalink

              Last one and I’m done on this thread.

              poke 8:24 am,

              It’s not dualist but it is Cartesian.

              I think you are mistaken. From the Stanford Encyclopedia of Philosophy’s entry on dualism:

              Descartes was a substance dualist. He believed that there were two kinds of substance: matter, of which the essential property is that it is spatially extended; and mind, of which the essential property is that it thinks….Bodies are machines that work according to their own laws. Except where there are minds interfering with it, matter proceeds deterministically, in its own right. Where there are minds requiring to influence bodies, they must work by ‘pulling levers’ in a piece of machinery that already has its own laws of operation.

              So you see, the Cartesian philosophy of mind is a form of dualism.

              Before Descartes our various mental capacities (for thought, decision making, etc) were construed as abilities of the whole person

              And following that form, why not abilities of the whole village or the entire human species or greater? Again, why stop at the body when you can extend these mental capacities further?

              Modern materialism identified the capacities of the Cartesian immaterial mind with the brain and mixed in some additional Empiricist confusions.

              You have just admitted you knew you were wrong when you said that Cartesian philosophy of mind was not dualist. Yes, it was dualist. And what are these Empiricist confusions you speak of?

              They give the brain exactly the same properties as Descartes gave his immaterial mind.

              What is the problem with that? Immaterialism is unnecessary to explain what we know about minds.

              If we do this then we can say, “The brain makes a decision before we are aware of having made a decision.”

              You are failing to distinguish between the unconscious processes in the brain and the conscious processes in the brain, and you are ignoring the problems with the (properly stated) conclusion as stated by several people elsewhere in this thread.

              I buy shoes, I don’t buy shoes for my body’s feet, as I would have to if I was a brain occupying a body.

              What about the unconscious part of that decision, the part that this study detected? You might be buying shoes for a reason (or an unconscious brain process) you will never know.

              All I would like here is for you to consider that the ordinary usage of the word “decision” does not fit with decisions being brain processes but rather with decision making being an ability of a whole person who has feet and wears shoes.

              I am not one who thinks that the mind is separate from the brain. Ordinary usage of the word “decision” may not reflect the behind-the-scenes unconscious work that goes into making “a decision” but they are one and the same.

              I say “the cluster of neurons necessary for decision making fires before we are aware of having made a decision”

              Who do you mean by “we”? Who are the ones being made aware of having made decisions? Wouldn’t it be better to say that a specific brain process called consciousness becomes aware of having made a decision?

              Some scientists and researchers … typically go on to talk as if the brain process is the real decision making and that our ordinary talk of decision making therefore rests in an illusion.

              Illusion meaning that it decieves our thinking about why it happened, not illusion meaning it didn’t really happen.

              what we see is a construction of our brain rather than the real world

              I don’t know about you, but when I look at a nail, it doesn’t fly into my eyes and get lodged in my brain.

            • poke
              Posted July 29, 2010 at 3:43 pm | Permalink

              Aratina, There was more to Descartes philosophy than dualism. “Cartesian” is not equivalent to dualist just as, say, “Marxist” isn’t equivalent to dialectical materialist. Dualism was part of Descartes philosophy but not the whole deal and not the part I was referring to. Perhaps I could have been clearer. The way you’re talking about consciousness, for example, is also Cartesian. The point being that you’re using ordinary language in a specialised (revisionist) way that is novel to Descartes philosophy and those inspired by him, even though you’ve (rightly) rejected his dualism.

              In order to make sense of his philosophy Descartes had to give his immaterial mind certain special features, such as that the mind (so construed), rather than the whole person, is the bearer of psychological attributes. Modern materialists have taken this odd construction and simply identified it with the brain. The result being that while they are not dualists, they are still committed to a number of erroneous ideas courtesy of Descartes.

              I wouldn’t attribute psychological states to a whole village because a whole village is not the bearer of psychological states. It is clearly the case that the whole person is the bearer of psychological states, as I have demonstrated repeatedly by using psychological concepts in the way in which they are employed in ordinary language. Villages do not decide to buy shoes, unless you mean that every person in the village decided to buy shoes. You seem to be missing the point. I’m not using these concepts in arbitrary ways. These concepts, in ordinary language, refer to the abilities of whole persons and not those of their brains and not those of villages. This is obvious and I won’t attempt to demonstrate it further.

