Hacker News new | past | comments | ask | show | jobs | submit login
Unconscious brain activity predicts decisions 11 seconds before action (qz.com)
211 points by laurex 13 days ago | hide | past | web | favorite | 125 comments





Michael Graziano’s Attention Schema Theory of consciousness (AST) provides an interesting perspective on this study. Although the study is limited and predicting a decision only implies a correlation rather than a causation, it is worth thinking about how brain activity and conscious action are related.

AST distinguishes between neuronal activity (attention) versus conscious experience (awareness), and postulates that awareness is both a model of attention and a director of attention. Essentially, consciousness as a simulation that has the startling ability to influence the simulator itself. Example: we experience seeing an apple (awareness) and consciously focus upon it, which amplifies the neuronal activity for sensing and processing vision (attention). Due to this amplification, we then gain a sharper experience of noticing more details about the apple. Awareness both models attention and directs it.

In terms of predicting decisions based on brain scanning, it would be easy to see how as a person’s awareness focuses attention on a particular image, even without quite realizing it, that the brain patterns of such attention would be clearly amplified.

This study seems to suggest that our conscious feeling of making a decision is a false experience. That we are simply watching a pre-ordained movie and are being given a false sense of agency. In AST, even though our consciousness is a synthesized model of our brain activity, and even though decisions may be made in neurons not part of that modeling, there is little meaningful difference between the model we experience of making a decision and the brain activity of the decision itself. Although there will always be a lag between a model of activity and the activity itself, we generally experience no lag.

it would be interesting to see in what edge-cases lag would really matter. I believe there are some odd experiments with buttons and blinking lights where the decision to press a button is experienced after the experience of seeing the light from pressing the button.

All the same, if AST is on target, then conscious agency is real, and this paper doesn’t have anything to say on the matter, even if the headline is highly suggestive that it does.


Where can I learn more about this?

“Consciousness and the Social Brain,” by Graziano is a great place to start.

I would want a study that focuses on the opposite: what is the lowest bar a person trying to deliberaty foil such a system can reach. My bet: around 50-150ms.

Hasn't this been known since Benjamin Libet's experiments in the 1970's?

https://en.wikipedia.org/wiki/Benjamin_Libet

Or am I missing something here?


It's primarily the significance of the amount of time before the action (or whatever). Prior to this new study, arguments purporting to show that the relevant notion of free will was undermined by experiments such as Libet's were unsuccessful. Al Mele (FSU) wrote the definitive work on this issue in "Effective Intentions" (2010, Oxford University Press). Mele convincingly showed that the ~500ms time delay in Libet style experiments was not long enough for the skeptical argument to go through successfully. In the last part of the book Mele commented on the implications of any studies of this type (perhaps heading off arguments based on future experiments such as this one). For sake of length I won't quote the semi-formalized version of Mele's argument in the last part of the book, but rather the analogy he uses to elucidate it:

"Consider an analogy. Max struck a log with his red ax, thereby causing the log to split. If Max’s ax had been green, it would have split the log just as well. But he was under strict instructions to split wood only with red axes, and he was committed to following the instructions. If his ax had been green, he would not have used it; in fact, he would have looked for a red ax and split the log later, after he found one. In this scenario, the fact that the ax is red is causally relevant to Max’s splitting the log when he does and therefore to the actual log splitting action he performed, an action that has a specific location in time. Similarly, in the imagined experiment, the fact that at t, Sam made a conscious proximal decision to press seems to be causally relevant to his pressing when he does and therefore to the actual pressing action he performs. I should add that although we do know that, other things equal, red and green axes split wood equally well, we do not know how effective unconscious decisions are. Nor do we know whether unconscious deciding (as distinct from unconscious nonactional intention acquisition) is something that actually happens. Also, for all we know, if there are instances of unconscious deciding, they are far too rare for there to be more than a glimmer of a chance that if Sam had not made a conscious proximal decision to press at t, he would have made an unconscious one."

----

It is worth pointing out that Mele is famous for remaining agnostic about free will (and certain fundamental questions surrounding it).


It's not as if they completely disregarded the existence of Libet; he is referenced many times in study and even the article posits that the research is supplementary to previous work done in the field.

I am speaking from memory, but iirc Libet's study wasn't without flaws and it was contested many times during the years.

More researchs like this are positive, as this is a far from solved field.


I had heard of this as a factoid about 30 years ago. Although I didn't know about Libet's experiments.

From the Wikipedia article -

That is to say, researchers recorded mounting brain activity related to the resultant action as many as three hundred milliseconds before subjects reported the first awareness of conscious will to act.


Thought I'd share one of my favorite short stories (written by Ted Chiang, most recently of Arrival fame) that bears on this: https://www.nature.com/articles/436150a

Wait. This device can be fooled easily: At any point in time, I can look at the LED. If it's off, I press the button exactly 1 second later. If it's on, I do nothing.

What am I missing?


You're simply missing the fact that by definition there won't be any case where you see the LED being turned off 1 second before you push the button. If you see the LED being turned off, you won't be pushing it the second later.

It’s interesting how many comments immediately defend the evidence of “free will”. I’m not saying this article comes close at all to invalidating “free will” but it begs the question how will humanity react if we do invalidate it one day? Seems on the same scale as “the world is round” or “Earth revolves around the sun”.

> I’m not saying this article comes close at all to invalidating “free will” but it begs the question how will humanity react if we do invalidate it one day?

It (or more precisely, the people who fully internalize the result and who have the power to act on it) will probably start treating other people a lot more like we currently treat machines.


What does free will mean in this situation? It’s still your brain making the decision, whether it’s conscious or not.

It seems to me that if you have a materialist theory of consciousness that there have to be unconscious essentially mechanical processes underlying it. But as long as those processes are contained within the being making the decision, it seems to me that it should still be considered ‘free will’.

Now, perhaps there is some level of determination there that makes the traditional idea of free will hard to maintain, but I’m not sure we have to toss the concept out entirely.

I’ve always largely considered it a convenient fiction useful for legal and ethical discussions.

If you’re designing a legal system, I don’t think you care as much about the first causes of human action compared to whether the action was done by a particular human being without being forced to do it by some external actor.


Why would you assume that "Free Will" is something that can be proven/disproven? Something that is in the domain of science?

To clarify the question, what kind of scientific experiment would you even imagine that would even resemble something producing evidence for or against free will?

(These experiments talked about in this topic do not at all fit it: they just show that there are conscious and subconscious parts of our experience of life and our experience of the process of choice. In what world would that be related to the degree of "free-ness" of that choice is beyond me. I thought we all knew already that there are conscious and subconscious parts of the brain and that many of activities of "I" are subconscious.)


One way is to show that, given a state of the universe, it is possible to determine the next state. From what I understand, this does not seem to be the case.

I think free will vs deterministic physics is a false dichotomy.

My instinct is that physics is performing a computation at optimal efficiency. Even if everything is deterministic, there is no other way to determine the outcome other than letting it play out.

In other words, even if a decision you made was bound to happen, there was no other way to find out what that decision would be beside letting you compute it.

So, your decisions can be both deterministic and unpredictable.

From this perspective, I don't think the existence or absence of randomness makes any difference on the question of free will.


Lets assume determinism is epistemological unfeasible, ie not possible to evaluate all the states due to limitations. Unpredictability makes no difference to the ontological question of whether actions are due to casual chain or pure irreducible randomness (if it exists in fundamental level). There is no other alternative. Free will is ill defined/semantic nonsense in libertarian sense of its use as there are only two possibilities.

Is this a strictly hypothetical thought experment-type of design? Ie, nothing that could be realistically built and tested? I am asking because our current understanding of physicas is that it is practically impossible to know exactlt the state of even a single particle, let alone of the whole universe, to the precision even theoretically required to "determine" the next state. (Outside of simplistic esperiments in a laboratory.)

This is not a limitation of technical prowess, this is a quality of quantum particles, of thermodynamics, etc.

My point is, if it is impossible to construct such an experiment, it means it is not a question within the domain of science. Science deals with hypothesis which can be realistically disproven.


>From what I understand, this does not seem to be the case.

Maybe. Determinism isn't dead, by any means. Depending on what model of quantum mechanics turns out to actually be correct, full determinism might be just fine. But even if one of the models that would invalidate full on determinism is correct, 'adequate determinism', similar to what Stephen Hawking believed in, might apply - the idea that the level any quantum fluctuations occur at makes it unlikely to result in any difference at the level that human thought, etc., operates at.


Additionally, an individual's next "move" (whatever that may be) being unpredictable doesn't make it "free will".

Universal predictability would definitely disprove free-will, but disproving universal predictability tells us nothing about free-will.


Without predictability there is no distinction between "free" and "random".

Exactly, which makes "free will" a semantic nonsense term. Understanding that neither determinism nor in-determinism of the universe show anything about popular conception of free will (libertarian sense) allows one to exit a semantic misconception.

I'm not sure I follow.

If determinism is correct, then given enough storage and computational power (or intelligence capable of doing this within their mind), an outside observer would be able to figure out every single action I would ever take before I took it.

If how my brain works is deterministic, and all of my encounters are able to be pre-determined, and those all deterministically shape my decisions and actions, then... how I can I possess libertarian free will?


I think it can't be the case because the brain is a chaotic system. But hiding behind chaotic behavior to preserve "free will" is farcical, isn't it? We can't know how the pendulum[0] will swing next, so we can say that it's got free will?

0: https://upload.wikimedia.org/wikipedia/commons/4/45/Double-c...


Well, chaos theory is explicitly not a denial of determinism, and speaks more to our ability to precisely measure state and accurately model it than anything else. The more precisely you can can measure state, the longer the Lyapunov time, and the longer we can accurately predict the future states of the system.

There are systems today that we can predict fully (i.e. we can predict what will happen faster than it can actually happen) that decades ago would have been unpredictable due to their chaotic nature. There are systems we cannot predict today that we will be able to in the future. Brains very well could be one of them.


It's not "farcical". If you could prove that physical states are deterministic, you could prove that there is no free will. That statement makes no claims whatsoever about what it means that physical states are not known to be deterministic. In other words, it could have been useful as proof negative of free will, but not as proof positive. And that's okay.

Wheter "free will" exists or not seems like a technicality, tied to whether the Universe is deterministic or not. Whether or not it's deterministic, a brain must still go through its thought processes to come up with any decision. The only difference that non-determinism makes is that the decision may randomly go one way or the other based on low-level fluctuations. Thinking and decision making still occurs even if the Universe is deterministic.

It sounds like we agree.

To be clear, I think that free will--by definition--cannot exist in a purely deterministic universe. Any semblance of free will in such a universe would be purely illusory. It also cannot exist in a universe where events can only ever be some combination of random events and determined events. Randomness is also not free will.

Free Will would have to be some other special category of events that is neither determined not random. We don't have a word or phrase to describe what these events would be like other than "Free Will." However, lacking the vocabulary and math to describe it does not make it non-existent.


>Free Will would have to be some other special category of events that is neither determined not random. We don't have a word or phrase to describe what these events would be like other than "Free Will." However, lacking the vocabulary and math to describe it does not make it non-existent.

That is not even non-existent, more like a misconception which one comes out of by thoroughly examining what "feels" correct. Like Wittgenstein said one of the jobs of philosophy is to free us out of semantic knots. I see free will such a knot arising due to our conception of a separate self from the causality, leading us to project a conception of freedom which makes no sense when examined upon.


I think this is some kind of supernatural "free will", and no mechanism has been found for such a thing in science. It's associated with mind–body duality, souls, homunculi inside the body, that kind of thing.

It doesn't have to be super natural either, though. By the nature of it's existence, the universe proves that there can be chains of events spawned that completely undermine our current treatment of causality. A realistic model must include at least one of the following:

1) Events with no cause such as the beginning of the universe or broader multiverse, depending on your beliefs or 2) Infinite regressions of causal chains if you believe that the universe or God (or a god) has always existed or 3) Some combination of the above.

In all of these, somehow there can be uncaused effects, but in 1) and 3) some causal chains can just begin existing, like free will would require.


If uncausal events were so common, wouldn't they have been discovered in laboratory conditions by now? They'd be messing up experiments everywhere.

You might even say that such events would be experimentally indistinguishable from random events. Or practically identical.

Seems nonsensical to me. You can predict 99.999% of our behaviors... as long as you make one decision your whole life that bucks the model, you are not the model.

Yeah, I don't even think it even touches the concept of free will. What it does show is that thought is layered, where you as en entity is only conscious of the "top" layer.

But the lower layers are all still you, molded based on your experiences and conscious layer.

The part of you that goes, "I choose A" is only the vocal/mental switch where you become conscious of the choice.


Sure, it's a construct, but do you really want to throw responsibility over board? Why punish people if their choices are just consequences of weird electron configurations?

Because not having free will won't change the fact that we act as though we do, nor will it change the desired outcome - punishment is meant as a deterrence, not just retribution.

That needs some citation. Or why then is justice so fixated at the right guy suffers at the end?

What does "free will" even mean in this context?

If I were to make a decision such as this, I would make a snap preliminary decision, then review it to see if there were glaring errors. That seems to fit with the results.

The results don't seem to be stating anything new or non obvious to me. Am I missing something?


The key thing is this "the researchers were able to predict which pattern participants would choose up to 11 seconds before they consciously made the decision"

So this goes into the question of "free will". Is your conscious mind making the decisions or is just trying to justify the decisions made by the unconscious mind.


The researchers did not decode it with 100% accuracy, so one interpretation is that the unconscious mind says "Instinctively, this seems like the right decision," 11 seconds later the conscious mind says either "Oh, yes, I agree with that" or less frequently "Actually I'll do something else", and the researchers are only tracking intuition.

It doesn't tell us anything about free will other than the fact that the process of decision making happens prior to our conscious awareness, which we already knew.

Ye, it's is kinda obvious that the decision is made before you "know it". How would it otherwise work.

If I understand correctly, the part that says

> up to 11 seconds before they consciously made the decision

should be interpreted as

> up to 11 seconds before they pressed one of the buttons

I can't imagine how they measure the time when the conscious decision was made but they definitively measure when the button was pressed.


I suppose the problem for me is, there isn't really a decision for the conscious mind to work on, the question is trivial without a right or wrong answer, free will doesn't enter into it.

This was choosing horizontal or vertical stripes, there isn't a rational reason not to go with my gut, although I could if I wanted to demonstrate free will.


Is the assumption then that "you" means only the conscious part of your brain?

What happens 11 seconds before the conscious decision can be part of free will, I guess.


OP is correct actually, it could be part of the process by which we exert our free will.

That's the decision process that the current chess world champion has explained he also uses, for most moves.

It seems pretty obvious that when confronted with a choice you generally take an immediate biased opinion and then spend some time reviewing the alternatives just in case. It wasn't 11 seconds until they made the decision, it was 11 seconds until the decision was confirmed and vocalized. It sounds like 'basic' image focus detection.

Maybe we just experience our brain making decisions. Everything else might be ‘unconscious’.

Implying that our conscious experience of the world has no control over our brains.

Otherwise conscious/unconscious implies some sort of duality in your head. Like the homunculus..


More like experience a "subset of the environment". We can barely handle more than a few digits at a time, a particularly small square of vision, or a particular song. We can focus our mental effort at really only one thing (the other has to be learned/unconscious so that it can go parallel).

I think the question becomes: Why are we conscious in the first place?


Unless there is something special going on.. experience seems to be a property of matter itself.

Which could mean everything is aware to some extent. People, dogs, rocks, etc.. But because of our big complicated brains that can hold memories and all that - we are aware of being aware.


The researchers don't know if the choice was made in the first second or in the final second: it doesn't make sense to say that the brain activity predicted the choice 11 seconds beforehand.

If feels pretty obvious to me to think that internal mood has something to do with decision making. It could take a couple of seconds for a reaction to cook up in order to become a decision.

Sure, and seeing some elecrical patterns in DRAM circuitry will predict changes before they show up on your monitor too. That's just how computers work.

Similarly, unconscious activity is just how consciousness and choice work. That doesn't mean that free will doesn't exist. It's a complete misunderstanding of the concept.


I came here to say this. Stating or notifying awareness of something in an external matter isn't what consciousness is. It is the end result of a process, not the beginning. Your analogy is a nice one. Another one is the decision by a jury happens minutes or hours before they actually tell the courtroom. Same concept.

No one thinks that computers have free will. Not yet anyway. The idea that the conscious mind isn't the one making decisions seems to have implications for free will. Can you explain why it's misunderstanding of the concept?

If I decide to type the words "free will", I don't tell my fingers, "move to coordinates <x, y, z> and apply force vector F", I just type the words. I'm not really even conscious of typing the individual letters unless I slow down and think about it. That's because, through months of training followed by decades of use, I've trained my unconscious to type without having to consciously think about anything other than what to type (and even some of that I can fob off onto my unconscious).

Does this imply it was not my choice to type "free will"? Would I be unable, through conscious effort, to learn how to type with a different keyboard layout?

I can't think of a reason to suspect that we don't also train our decision-making processes the same way. (This is especially true for such trivial, inconsequential decisions as to which pattern to select in an experiment.) And I suspect we could train ourselves to always prefer a particular kind of pattern fairly easily through conscious effort.

So even if you limit free will to consciousness, I am not sure how you can say anything specific about free will as a result of the study.


> The idea that the conscious mind isn't the one making decisions seems to have implications for free will. Can you explain why it's misunderstanding of the concept?

Because "you" are not just your conscious mind, you are your entire body and mind, including your unconscious; these are holistic and interdependent systems. The tendency to want to separate these is a common mistake reminiscent of an absurd mind-body dualism.

"Free will" is making a choice according to what "you" want, not according to what your conscious mind thinks it wants.


conscious and unconscious have no discrete definitions. They are fuzzy concepts. Separation of those concepts would require an interface between them. Now you've got three states to your fuzzy system. At some level of granularity, the model might be accurate. But going one level above a single blob is already an achievement.

I find that the intuitive idea of free will doesn't make all that much sense anyway. Leave aside the discussion of whether people have free will in this universe, how would a universe in which people actually have what most people call free free will even work?

Assuming that "free will" doesn't include manipulating the laws of physics, pretty much like the one we live in? I can see where people are coming from when they say there is no such thing as free will, but it sounds like they are saying the no one has the ability to consciously override their impulses. I don't smash the skull of everyone that makes me angry, for example.

but why is bifurcation required? is the one stopping the impulses any different than the impulses engine?

the reason you don't smash all of the skulls is as causal and determined as the impulses afaict.

to me it seems that life is just sufficiently complex self contained units that are completely bound by deterministic laws. we have a sense of self referential capalbility which may give the illusion of free will but I'd say that it's not your decision to stop the impulses of smashing skulls rather it is the inescapable result of your unique determined state or rather process that your brain has been configure d for based on life experience and genes.


You're assuming free will is mutually exclusive with determinism. This is probably not the case. See Compatibilism, which enjoys majority support among philosophers and studies of lay people.

This works as an assumption, and is rather useless in practice, because we cannot determine the deterministic state to its full extent, even in thought experiment, without limiting the degrees of freedom converging to zero, to then say, well I had free will after all, when I didn't know any better. I really don't get it.

I mean, support for Compatibalism is in effect not the support of determinism that it claims to be. It's just an effort of philosophers to safe free will, out of free will, which means, to me, out of ignorance. So the only real insight is that we are sufficiently ignorant. How is that still sold as a positive thing?

That still didn't come out right.

I was trying to say, that Compatibilism is a meta argument. If it isn't then because it determined free will as a deterministic fact. I understand that philosophy is rather complicated, and at the very least I doubt that, if Compatibilism is true, that many people understand it sufficiently to be in the position to support it, other than as an educated guess.


I think the problem is that most people misunderstand philosophical inquiry. The free will debate is about defining what free will means such that it can serve as one factor to ground moral responsibility and how we employ our moral reasoning.

Long ago, some philosophers suggested that we cannot be held morally responsible for factors that our outside of our control, and further, that any effects that follow from factors outside of our control, are thus also outside of our control. Determinism is the quintessential example. Numerous arguments have since been presented that debunk this need for control, the ability to do otherwise, etc. These properties simply aren't necessary for moral responsibility.

Furthermore, studies in experimental philosophy have shown that lay people actually employ Compatibilist moral reasoning, debunking the long held assumption that people are incompatibilists because of religious upbringing.

So I disagree that Compatibilism is some sort of meta-argument, as it fits squarely in the expected domain of discourse for free will. It is precisely fulfilling the philosophical research program of identifying that which is necessary to make sense of choice in our moral language.


Now I know what other people feel like when they listen to programmers talk. :D

Do you know of any materials that present the debate in a form that a ten year old could understand?


the danger is ascribing too much agency to a person's actions. We are all a product of circumstances. Instead of saying "You're lazy!", it's more productive to increase their living conditions.

I think your response often conflates two things: it's correct to classify a lazy person as lazy, but a diagnosis is not a treatment.

For instance, diagnosing someone with a sprained muscle as being out of shape doesn't help get them in shape, and in fact, the immediate treatment is more rest until they're healed.

So I think it's incorrect to say that people don't have agency, but it's equally incorrect to say that merely identifying that they have agency should suffice if such a person is making incorrect decisions. Sometimes it might, but not always.


It wouldn't, free will has to be a product of Newtonian cause and effect or quantum randomness, neither of these things i define as 'me'.

Even adding a soul wouldn't help, i imagined a soul existing in another universe affecting this one; same problem, just 1 universe over, cause and effect or random is still not 'me'.

I think the only way to square any version of free will is to radically redefine what the term 'me' means.


The problem with this line of thinking is that everything we do in our lives, especially rational thought and communication based on it, pre-supposes some sense of free will.

In particular, it is well known that if you don't believe in any degree of free will, the stranger quantum effects (or basically any kind of statistical thinking) can be ignored - if the experiment to be run is itself pre-determined, then hidden variables that correlate the behavior of the experiment with the 'decisions' of the experimenter can't be ruled out anymore.

Perhaps more fundamentally, if you believe that there is absolutely no free will, then of course this conversation is essentially pointless, eve though we were forced into it by the big bang.


The presupposition of everything we do in life may be essentially wrong.

>if you believe that there is absolutely no free will, then of course this conversation is essentially pointless

If you are absolutely devoted to defining yourself as the ghost in the machine in a world where ghosts have been disproved; But if you can redefine 'you' to mean the machine, then your actions are an extension of who you are. So you can't choose to be someone other than yourself, but what value did that add anyway?


> I think the only way to square any version of free will is to radically redefine what the term 'me' means.

The problem is that you're employing an incoherent or invalid conception of free will. Free will can be compatible with determinism, eg. Compatibilism.


>Compatibilism(from Wikipedia); They define free will as freedom to act according to one's motives without arbitrary hindrance from other individuals or institutions.

This only works if we radically redefine the term "one's motives". Which isn't different from redefining what 'me' means.

The problem is of course that in a Newtonian universe one's motives is a result of arbitrary past actions outwith our current definition of one's motives. And in a quantum universe one's motives are arbitrary (randomly) generated outwith our current definition of one's motives.


> The problem is of course that in a Newtonian universe one's motives is a result of arbitrary past actions outwith our current definition of one's motives. And in a quantum universe one's motives are arbitrary (randomly) generated outwith our current definition of one's motives.

Fortunately, that's not a problem at all for free will and moral responsibility! That's why Compatibilism is now the prevailing paradigm.


Nothing is a problem if we are willing to stick our fingers in our ears and quietly hum la la la. But anyone who wants to square and explain how the common definition of free will works in this universe, still has a problem.

A radical implication is then to locate free will in the unconscious. The true acts of freedom are choices made unaware. The consciousness merely registers the choices that are already made.

It wouldn't necessarily have to push it to the unconscious. Free will could be an infrequently-exercised veto power.

Which might not even be an unreasonable way to look at it. Many people make a contrast between habit and will. There already seems to be a common idea that much of the time, people do things the way they always have, and it is only at certain key moments that the will asserts itself and changes course.


> The consciousness merely registers the choices that are already made.

A point made in the Matrix movies: you're not here to make a choice, you've already made it. You're here to understand why you've made that choice. I think that's probably a pretty good analogy of how consciousness operates.


Brain has much more unconscious processing than conscious. Some of that processing deals with complex high level topics. Once the processing is done the result "pops" into conscious mind. It does not mean the conscious mind does not have free will, rather, that it works as an editor for a bunch of autonomous agents running in the background. Some of the feedback from those agents is bullshit and we are taught to fight against those impulses.

> It does not mean the conscious mind does not have free will, rather, that it works as an editor for a bunch of autonomous agents running in the background.

I rather think of it as an integrator, in that, it collects and integrates the results of various subsystems into a coherent "rationale" for why you've made a choice, and that rationale is then stored in your memory as "my choice".


I think the idea in dispute is that the area scientists had called unconscious isn't just exclusively unconscious, but also takes part in conscious thought.

I guess the idea that my conscious mind isn’t actually making decisions isn’t so bad, as long as the decisions are coming from somewhere in me.

Then again, I may be biased.


An interesting question then is in what way unconscious you is different from unconscious me.

Computers are ideally non-deterministic, but due to NP-completeness, they can't predict their own results.

It's actually due to the undecidability results that there can't be a general algorithm to predict the result of any algorithm. If the problem were merely NP-complete, it would mean that, as far as we know, it would take a very long time to do it, but it would still possible.

>Sure, and seeing some elecrical patterns in DRAM circuitry will predict changes before they show up on your monitor too.

is it? you should patent and sell it, you could call it fast-forward computing. FPS game players would pay a fortune for computer being able to display events from the future!


Are the action potentials spinning up like a turbo in a car engine?

11 entire seconds is a life for brain activity. Superslow.

I don't think this paper is related to free will, or if they even claimed it is. It's just sensational title.

The choice is completely non-consequential, it's picking between 2 patterns. The brain choose based on it's initial biases. It's like when you are turning off your brain watching youtube videos, it's likely that you will open another video after that instead of going back to your work because there is a strong pre-existing brain activity related to whatever you're watching.

What I'm saying is it isn't like some directive from the subconscious that you cannot go against, it's more like one room is filled with sweets and the other is filled with nothing. The choice is still taken at the moment.


This is just one of the many studies. You can detect the choice from the brain up to seven seconds before when people are asked to push a button any time they feel the urge. https://www.nature.com/news/2008/080411/full/news.2008.751.h...

> The choice is still taken at the moment.

There is no reason to expect that 'the moment' is the same as when we become aware of it. It's questionable if we can make make any choices within two or three seconds. We just have behaviour that feels like it's our own volition.


>You can detect the choice from the brain up to seven seconds before when people are asked to push a button any time they feel the urge. (...) There is no reason to expect that 'the moment' is the same as when we become aware of it. It's questionable if we can make make any choices within two or three seconds. We just have behaviour that feels like it's our own volition.

Of course it's our own volition.

Do you see anybody else there telling the person what to do? Is there some "eternal soul" that should have been informed to make the final decision?

The misleading dichotomy here is that the conscious urge is something different from the "choice from the brain up".

Whereas the conscious urge is just a later manifestation of the "brain up" process. It's just a nicer UI output we give ourselves of the choice process.

It's still our brains, that is us (there's no third party involved), that makes the decision, based on our past experience, knowledge, preferences, memories, etc as wired in our brains.

Whether we do it at the conscious layer or we do it at the lower level and then project it at the conscious layer doesn't make it any less of "our own volition".

To me it's like saying "the computer didn't make this picture, because the picture was already created in the memory as a structure before it was send to the screen".


>Do you see anybody else there telling the person what to do? Is there some "eternal soul" that should have been informed to make the final decision?

Obviously not, but commonly when people refer to so called 'free will', they don't just mean to imply that you, as a physical entity carry out things by your own volition, it is taken to mean that you consciously deliberate on your choices and have some insight into your own decision making.

These experiments suggests that the actual decision making happens at a much more secluded, black box like level.

Free will, in a meaningful sense, is taken to mean 'rational control', not just 'that guy over there acts in some way because his brain commands him to', because that's true almost by definition if you exclude religious explanations.


I don't see how any of that negates the idea of rational control in the sense that we commonly understand it. The series of physical events in the brain that lead to a decision are not subject to an out of band process that can override the final conclusion; it doesn't seem surprising that there is a delay between the unconscious processes that underpin our rational thoughts and the eventual culmination of a decision, IMO the opposite would actually be the surprising result (i.e. our conscious mind is not a supervenient product of our unconscious brain and supporting biological processes)

>Obviously not, but commonly when people refer to so called 'free will', they don't just mean to imply that you, as a physical entity carry out things by your own volition, it is taken to mean that you consciously deliberate on your choices and have some insight into your own decision making.

Well, it's both: as people we can consciously deliberate on our choices for months, torment ourselves at a very conscious level, express it out to others who couldn't care less, listen to moody EMO music, even write lyrics about our predicaments before we make a choice.

But we can also make some much easier decisions using our pre-conscious processing. And that might involve deliberation too -- it certainly involves prior experiences and stimuli.

That's orthogonal to whether free will exists or not, though, and people saying that such pre-conscious choices it's proof free will doesn't exist are wrong. There's no necessity that free will has to be expressible as conscious deliberation. Just that it comes "freely" from the person. There might be lots of conscious deliberation behind those "non-conscious" decisions too.

Free will might very well not exist, but not for that reason.

We are ourselves, the whole time, sub-conscious or conscious. One's self doesn't resolve to just the "voice 'speaking' in our thoughts".


Especially if you start to work at your inner self and tune your reactions to be who you really want to be, it gets deep and very much in tune to what you are saying IMHO.

"Time between when a decision was made and the awareness of it" != "Not having free will"

Given that we have to make decisions all the time it's natural an unconscious heuristic could exist, but that, again, doesn't mean "not having free will".


The feeling of being in control and acting on your free will on the spot is demonstrably not connected to actual decision making will. The whole concept of free will is so vague that debating if it exist or not is useless.

If you think that "free will" is becoming aware of your decisions just when you make them, consider the fact that the human perception of "Now" spreads over 3 seconds or more. The ordering of two events can swap up to 5-7 seconds in some cases (ever feel that you are thinking about somebody and then they call you).

There is some evidence that your brain does stuff parallel and out of order and then marshals them into stream of consciousnesses where causality of internal processes seems right.


> The whole concept of free will is so vague that debating if it exist or not is useless.

Yet the idea of free will is so empowering that I choose to go with it even when it would make sense short term to deny it.

Seriously, even if it wasn't true and somebody here starts to believe it - even if that thought weren't a real decision but just a response to my persuasion skills - even then it would probably make a positive impact on their life.

Is everything in my life a product of my own decicions? No. I've been dealt a good hand. But after I learned to decide for myself it became better.

FWIW I've also no idea who would downvote you, your position seems reasonable even if we disagree strongly. If anyone think what you say is wrong I'd urge them to pst evidence instead of just downvoting.



“A man can surely do what he wills to do, but he cannot determine what he wills” – Schopenhauer.

For we act not only under external compulsion but also by inner necessity.


The parent's point was that it shouldn't be surprising that inconsequential decisions arise long before we have conscious awareness of it. But that has no relevance to the question of free will, i.e. the capacity to deliberate on decisions. A more relevant test would be to test when explicitly moral decisions are made in relation to our conscious awareness of the decision.

I’ve been sitting with an Advaita teacher for a few years. One of the teaching’s pointers is to observe one’s choices.

I’ve been doing this for at least 3 years and have not yet encountered a choice made out of “free will”, and not by desires, basic needs, thoughts, trauma, conditioning etc.

My conclusion is that in the West, “free will” is an un-investigated label for any combination of the above.

This teacher often speaks of a sense he calls “the false sense of authorship” which claims the responsibility of our actions, thoughts and words.


> asked to push a button any time they feel the urge.

Still completely inconsequential choice. Nothing is going to change in lives of these people based on this choice. This is not a "free will"-type of choice people usually think about.

I think this all started with Sam Harris using this argument in his crusade against free will, and some people just picked it up without giving it a second thought.

The argument is that if the choice is unconsequential (like thinking of a random city or pushing one of equally meaningless buttons), Sam sees that as somehow being the "most free of a choice you could think of". But it is the complete opposite to how the real world and real life works!

I would argue that choosing which school to go to, which career to pursue, which food to eat, which person to spend time with - those are real choices, that's where the free will is used. In choices which have consequences. Choices that matter. Choices that require using the consciousness, not appropriating its gargantual power as being a glorified random number generator. Not in a choice that requires no effort, and can be forgotten the moment it is made, because it will have no consequences whatsoever.

Good luck making those real choices being predictable by a computer. When a computer can predict what career I am going to choose or who am going to hire (given real viable choices) - then we can come back and talk about free will. Of course at that point, by definition, we have Strong AI which can predict almost anything and the world will be completely different, beyond our wildest imaginations because we can't see it behind the veil of technological singularity.

These experiments say nothing about free will. (Of course the papers themselves don't pretend to either, just the sensationalist journalists weakly versed in popular philosophy.)


the fact that it's 7 seconds... that kind of proves that it isn't a directive. A directive should be executed immediately. That would make more sense to me, anyway.

Our brain is a massively parallelized computer that runs the entire body yet has a cache of ~7 items and runs on 20 watts of power. I would expect a degree of latency in its more complicated operations.

The brain runs on 120 watts, not 20. In addition, working memory has a capacity of ~7 toplevel item groups, not items.

I concede the ambiguous 7 item thing, but where are you getting 120 watts from? The only figures I get from my googling say the entire human body only uses 100 watts on average, with 20% of that used by the brain.

Human body consumes 100W at rest, ~300W at sedentary work, ~1kW at full power. Full power figures are based on appropriation of longitudinal draft ratings.

Yes, it's basically a demonstration of brain activity monitoring technology.

choice can be predicted with eye tracking. so what.

N=14. This is a preliminary study, not something you should change behaviour or beliefs nor cast policy based on.

Even N=1 would be interesting. Not everything has to use a cohort of thousands - when you're demonstrating the existence of a phenomenon, a single case will suffice.

Oh thats not like this, if you give me N=1000, i can probably prove anything for N=10 (considering bad intentions)

So basically for big N's, study is much trustworthy.


Give N=7.5 billion find me N=1 person that is 300 years old, or one that can fly.

For a large N, the study is only less likely to produce a false positive. It is more likely to produce a false negative though. There is an optimal N where type 1 and 2 error are minimized which is not infinity.

> "In, other words, if any pre-existing brain activity matches one of your choices, then your brain will be more likely to pick that option as it gets boosted by the pre-existing brain activity.”"

We have biases. Sometime we are aware of them. Sometimes, evidently, in trivial situations, we don't notice such patterns.

"Be mindful of your biases" just took on a wider scope.


- They are showing 2 images.

- maximum duration to choose is 20 seconds.

So basically if you look one of them 10 seconds, seconds one 10 seconds. And make your choice. They can see your choice when they go back in your brain activity 11 seconds before.

(If they can with MRI can see how much you like something while you are looking)

So basically nothing much value here.

(PS: when i see 'upto 11 seconds', this sounds like marketing talk)


I understood that they show both images simultaneously. Can someone look at the research paper?

(I also don't like the "up to 11 seconds".)


This is a good tldr comment for people who are just reading the headline and drawing conclusions about free will



Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: