Neural correlates of free will

  1. Ken G

    Ken G 3,601
    Gold Member

    A recent thread on free will did not meet the criteria of the forum, but perhaps a similar discussion could be generated by borrowing directly from the example in the rules:
    "The research of Benjamin Libet suggests that our decisions to act occur before our conscious awareness of them. Isn't this a serious problem for the idea of free will?
    http://en.wikipedia.org/wiki/Benjamin_Libet". Specifically, does this imply an either/or battle between determinism and free will?

    One might try adopting the approach that the neurological correlates of free will are deterministic (if one does wish to adopt a kind of dualistic picture where all that is physical is deterministic and free will is housed in some extra-physical seat of conscious choice). To summarize, Libet's research, and a host of follow-on studies of various kinds, suggest that when you look at the neural correlates of consciousness, as they have been identified (a tricky business in itself but the only one that neuroscience can access directly), you typically find that the experience of consciousness is a rather slow process, compared to the timescales on which we sometimes have to make (split second) choices that nevertheless are considered "conscious", albeit "snap", decisions. Does this mean that a decision we must make rapidly, say a life-or-death choice about whether to jump into a river to save someone, are not things we can attribute to free will, when a more long-term decision (like who to marry) might be?

    That might be an interesting issue, to try to draw the line between what is "conscious" and what is just "instinct", but personally, I would look in a different direction-- I would look critically at the very assumption that physically identifiable processes are deterministic in some "absolutely true" way, such that they could preclude a concept of free will. Instead, determinism is a model, just like all scientific concepts. It was never intended to describe how reality actually is, and it is never used for that.

    Instead, the model of determinism is applied to making predictions about outcomes, and in some situations, we know this model leads to good predictions, and in others, it has more limited success. For example, determinism in weather prediction leads to the fairly absurd concept that a butterfly can "change the weather", when a much more natural conclusion is that the butterfly is schooling us in the limitations of deterministic thinking. And in quantum mechanics, that which is deterministic is a process that does not lead to definite predictions, such that the outcomes of experiments are not in fact determined, are are of uknown determinism at best. I feel these should all be taken as a cautionary tale about the fundamental incorrectness of the equation physical = deterministic.

    So in this sense, I would agree that there is something fundamental in how we do science that requires the determinstic model, but I disagree that this means we know the universe itself evolves deterministically-- rather, our scientific understanding of the universe involves placing the template of determinism against the universe, so a deterministic universe is what we get-- with all its blind spots. The deterministic approach produces a powerful, but the evidence is, an incomplete, way to think about things, as per the chaos and quantum mechanics examples.
     
  2. jcsd
  3. apeiron

    apeiron 2,432
    Gold Member

    It would be great if people could actually discuss freewill in the light of what is actually known about the neurology of choice making and the focusing of action plans, rather than applying the lens of reductionist physics. All the philosophical conundrums would rapidly vanish.

    But if people want to debate in terms of universals, or mathematical constructs, then a good place to start is recognising that determined~random are two extremes of a spectum of constraint. Determined is to be locally constrained in some regard. Random is to be unconstrained in that regard.

    There is then a third story - that of complexity. Here you have a systems view, rather than the reductionist's view, where there is a self-organising, equilbrium-seeking, interaction between global constraints and local degrees of freedom.

    This is what the standard reductionist dichotomy (of determined vs random) cannot see. By only viewing causality as a local issue, the global or holistic aspects are missed (or treated as merely "emergent", epiphenomenal even).

    So the whole freewill debate - which has been run through 1000s of times on PF - is motivated by a "too simple" reductionist view of reality. One that really works for physics, but does not work for neurology or the other sciences of the complex.

    And anyone who is actually forced to study the neurology will quickly realise this.

    As to focusing on Libet, his evoked potential approach is easily misinterpreted as it tells you so little about what is actually happening in the brain (it just says something happens ahead of the time it becomes reportable - but we knew that from psychology's very first experiments conducted by Wundt on perception/aperception).

    As a laugh, here is an example of the current literature where you can see the kind of detail people are having to get into to explain the brain as a system that develops its actions.

     
  4. Ken G

    Ken G 3,601
    Gold Member

    I agree with your points about the limitations of the determined vs. random dichotomy. Often a debate on free will seems focused on that dichotomy, as if either was directly relevant (it's not even obvious which one provides more room for free will-- determinism is usually seen as anathema to free will, which then implies randomness somehow supports it, but of course we would always hold a person more responsible for a decision that emerged from the constraints of their persona moreso than something they did at random). I'm sypathetic to the claim that free will and conscious choice may require a different analysis than reductionism.
     
  5. apeiron

    apeiron 2,432
    Gold Member

    Well let's see if there are any takers for a non-reductionist discussion for a change :smile:.
     
  6. Q_Goest

    Q_Goest 2,991
    Science Advisor
    Homework Helper
    Gold Member

    Hi Ken,
    I think what you're trying to suggest is that chaotic systems might not be fully deterministic, but that's not true. They are.
    Ref: http://en.wikipedia.org/wiki/Chaos_theory
    Baranger wrote a decent paper covering the various concepts of chaos and complexity called "Chaos, Complexity, and Entropy, A physics talk for non-physicists" that you might be interested in. About the "butterfly effect" or more appropriately, "sensitivity to initial conditions" he states:
    Ref: http://www.necsi.edu/projects/baranger/cce.pdf

    So regardless of whether or not a system can be called chaotic, complex, nonlinear or whatever, at least given classical mechanical models, those systems evolve deterministically over time. A computer system for example, such as the one sitting on your desk, is utterly deterministic. Even random number generators in these machines are not random, but only pseudo-random, using 'environmental' cues to create seemingly random outputs. If classical mechanics governs mental 'decisions' then 'free will' (or perhaps more appropriately, "mental causation" which is the concept that phenomenal mental states have an influence over physical states) is false. There is no "downward causation" in the sense that locally efficient causes can be subordinated by global physical states.

    The takeaway from all that is generally that mental causation, including free will, is epiphenomenal. A paper quoted by apeiron (Farkus, "Mental Causation in a Physical Brain") for example, suggests:
    In this case, "top-down effects" should not be confused with "downward causation". Not sure why apeiron suggests mental causation isn’t epiphenomenal since he’s quoted this paper as being the “systems approach”. Perhaps he can help explain what that really means. Regardless, it would seem that if our expectation of the brain is that of a chaotic, complex system that operates as we understand it today (ie: through the classical interaction of neurons) then the brain is deterministic as would be "free will".
     
  7. apeiron

    apeiron 2,432
    Gold Member

    Conflating complexity and complication is precisely where reductionists get it wrong.

    Chaos is merely an example of the complicated. The global constraints are simple and unchanging (holonomic).

    Complexity by contrast involves non-holonomic constraints (as argued by Howard Pattee for example). Top-down causality is qualitatively different when we shift from the holonomic to the non-holonomic case.

    Baranger's paper shows he has an intuitive grasp of this, but has not actually studied the subject from a theoretical biology standpoint. So this part of his presentation lack precision.

    I don't follow you here. Perhaps "effects" does seem a too-loose way of talking about global constraints (holonomic or otherwise), but it seems acceptable enough in context. And indeed, it would be exactly the right term if you wanted to draw attention to the crucial systems fact that the top-down action is having an "effect" on the local scale. Because this is the whole point. Top-down constraints do result in something at the local atomistic scale. That is, it creates what is there via its constraint of local degrees of freedom.

    And likewise, I don't get your crack about epiphenomenal mental states. Farkus argues that the epiphenomal part of it all is that philosophers end up talking about something that does not in fact exist separate from the system.

    So if you say "mental state" and I say "non-holonomic constraint", or even "top-down effects", only one of us is speaking the language of systems science. The other is stuck with the epiphenomenalism that is "solved" only by ontological dualism.
     
  8. No. All it really tells us, is that the decision making process is distinct from the self-reflective process. It actually makes sense that the latter would require more processing. Compare how much more difficult it is to learn to drive a car... than it is to drive one after you have learned. In the former case, you have to 'be aware' of everything you are doing. In the latter, your decisions seem 'more unconscious', even though a truly unconscious driver would be in a lot of trouble. The real problem is that the conscious/unconscious dichotomy is overly simplistic. We're only scratching the surface of what consciousness actually is, so this is not surprising.
     
  9. I agree with Joe about the conscious/unconscious dichotomy. I think there is a problem of what awareness and sensory stimuli (qualia) is, but I also think there is a whole lot of confusion as to what a conscious action is, how humans make a choice, and just what it means to be able to think but not able to control ones brain directly in any way. It's very hard to define what choice is, how one makes a choice, and what control is in terms of the influence a conscious mind might have on physical brain states.

    I do think on some fundamental level that if everything is determined that the mind doesn't really have any true choice, but the problem then becomes the question of what other choices would we have that would truly be free? The mind is by definition limited to the choices it can see, and those choices are limited by the physical opportunities presented, no matter how free the mental mind might be.
     
  10. Exploring free will from the perspective of modern neurology is a bit like trying to divine character traits via Phrenology... "scratching the surface" is generous IMO.
     
  11. Suppose an single-vote election design to elect one of two candidates. Let's modify slightly the usual rules so that a randomly chosen schedule determine when each voter will have access to the polling station. Suppose also that there is 108 voters. Finally suppose two hidden observers who have prior access to either the first 105 ballots or the next 105 ones.

    In this example free will is what makes each voter decide on way or the other, consciousness is the public outcome publicised after the end of the pool, and everything in between is purely deterministic.

    Because of the law of large numbers, there is no doubt that the hidden observers will be able to predict the outcome of the election, despite they look at only 0.1% of the individual ballots. Is it to say that the decision is taken by only 0.1% of the individuals? Of course not: the two observers does not even look at the same ballots, but both can predict the outcome.

    What Libet showed is that one can predict the outcome of a decision before the decision becomes consciouss.* The shortcome is not with determinism. The short come is to equate decision with the ability to predict the decision.

    Best,

    *for the sake of clarity I lay aside the usual issues about this claim
     
    Last edited: Mar 8, 2011
  12. In other words, true uncertainty vs. measurement problems?
     
  13. apeiron

    apeiron 2,432
    Gold Member

    How so when the previous two posters raised a crucial distinction between attentional control and automatic or habitual behaviour. And you can go straight to your neurology textbooks to learn in intricate detail about the way the brain handles the two so that it no longer seems a mystery.
     
  14. I find it mysterious because those distinctions don't always hold, and we know from observation that greater complexity exists... it's just an approximation of those intricacies. I'd add, for all the mapping that's done, the interactive and adaptable nature of the brain continues to stymie, and when you touch on consciousness and tne mind... yech.

    I love neurology, but it has limits in terms of what is being looked for or screened during imaging, and the tendency to hyperfocus on one system.

    Hell, we could probably debate all day about what constitutes the Limbic system... it's too messy for my tastes.
     
  15. apeiron

    apeiron 2,432
    Gold Member

    What are you thinking of here? Broadly I believe the attention~habit dichotomy holds up pretty well.

    The striatum can "emit" learnt behaviours in a habitual fashion in time frame of 120 to 250ms, the cortex can "evolve" novel states of response in 300 to 700ms. The distinction in terms of architecture, time frames, and processing logic looks robust.

    Though I would agree there is a third level of responding probably in the reflexive - responses evolved over genetic timescales, such as all the brainstem and spinal level stuff (with time frames of 30 to 100 ms). And mostly very resistant to any learning, or top-down higher brain constraints.

    The limbic brain is a pretty useless construct when it comes to the brain's architecture I agree. I prefer understanding the "emotional" aspects of the brain in terms of orienting responses.

    But the thread is about the "willing centre" of the brain :smile:. And that becomes a bit of a laugh in the literature. Ohh, we can see the anterior cingulate light up, the nucleus accumbens, the insular cortex, the DPFC. When actually the whole brain is the willing organ - and divides according to grades of willing from the reflexive, to the habitual, to the "conscious" or attentive.

    And all these grades are consciously reportable (if we learn to attend to them), but not so easily consciously controlled (because reflexes and habits are not meant to be attentively controlled by definition - that is just a mistaken cultural belief fostered by the socially constructed notion of freewill).
     
  16. Certainly that is the prevailing view, but I suspect that as you say, the combined action of numerous systems gives rise to the possibility of will, conscious or otherwise. When and how a behavior is initiated is a part of the answer, but it leaves major gaps that I don't think reflex alone can fill. PTSD is an example of "learned" reflexive behavior IMO, and yet it only begins to make sense once filtered through many of the regions you mentioned earlier.

    In short, I'm not arguing for free-will... I'm saying that the whole mess is sufficiently complex that we can't pick it apart yet with scientific tools. This is a time when philosophy, psychology, and biology (in the form of neurology) have to find some kind of uneasy balance. How each part adds up to a 'willing' brain, or not, is beyond the ability to extrapolate based on imaging to this point.
     
  17. Ken G

    Ken G 3,601
    Gold Member

    There is an important difference between a "chaotic system", which is something physical, and chaos theory, which is mathematics. Of course chaos theory is deterministic, the issue is whether or not the physical system is deterministic. How would you show that the chaotic system is deterministic? You cannot, you can only show that the chaotic analysis leads to useful results-- that's no surprise, holding the template of determinism to physical systems has led to many successes, none of which show that the physical system is actually deterministic. Indeed, as I argued above, there is considerable evidence that the physical system is not actually deterministic, and imagining that it is gives us that butterflies change weather. They do not-- instead, weather is fundamentally statistical, and no butterfly changes the statistical tendencies of the outcomes. Determinism is simply a limited concept.

    The computer is built to be deterministic. The weather is not.

    Again, there is no evidence that classical chaotic systems are deterministic, that's just plain not true. How would you show that a chaotic system, a physical system not a theory describing it to some degree of usefulness, is deterministic?
     
  18. Ken G

    Ken G 3,601
    Gold Member

    I agree with what you have said here. That's why I tend to see the whole issue of free will as orthogonal to the issue of deterministic vs. random-- the latter are templates we use in reductionists analyses, the former may involve phenomena we are quite far from having much of a handle on.
     
  19. Ken G

    Ken G 3,601
    Gold Member

    I can see where you are coming from on the first two issues, but the basis for the final claim just depends on what one thinks determinism really means. I would say determinism is quite demonstrably an analysis tool, not a description of how things happen (and you might be saying something similar, but then we cannot see that what happens is "deterministic", we can only say what happens admits to useful analysis via determinism). Indeed, when we attempt to use determinism as a literal description of how things happen, it invariably breaks down at some point along the way. Even when we say we are sure an apple will fall when we drop it, we are speaking in terms of probabilities-- it is highly probably the apple will fall, but we don't know that something else could happen we did not expect, like a bomb might go off that blows the apple upward instead. Probability is always an assessment of what you don't know as well as what you do, and determinism is an analysis tool that is intentionally blind to this fact.

    Randomness is also an analysis tool-- my point is merely that being able to predict an outcome with high success is an example of the usefulness of the concept of determinism, not an example of a deterministic process. If you decide what movie you will see today, and I know you quite well, I might be able to write in an envelope what you will choose, based on my knowledge of you. That means I can determine your choice with high success rate-- it does not mean you are not exercising free will. The issue of determinism is nothing but predictability, and is quite orthogonal to issues of free will.
    Here I would agree with you.
     
  20. apeiron

    apeiron 2,432
    Gold Member

    My feeling is different having studied precisely this question of how the brain "wills" actions. We already know more than most people could ever want to know.

    I would just say pick up Luria's The Working Brain, published in 1973, and read chapter nine. The broad outlines were worked out 50 years ago, and the gaps have been filled in by electrophysiology and animal studies much more than neuroimaging. Read Graybiel on the striatum or Passingham on the frontal lobes for example.

    The neural correlates of freewill are one of the "easy problems" even if you are a Chalmer-ite by persuasion. But who really reads neuroscience textbooks?
     
  21. I guess wherre you see filled gaps, I see them as bridges to ever widening gaps in our knowledge... we know a lot, but not enough to really explore what the mind is. Well... we can explore, but not in what strikes me as a meaningful way.

    Oh, and... nerd that I am, I read them... I read and read them, often for fun. So... that's me... that's a serious bias on my part I guess.
     
Know someone interested in this topic? Share this thead via email, Google+, Twitter, or Facebook

Have something to add?