              I suggest reading either MR Bennett and PMS Hacker’s “Philosophical Foundations of Neuroscience” or Anthony Kenny’s “The Metaphysics of Mind.” They express these ideas much more eloquently than I can and in much more detail. They are really worth grappling with, even if you do ultimately conclude that they are wrong, or if you haven’t found my presentation convincing. At the very least, they provide a better foil for materialism than dualism does, and one that’s completely in keeping with a naturalistic worldview.

            • Jon
              Posted July 29, 2010 at 3:49 pm | Permalink

              “Again, why stop at the body when you can extend these mental capacities further?”

              This is dualistic to me–the point is that our “mental capacities” aren’t separate from the body to begin with, and that our sense of the body is essential to “extending our mental capacities.”

              “I am not one who thinks that the mind is separate from the brain.”

              One thing I think we get from Cartesian dualism is that mind is the brain. Why does it have to be? Why can’t other parts of the body include what we experience as the mind? (Like the glands I mentioned above and nerve endings associated with them.) The problem with this is that this conflicts with the computational model of the brain–as if the brain contains the entire setup for our conscious experience. But that’s not from empirical observation, I don’t think. And it gives us a sense of the mind as like a mechanism–something I’m frankly skeptical of. I think it’s a lot squishier and less cut and dried…

              If we look at the work of a neuroscientist like Alan Baddley, some parts of cognition do appear to operate like a mechanism–but the nearer you get to an “executive”, the more handwavy things become. At one point he calls the executive a “ragbag” of things he can’t explain. That’s a pretty important piece of the puzzle left unresolved, almost making you laugh: “I’ve discovered everything about thinking, except consciousness.” (I’m laughing with him, not at him–what he accomplished is non-trivial.)

              The body seems to me essential to personhood and mind, not accidental–it’s an essential experience in our subjectivity. It makes it possible for us to empathize and share experience with other persons, which begins with our parents (hence the importance of this relationship that early psychologists like Frued placed on it). In this sense peoples’ minds *are* very different from brute matter, because it’s how humanity and culture gets poured into us. (I’ve heard one of the reasons why German philosophy is so different than the English tradition–besides England’s characteristics that favor individualist “common sense” inventor types like Newton–is because words can be coined like “humanityandculture” as one word and they don’t sound clunky.)

              But having things be this irreducably subjective bugs people like Daniel Dennett, who feels we can blitz past Geisteswissenschaften and make everything all Naturwissenschaften and evo psych (which conveniently puts ubernerds like himself in the vanguard)…

    • Torbjörn Larsson, OM
      Posted July 28, 2010 at 10:28 am | Permalink

      The biggest problem with this, besides the confused talk of decisions “originating” in the brain and so forth,

      But that is what experiments such as in the post tells us, we can see them originating in the brain and lead to to the actions decided.

      is simply that the notion of being consciously aware of making a decision, in the sense used here, is confused.

      There is a lot of conflation in these mechanisms, sure.

      That is one reason why I define “free will” as the simpler “making choices”. That can be as simple as a bacteria reaching a threshold for a sensor activating motility. That “choice” is simple process, yet it means making a decision (go/not go) and affecting an outcome.

      When you use “decision” instead of choice, you start to drag all sorts of social constructs into it. What you describe as a decision was actually the circumstances surrounding a choice.

      For the purpose of the experiment “decision” was suitably well defined. (I haven’t read the paper, but the post describes the circumstances well enough.)

      • Dan L.
        Posted July 28, 2010 at 11:47 am | Permalink

        But that is what experiments such as in the post tells us, we can see them originating in the brain and lead to to the actions decided.

        Umm, we see a neural correlate to the decision-making that precedes it in time. Is that really enough to conclude that it’s the origin or cause (either one, I’m not trying to conflate the terms) of the decision/choice (your preference here, I don’t see that a clear distinction has been made)? Or are we using language somewhat loosely because we don’t have a vocabulary to talk about this kind of thing rigorously?

  10. KP
    Posted July 28, 2010 at 9:31 am | Permalink

    There is also a related article in the July 2 issue of Science. Custers R. and H. Aarts. 2010. The unconscious will: How the pursuit of goals operates outside the conscious awareness. Science 329:47-50.

  11. Kevin
    Posted July 28, 2010 at 9:32 am | Permalink

    This supports what I said yesterday. The higher-level suite of cognitive skills that involves decision-making does not necessarily require conscious awareness to be operative.

    And this says something about free will how, exactly?

    I can envision many scenarios where this ability would provide a positive survival benefit even aside from awareness of potential predators. But even that single example would be more than enough for it to be evolutionarily useful enough to be passed down from generation to generation, favoring those with the ability contra those who didn’t.

    Sometimes, we call this process “intuition”. Making a decision in the apparent absence of deliberate (self-aware) thought.

    You do this all the time, every day. Anyone who is an experienced driver makes un-self-conscious decisions about changing lanes, accelerating, placing the foot over the brake, and shifting gears for those of us who insist on driving manual transmission cars. If we had to wait for our conscious brains to make all of those decisions each and every time, we’d never be able to drive more than 20 miles an hour without whomping into a bridge abutment. You certainly wouldn’t be able to drive and talk on the cell phone (which you can’t anyway, so don’t tell me YOU can), or listen to the radio, or talk with your companions.

    Another example is touch-typing, which I am engaging in right now. I don’t “decide” which keys to press consciously. My fingers “know” that I want to type the word “predestination”, and go off on their merry ways without me having to think “where the heck is the ‘p’ key” each and every time I wish to use it.

    Again, this says nothing about “free will”. Your brain is making the decision freely and without any sort of pre-determination; it just doesn’t require the self-aware portion of the brain we call “consciousness” in order to operate.

    Any of us who are fans of the “Inner Game” series of books (Tennis, Golf, Music) will see Self 1 as the self-aware “checking” portion of the brain and Self 2 as the workhorse which is really doing the heavy lifting. Self 1 takes credit for the good decisions, and berates itself for the bad ones, when most of the time, it’s not even involved in the process.

    But there’s no predetermination involved, and most certainly (which I’m sure you’ll agree) no external agent (ala Cartesian dualism) driving the process.

    I think, therefore I am…but most of my thinking doesn’t need to be in the front of my brain. Otherwise, I’d never get anything done.

    • Torbjörn Larsson, OM
      Posted July 28, 2010 at 10:50 am | Permalink

      Good, I was about to mention another related point: not sure if it was good science, but I seem to remember that one can possibly observe that making decisions “on the spot” in stores makes for more satisfactory buys.

      This is fraught with problems of analysis, but it could be that unconscious or at least not heavily deliberated but emotionally decided actions are also _objectively_ better.

      For this example, our emotions are likely adapted to, if not stores, what people and situations we trust, and what items we would like to collect for our living way back in deep time. All under time pressure.

      I feel, therefore I am!?

  12. Eric MacDonald
    Posted July 28, 2010 at 9:33 am | Permalink

    What is making the decision, if not our own conscious selves?

    Is this a scientific question? Or is it a conceptual/philosophical one?

    Dennett titled his first book on free will: Elbow Room: The Varieties of Free Will Worth Wanting. And in his The Problem of the Soul Owen Flanagan lists 10 different features which he thinks comprise what we mean by free will (144), and he claims to have dealt with them, that is, provided for all the characteristics that are worthwhile about the notion of free will.

    Free will as understood in Christian and other religious traditions is a very strange idea altogether, and perhaps we are misled by this strangeness. It is as though we were like gods, the absolute originators of our actions, with no causal antecedents whatsoever. But we know that can’t be true, because we expect people to act similarly in similar situations.

    I guess my question is whether the Soon et al. paper has far reaching consequences for all concepts of free will, or only for the odd notion that we are absolute originators of actions? After all, we know, even without fMRI scans, that our ‘decisions’ to do certain things precede our conscious thought about doing them. Think about catching something that has just fallen from the table. It seems remarkable that we should have been successful, if we had to have the thought first, and then made the conscious move to catch it. So, we are obviously linked causally, interactively, in many ways, with our surroundings. It could scarcely be otherwise. We’d never have survived or been so successful as a species.

    However, it does not follow from this that there are not situations in which a whole host of counter-factuals might well have been true, depending on how we react to a situation or make ‘choices’ within it. For many of those, the likelihood of them having been true is very small, since most people act consistently, and give reasons for their actions. Are all those reasons simply rationalisations? Or do reasons and causes sort themselves out in a way that it is not simply a case of post hoc ergo propter hoc?

    I think the question about what is meant by ‘free will’ is probably more important than conclusions drawn from experiments like Soon’s (et al.) acknowledge, although they do not draw them. But, as I say, I speak from ignorance, and probably should read the books that have been sitting on my shelves all these years!

    • Kevin
      Posted July 28, 2010 at 9:48 am | Permalink

      Yes. A major problem with discussing “free will” is the extremely heavy theological load that comes with the term.

      After all, it was invented to try to describe why humans commit evil acts contra god’s will. If a kind and loving god wants everyone everywhere to end up in heaven, and if certain acts consign a person to hell, why not just impose a supernatural brake on our behaviors? For an omnipotent god, this would be a trivial thing. A mere snap of the fingers.

      Free will is the ultimate theological dodge around this apparent limitation of an omnipotent god.

      It does, however, pose problems for an omniscient god. Because if you’re “free” to make a choice, but that choice is pre-known (and therefore predetermined) by god, is it really “free” at all?

      Anyone who supports the theological concept of free will cannot support the theological concept of an omniscient god without Grade A, USDA Prime cognitive disconnecting.

      Not that they don’t try…oh, how they TRY!

  13. ennui
    Posted July 28, 2010 at 9:47 am | Permalink

    I always choose to be who I am,
    And self-define my personality
    Thru fits and storms of rationality,
    You free will deniers be damned!

    Tho’ chemical pathways sometimes loom large
    And genetic expression will have its say,
    Hormones, receptors, and physics give way
    ‘Cause they know I am in charge.

    We are only electric meat to some
    Who greedily want to reduce the issue
    To just environment and tissue, but
    They couldn’t be more dumb:

    Homunculus is not some passive rider–
    It is I who am the DECIDER!

    • Jon
      Posted July 28, 2010 at 9:58 am | Permalink

      “It is I who am the DECIDER!”

      I take it that’s a reference to the “white knuckle drunk” who was our previous president…

  14. Tacroy
    Posted July 28, 2010 at 10:07 am | Permalink

    It took people a full ten seconds to press the left or right button? What the hell is wrong with their reflexes?

    As an occasional player of “twitch” games, I might have slightly better reflexes than most – but still, taking ten seconds to press a button is ridiculous. I know from experience* that it should take less than a second to press the button – your fingers are right on them!

    Maybe I should actually read the paper.

    *I did this sort of MRI study once, it was pretty cool and I even ended up taking a nap in there. You just have to keep a tight reign on your claustrophobia.

    • Neil
      Posted July 28, 2010 at 10:35 am | Permalink

      My thought exactly. If a ten second delay between neuronal activity and conscious decision making were required, we’d all be dysfunctional.

      • Torbjörn Larsson, OM
        Posted July 28, 2010 at 10:40 am | Permalink

        But it wasn’t required, it was observed. One way that could happen was if consciousness is modeled so that it solidifies ~ 10 s after our actions. It is constantly rewritten until our actions and observations suits our model of self.

        • Neil
          Posted July 28, 2010 at 10:56 am | Permalink

          I know all that. My point was that the observation is peculiar and not general. My guess is that the bored subjects were half way to that state we’ve all experienced where we’ve driven home from work, lost in thought, and cannot remember ever being conscious of making crucial driving decisions.

          • Posted July 28, 2010 at 11:33 am | Permalink

            Agreed with Neil.

          • Torbjörn Larsson, OM
            Posted July 28, 2010 at 1:30 pm | Permalink

            It’s the one observation we have, so how can it be peculiar (and not general vs what we have)?

            Again, how would we be able to tell if the “self” model is postdated? Sometime it would converge fast, say on scale of tenths of seconds. Sometimes it could well take tens of seconds before all neurons are done with the event in question. We would see no difference, it would all seem instantaneous afterwards, as if we made the decision ‘now’.

            In that light it would be impossible to denote peculiar or general merely by introspection.

            • Neil
              Posted July 28, 2010 at 5:53 pm | Permalink

              Absent an MRI machine, a quick game of Simon Sez is enough to convince me that there is something wrong with the 10 second finding.

    • qbsmd
      Posted July 28, 2010 at 12:16 pm | Permalink

      I’d be interested to see the decision making occur in the context of something faster as well as less random. Maybe MRI two people playing tic-tac-toe or a made up game with similarly simple outcomes.

      • Tacroy
        Posted July 28, 2010 at 12:31 pm | Permalink

        I would be really interested in the outcome of that, actually – put people in an MRI machine, and have them play speed tic-tac-toe against a computer (or another person!)

        How are the outcomes different from just pure “match this against this”?

  15. Posted July 28, 2010 at 10:12 am | Permalink

    Personally, I don’t see why the definition of will must be constrained to only our conscious decisions.

    If you broaden your view of will to include one’s non-conscious brain activity, as I think you properly should, the challenge to free will in light of these studies fizzles.

    Furthermore, these experiments are not predicting decisions based on a computer model of the individual’s brain and a simulation of stimuli to it, they’re predicting decisions by looking at the brain activity itself and forming a prediction based on watching the brain activity at specific spots deemed likely to represent a decision for action. The interesting part is that the decision happens before the subject is conscious of it, not that the machines can predict the subject’s behavior based on a scan of the subject’s own brain. That’s no more notable than my prediction that a drag racer will launch his car when the green light on the Christmas tree turns on.

    There may still be a valid discussion about free will in a deterministic or stochastic universe, but that is a separate question from non-conscious decisions made in one’s own brain.

    • Torbjörn Larsson, OM
      Posted July 28, 2010 at 10:37 am | Permalink

      Without having definition of “free will” from you, it is impossible to know what you mean by “the challenge to free will”.

      Coyne see no challenge to free will, as he defines it so these experiments test the prediction there is none. (See the previous post.) Similarly I don’t see no challenge to free will here, because my definition (“making choices”) predicts that both conscious and non-conscious acts can incorporate it.

      What is the challenge?

  16. Jeremy Manier
    Posted July 28, 2010 at 10:27 am | Permalink

    This also reminds me of the “split-brain” experiments where a subject will reach for an object as instructed with one hand, even though he/she isn’t verbally “aware” that the action is happening. It happens even though the person can’t articulate any underlying volition.

    One question in that case and the one Jerry describes is whether you can separate awareness and will. I’m pretty sure you can.

    A somewhat bigger question is whether the reasons that we think drive our actions are real, or just cover stories created after the fact. For example, I’m pretty sure a major reason I didn’t attend law school is that my sister said to me, “I just don’t see you as a lawyer.” It made me think, and I decided she was right. It seemed like a perfectly deliberate and rational process. But if in fact some non-conscious, constrained thing inside me made the decision, I’d want to understand that better. Right now I don’t see how that would work.

  17. Pliny-the-in-Between
    Posted July 28, 2010 at 11:28 am | Permalink

    I guess this really isn’t that surprising to me at least from a logical sense as others have observed. It stands to reason that action responses to stimuli would evolve before what we call consciousness. And there is no particular reason to think that these more automated responses would have to be wired into the same parts of the brain that manage the more analytic areas.

    It would just seem to support an evolutionary model of cognition. Not exactly the age of enlightenment model we might wish but we need to stop mystifying what we are and start approaching our strengths and weaknesses with a more biological perspective.

    Thanks for sharing this with us.

    • Torbjörn Larsson, OM
      Posted July 28, 2010 at 2:06 pm | Permalink

      Exactly! I’m reminded of a paper that I’m enamored of, that shows how a physiological neuron net model of the prefrontal cortex self-organizes symbolic thinking. The cortex is of course a later brain trait.

      It was (~2003, IIRC) the first time a neuron net was able to do that.

      Ordinarily the whole net participates in learning features on “training sets”. This topology easily becomes over-trained. Add more training and the efficiency goes down (!) because the feature set it triggers on becomes fuzzy, the net makes more mistakes in classifying patterns. Which in turn depends on that training adjusts the weights of _every node_ in a complicated and messy fashion, which can only do so much.

      The new topology was, spontaneously, mapping “near” features to “near” nodes. More training means more efficiency until it eventually hits a plateau, precisely as for us.

      And that localization is analogous to a symbol, something that represents a set. Add members to a specific “training set” and they are added to a specific set of nodes in the form of weight adjustments – they belong to a certain “symbol”.

      Not consciously representing a set, mind you. But if our brain is basically using symbolic thinking, self-organized to boot, it is easy to think that a conscious development of the same (say, in a model of “self”) is close at hand. That “only” had to map to what was already taking place in the substrate, the cortex processing.

      Or, if an exact mapping didn’t happen, at least it shows symbolic thinking is easy and can self-organize, it is “natural”.

  18. Posted July 28, 2010 at 11:28 am | Permalink

    I proposed on another post a similar experiment which proves that free will definitely exists (whatever the result is) :

    http://whyevolutionistrue.wordpress.com/2010/07/27/new-york-times-to-readers-of-course-you-have-free-will-2/#comment-35862

  19. Posted July 28, 2010 at 11:38 am | Permalink

    I read the original work and the subsequent froth regarding it. The fundamental research defines the concept of the decision based on 1) the awareness of the decision and 2) the act. As we know, eyewitness testimony is the least reliable evidence. Consequently, the person’s “awareness” of the decision is questionable.

    As to whether the act constitutes the decision (made up to 7 seconds prior to the act) is also questionable. The facts are inescapable: the researchers were able to predict with good accuracy the resulting act up to 7 seconds prior to the act. That is very interesting.

    But does it constitute the advent of the “decision”?

    Humans act from a myriad of source material, some conscious, some unconscious. I could easily building a bias circuit based on input where the bias becomes evident a fair time prior to the time of action. Does the detection of the bias in the circuit constitute a decision?

    I’d argue not.

    A corollary to the experiment would be very interesting. The researchers could flash a sign at various times prior to the predicted time saying “DON’T PUSH THE BUTTON!”. The number of people who still push the button, and how the time varies, would tell whether or not a decision has actually been made.

    Without a confounding component to the experiment the data as it stands only details possible prediction. Nothing is said about a “decision” at all.

  20. MoonShark
    Posted July 28, 2010 at 12:23 pm | Permalink

    I wasn’t so sure about those previous “free will” threads, but this one seems to have a lot of thoughtful comments all around. Thanks folks, your inquisitive minds are keeping mine entertained :)

    • whyevolutionistrue
      Posted July 28, 2010 at 12:27 pm | Permalink

      And edified!

  21. Posted July 28, 2010 at 2:13 pm | Permalink

    Although I have not examined this study beyond a superficial glance and thus can be horribly wrong in everything I say, it may share the same methodological limitations as the Libet et. al. study may have had in that

    * it may be presupposing that there exists a certain time t (for the subjects side) where everything comes together and consciousness (or conscious decision) happens i. e. Cartesian materialism, which may be debatable.

    * may not actually differentiate between the time when X consciously decided to do Y and the time when the brain X experiences the model of what it just did. It may have assumed that the latter is identical to the former.

    * may not have differentiated between the following to situations: (1) person X decides at time t1 to act at a later time t2, then acts at time t2 (2) person X decides at time t2 to act as soon as possible. In computational terms, when the act is executed, the program might have already been running for some time (like a time delay).

    Even these possible suggestive criticisms are themselves a bit Cartesian in nature and thus probably somewhat flawed, showing how very hard it is to get away from this type of thinking.

    Sure, this study falsifies libertarian (acausal) free will, but I doubt that anyone outside the religious right or academic left believed in that type of free will in the first place.

    Carrier (2009) and Dennett (2003) have argued that these types of data is consistent with compatibilism.

    Carrier, Richard. 2009. Rosenberg on Naturalism. richardcarrier.blogspot com/2009/11/rosenberg-on-naturalism html. (Accessed July 28, 2010).
    Denett, Daniel. 2003. Freedom Evolves. New York: Penguin Books.

    • Posted July 28, 2010 at 3:04 pm | Permalink

      “Sure, this study falsifies libertarian (acausal) free will, but I doubt that anyone outside the religious right or academic left believed in that type of free will in the first place.”

      Libertarian intuitions might be more widespread than you suppose. Experimental philosophers are conducting surveys on beliefs about free will. In one study, undergraduates were asked which of two universes, A or B, is most like the universe we inhabit:

      “…in Universe A every decision is completely caused by what happened before the decision – given the past, each decision *has to happen* the way that it does. By contrast, in Universe B, decisions are not completely caused by the past, and each human decision *does not have to happen* the way that it does.”

      The results:

      “…nearly all participants (over 90%) judged that the indeterministic universe [Universe B] is more similar to our own.”

      This is from a paper by Shaun Nichols and Joshua Knobe, “Moral Responsibility and Determinism:
      The Cognitive Science of Folk Intuitions” at http://dingo.sbs.arizona.edu/~snichols/Papers/moralresponsibilityFinal.pdf

      • Torbjörn Larsson, OM
        Posted July 28, 2010 at 8:29 pm | Permalink

        By contrast, in Universe B, decisions are not completely caused by the past, and each human decision *does not have to happen* the way that it does.”

        Well, the problem with that is what Coyne mentioned in an earlier post, deterministic chaos can be used to pry that open if you want.

        First, it shows that you can’t replay a universe empirically, so the model becomes physically meaningless. And so does the “have to happen” which is based on replay comparison, not determinism as such which looks at trajectories in phase space. (Or structure of model, if you want.)

        Second, any quantum fluctuation can be magnified to prevent “have to happen”, the same trajectory, even in an unrealistic gedanken experiment.

        Third, it doesn’t look at predictability, which is what we would like to do, née Coyne.

        FWIW, the B) choice is the more realistic description.

      • Posted July 29, 2010 at 5:37 am | Permalink

        This was very interesting news and I immediately concede the point.

        However, the questions asked may be somewhat problematic in that it may have been viewed as a hard determinism / libertarian free will dichotomy and left compatibilism out of the picture. If one takes the intuition that hard determinism implies fatalism to be common, then that may help to explain the result.

        • Posted July 29, 2010 at 11:07 am | Permalink

          Agreed, lots of subtleties to be investigated in all this. But anecdotally, I’ve found lots of people outside the religious right or academic left who have at least initial libertarian intuitions. When you point out the difficulties and absurdities of the causa sui, some of them slowly revise their views, but others hang on for dear life.

  22. Nick (Matzke)
    Posted July 28, 2010 at 3:05 pm | Permalink

    The philosopher Mary Midgley has written a bunch on free will. I know she’s not popular because she has argued with Dawkins about various things, but she is very good at pulling out common modern assumptions that make topics like “free will” so puzzling.

    This chapter on “Mind and Body” is pretty good I think:

    http://books.google.com/books?id=sv37SfzaV8MC&lpg=PA102&dq=mary%20midgley%20%22free%20will%22&pg=PA99#v=onepage&q&f=false

    • Reginald Selkirk
      Posted July 28, 2010 at 6:52 pm | Permalink

      I know she’s not popular because she has argued with Dawkins about various things

      Um, no. She is unpopular because she has written bone-stupid things about atheists and atheism. Sugar-coating will not help.

    • Jon
      Posted July 28, 2010 at 7:39 pm | Permalink

      I don’t know about Dawkins, but if she puts human subjectivity (and “intersubjectivity”) into its own special category, it would make sense that she’d be at odds with Dennett:

      The Germans divide learning into Naturwissenschaften, the natural sciences, and Geisteswissenschaften, the sciences of mind, meaning, and culture, but this sharp divide — cousin to CP Snow’s Two Cultures ( 1963 ) — is threatened by the prospect that an engineering perspective will spread from biology up through the human sciences and arts.

      I’ve heard Dennett say he doesn’t believe in this distinction. This seems like an important divide. One camp seems to want to keep these categories well divided, the other doesn’t, or even wants one category to dominate and encroach on the other.

    • Torbjörn Larsson, OM
      Posted July 28, 2010 at 8:55 pm | Permalink

      The philosopher Mary Midgley has written a bunch on free will. I know she’s not popular because she has argued with Dawkins about various things, but she is very good at pulling out common modern assumptions that make topics like “free will” so puzzling.

      That was news to me. I have heard of her here:

      The summary I read doesn’t mention the name Thompson even once, but I can see him standing tall in the concepts Fodor is crowing over. My inference was confirmed in a review by Mary Midgley (who, it has rumored, has actually written some sensible philosophy…but every time I’ve read her remarks on biology, comes across as a notable pinhead).

      Besides this — perhaps even more interestingly — the laws of physics and chemistry themselves take a hand in the developmental process. Matter itself behaves in characteristic ways which are distinctly non-random. Many natural patterns, such as the arrangement of buds on a stem, accord with the series of Fibonacci numbers, and Fibonacci spirals are also observed in spiral nebulae. There are, moreover, no flying pigs, on account of the way in which bones arrange themselves. I am pleased to see that Fodor and Piattelli Palmarini introduce these facts in a chapter headed “The Return of the Laws of Form” and connect them with the names of D’Arcy Thompson, Conrad Waddington and Ilya Prigogine. Though they don’t actually mention Goethe, that reference still rightly picks up an important, genuinely scientific strand of investigation which was for some time oddly eclipsed by neo-Darwinist fascination with the drama of randomness and the illusory seductions of simplicity.

      Her whole review is like that; she clearly adores the fact that those biologists are getting taken down a peg or two, and thinks it delightful that poor long-dead Thompson is the stiletto used to take them out. I’ve got a few words for these clowns posturing on the evo-devo stage.

      and here:

      Is Mary Midgley supposed to be the epitome of philosophical confusion and bungling incomprehension? She’s like the Emily Litella of science criticism, always going off on harebrained tangents of her own invention, but unlike Litella, nothing ever compels her to offer a meek “Never mind”. Midgely has done it again with another tirade against the New Atheists.

      and here:

      True confession: Nick could be correct, because I have not read any of Midgley’s books. I’ve read many of her short articles, however, and from those I think it eminently reasonable to conclude that her longer works will be much more of the same, and not worth wasting time upon. I have also encountered many people who differ, though, and say that her books are excellent and interesting…curiously, none of them ever goes on to say why. It’s a very weird phenomenon.

      And not once was Dawkins mentioned. Oh, but for PZ Myers mentioning Wilkins as an example of

      an unintentionally amusing counterpoint, though, read this article by John Wilkins, about an atheist who was so annoyed by a Dawkins talk that he decided to call himself an agnostic. What did he discover when Richard Dawkins spoke that was so awful? Why, that these New Atheists were providing a sense of community, instilling values, and talking about beauty and truth in the natural world. He doesn’t cite the idea of providing a framework for social action, charity, and politics, but the Richard Dawkins Foundation is doing that, too. Damned if we do, damned if we don’t.

      Lots of Midgley, lots of Matzke;
      “a notable pinhead” (on biology), “the epitome of philosophical confusion and bungling incomprehension”, “harebrained” (on science and New Atheism);
      but very little of Dawkins.

      Or, what Reginald Selkirk said.

      It looks to me that general criticism is taken for criticism of person (“because she has argued with Dawkins”). But argumentum ad accommodationism is, however mechanically repeated, still a fallacy.

    • Jon
      Posted July 28, 2010 at 10:56 pm | Permalink

      dialectic time is over, resume normal mosh pit…

  23. Reginald Selkirk
    Posted July 28, 2010 at 3:50 pm | Permalink

    This experiment bears on whether a decision is made by the conscious or the unconscious mind. Since neither of those require a supernatural explanation, I do not see how this has anything to do with contracausal free will.

  24. Albert Kong
    Posted July 28, 2010 at 9:30 pm | Permalink

    Basically people want to know why we do the things we do and not something else, is it because we have exercised free will?
    Studies of the multidimensional non-linear information spaces produced by the senses indicate that they are very orthogonal. That is, basically very very isolated clusters of information and the rest blank. To understand this think of different 2D functions in the range 0-1.0 multiplied by each other many many times (say for example 100). The result would be mostly zero since even 0.9 multiplied by itself many times is close to zero and spikes where functions happen to coincide very close to 1.0.
    Imagine trying to find these non-zero features by random experimentation (free will). By random experimentation alone we would never find them (they are too isolated). We must start off with random experimentation though and use the results to tell us where to try more directed experimentation.
    This is a totally new concept, anyone can tell you that, in all of the sciences we use experimentation to identify processes. Random experiments (Gaussian signals)are applied and from the results we regress the model. In this case we are talking about using the results of regression on experimentation to not identify the process (impossible for non-linear high order multidimensional systems)but to say what further non-random experiments to try.
    What identifies the process are the non-random experiments not the random ones. We can do random things, yes, very evident when infants (random motions that do nothing) but they are inconsequential and only the non-random things are consequential (balance upright) and are not free. The same thing happens when we discover love, hate etc. These are not free at all but as a result of directed experiments. We don’t choose to love or hate it just happens. What we do with it depends on how well we have identified the underlying processes. Imbued with great power before these processes are understood and we might get someone like Hitler, for example. Not necessarily evil just ignorant. So the next time you hear one of your friends say they are going skydiving, it is not a dramatic expression of free will, but directed experimentation to further explore gravity. They may even decide to do it many times over if they are having difficulty identifying the experience. I was very fortunate to have fallen off a mango tree when I was a kid and have fully identified the gravitational force so I would not be trying it personally.

  25. MadScientist
    Posted July 30, 2010 at 8:52 am | Permalink

    I don’t see how this affects the concept of “free will” at all. For me it simply suggests that we apes take a long time to act on things in some circumstances and I can’t imagine anything unusual about that at all.


3 Trackbacks/Pingbacks

  1. […] fact, there’s been some interesting research that Jerry Coyne has been discussing on his blog of lateabout the idea that our brains are so bloody awesome they may have ruled out free will . I […]

  2. […] Free Agents or Biological Computer Programs? And Does It Really Matter? I read an interesting article that seems to show that our brains unconsciously make decisions before we are conscious of it. Some […]

  3. […] The “free will” experiment « Why Evolution Is True ¿Existe el libre albedrío? Pues mira… […]

Follow

Get every new post delivered to your Inbox.

Join 30,263 other followers

%d bloggers like this: