YAFWSA

Yet another free will science alert . . .

Just thought you should know that once again scientists have found the location of free will in the brain. It seems to have moved a bit from last time. In all fairness, the article only claims to have found free will's "center" in two places (neat trick, that).

Comments

[Neil]

I noticed this paper, but I don't understand it. When subjects decided to do something, one part of the brain was activated. When they changed their minds, another part was active. Now what exactly is surprising or interesting here? That people have brains?

[Manuel Vargas]

I think the discovery is that it doesn't all correlate with pinneal gland activation. But still, that they've proven that free will is located in the brain is totally awesome.

(Actually, the issue of where free will is located is, I think, a pretty interesting issue. But that the brain is a plausible candidate location, and that basic mental operations are located in different parts, doesn't seem all that notable, I agree.)

[Neil]

I doubt free will is the kind of thing that can be located. Here's the best we can do: There is, everyone should agree, a power that agents have to deliberate and choose; that power is constituted by or supervenes on a set of properties centred on the central nervous system but extended beyond it and including the agent's body; in most contexts, it might plausibly be held that that power is constituted by properties in the agent's environment as well. So we might say that free will is located there.

[Andrei Buckareff]

The notion of locating "free" will strikes me as perhaps conceptually confused. Locating the hardware, so to speak, whereby we exercise executive control makes some sense to me. But whether or not one's exercise of control is "free" would not be something one can locate. It just turns out to be a property of one's exercise of control. So for reasons similar to those expressed by Neil, I start to scratch my head and get a little nervous when anyone--philosopher or scientist--starts to talk about locating free will.

[Jim]

Locating the "free will" center is certainly misleading. What seems to be left out of this discussion is any description by the subjects themselves as to what they are experiencing. It strikes me that a subjects consciousness must somehow switch into observor mode (becoming aware of the impulse to press the buttom before actually pressing or not pressing it). It is this awareness which is perhaps the foundation of "free will" or "free won't."

[Paul Torek]

I was struck by the bizarre instructions given to the subjects. Signal that you plan to press the button, but then sometimes, don't actually press it. That's my paraphrase of what was, apparently, already a paraphrase. Forget the philosophy for a moment: that just seems a strange way to do neuroscience. To wit, setting up an experiment which has no obvious analogue in day to day life. This smacks of a test of hypotheses originating in earlier research (Libet's?), but sadly the reasons for framing the experiment this way are left unreported.

[Jonathan Jong]

I'm certainly the most philosophically naive here, but I'm not sure I understand the disapproval over locating the part of the brain that is involved in activating subjectively free actions, as opposed to any old behaviour. Sure, this study isn't the most insightful in the world: The frontal lobe is involved in inhibition, big whoop. But still don't we think, in principle, that there are neurological differences between subjectively free (though our subjective experiences of free actions could be illusory) actions and subjectively non-free actions? And if so, wouldn't it be cool to know what these brain regions are? Sure, neuroimaging studies might not allow us to make grand sweeping causal statements (e.g., "The pineal gland causes free actions."), but surely it'd pave the way for that sort of work. Am I missing something here?

[Enigman]

I agree with Jonathan. The more we find out, the closer we get to the truth. I doubt if we'll have much of an idea of what precisely we're looking for, until we find it...

[Neil]

Jonathan, none - or perhaps only a few - of us believe that empirical work on agency, including neuroscience, has nothing to contribute to the philosophy of agency. My worry, and I think it is shared, is that the exciting and widely reported result is exactly what we would have predicted given what we already know. It's just not interesting, not because it's empirical, but because it's not interesting (and that's setting aside Paul's worry: how do we, or the subjects, know that they had neural events corresponding to intending-then-vetoing, rather than a single pretend intention event).

[Arno Wouters]

Who was it who said that one difference between social scientists and natural scientist is that the first use complicated terms for familiar phenomena and the latter familiar terms for complicated phenomena?

Neuroscientists talk about 'intentions', 'actions' and 'decisions' but they do not use these words in their ordinary sense. By action they mean a bodily movement. Such actions are prepared in the Supplementary Motor Area (SMa) and pre-SMA in the brain. The result of the preparation is called an 'intention' or an 'action plan'. Think of it as a series of instructions that when transferred to the Motor Area are executed. Now it turns out that an intention in this sense is sometimes carried out and sometimes not. One can think of several explanations for this. One possibility is that the formation of an intention is followed by a go or non-go signal from another area of the brain, another is that intentions are normally carried out unless a non-go signal interferes. The research in this paper (Brass & Haggard "To Do or Not to Do", Journal of Neuroscience, August 22, 2007, 27(34): 9141-9145) indicates that the second explanation applies. It also indicates where the inhibition signal is generated (namely in the dorsal frontal median cortex) and when (after the subject has become aware of the fact that an intention has been formed).

I found this interesting because it gives indications about the neuromechanisms that result in the execution of an action and the degree and kind of control we have over such mechanisms. As the authors note, insight into such mechanisms can, in the distant future, lead to improvements in psychotherapy and in our dealings with criminal behaviour.

It might also give insight in the role of consciousness in the generation of actions.

In addition provide a counterargument against Libet's free-will-of-the-gaps (just as Newton favoured a God-of-the-gaps). Libet explained the inhibition of prepared actions as the result of the direct intervention of our non-material mind in the neuroprocess. Brass and Haggard's research indicate that "the hypothesis of a special, non-neural veto process" becomes "unnecessary" (compare this to Laplace's "I don't need that hypothesis".

[Neil]

Let me expand on my response to Jonathan. I should have been more careful. Though I think this is reasonably boring science, I don't think it's bad science. It contributes a little to our understanding of the neural bases of agency. My ire should really be directed at the press who picked it up as though it was revealing the essence of self-control. It's just doesn't warrant that kind of coverage (though this is nowhere near as bad as the press coverage of the truly awful study on gender and color, which has got a lot of play). But the second point is that neural localization studies are rarely philosophically interesting. Not never - I have published work that turns on focal lesions in the right inferior parietal - but rarely. This is only to be expected, since neural localization studies are rarely *psychologically* interesting. For a defense of this claim, see Max Coltheart (2006),"What has functional neuroimaging told us about the mind (so far)?" Cortex, 42, 323-331. Coltheart's answer: nothing.

[Arno Wouters]

Regarding the question "how do we, or the subjects, know that they had neural events corresponding to intending-then-vetoing, rather than a single pretend intention event":

The article says that subjects were asked to press a key while watching to a rotating clock hand. They were also instructed to cancel the initiated activity at the least possible moment in one or more trials of their choice (they did so in 45.5% of the cases). After each trial they had to report the clock position at the moment they formed the intention (intention in a more traditional sense). This may sound complicated but apparently a number of subjects were able to follow these instructions in the sense that they produced the activity in the SMA/pre-SMA area known to be characteristic of forming an intention (in the neuroscience sense) and yet did not push the button (the experimenters recruited 19 subjects, 4 of them reported to be unable to follow the instructions). The temporal resolution of fMRI does not allow to determine whether the inhibition started at the same time as the forming of the intention (in the neuroscience sense) or later. So indeed it is not clear whether we have intention-plus-inhibition or intention-then-vetoing. One might interpret the first possibility as 'pretending to intend' but note that this pretention involves the generation of a real intention (in the neuroscience sense) and inhibiting it, rather than reporting that you have an intention without having it.

[Jonathan Jong]

Sorry for the late reply: I was trying to get a hold of that Cortex paper, but for some reason Otago doesn't seem to subscribe to it electronically, and someone keeps holding on the the hard copy whenever I pop into the library for it. If anyone wants to send me a pdf of that, that'd be nice. :)

Anyway, I agree that this particular study was uninteresting in terms of "locating free will." Like I said: The frontal lobe is involved in inhibition, big whoop. And if the suggestion is that neural localization studies are also uninteresting in the free will debate, I'd agree with that to a great extent too. But would there be an objection to the usefulness of neural localization studies that involve timing as well? Think of Libet's studies, but with fMRI, say. Erm, I should say, a very, very well designed version of Libet's study, rather. Would this tell us something interesting? Neil, if I remember from your chapter from Neuroethics on Libet and Wegner (which I read a year ago for an undergad assignment. Thanks for the help, by the way.), you said these studies are not very interesting either. I think. Why was that again?

[Neil]

Jonathan, the 'Cortex' paper is on it's way to you. The reason the Libet studies aren't all that significant (even with the problems identified by Al Mele set to one side) is that the claim that all conscious processes start with unconscious antecedents is just obviously true. What's the alternative? This is a conceptual claim, not one that assumes the truth of physicalism: choices made in the light of reasons assume unconscious (or at any rate non conscious) assignment of weights to the alternatives, and this would be so even if minds were non physical substances. Moreover, the actual decision itself, if it is to be rational, is also carried out by non conscious processes. Consciousness just isn't a decision making mechanism. I defend these claims, relatively briefly, in the Journal of Consciousness Studies (2005).

I don't go as far as Max, by the way: when it's combined with lots of other evidence (from social and cognitive psychology and from double dissociation findings) neuroimaging can be informative. By itself it's never informative.

[Jonathan Jong]

Presumably the alternative to consciousness not being a decision making mechanism is that it IS a decision making mechanism. At any rate, we do want to know if consciousness plays a role in decision-making, and we don't know this a priori, right? So, doesn't the ideal Libet study (if one is possible in principle) give us empirical evidence for whether or not (and how) consciousness is involved in decision making?