Free will experiments reveal how little we know about our minds

http://www.bbc.com/future/story/20150806-why-your-intuitions-about-the-brain-are-wrong

by scorchio • 11 years ago

34 68 11 years ago

dmfdmf• 11 years ago

Would the statement "Free will is an illusion" be less of a self-contradiction than when stated as a question? Hard to say.

scorchio• 11 years ago

Sam Harris' book on the same subject - http://www.samharris.org/free-will

andy_ppp• 11 years ago

Yes, it's a brilliant book that proves what we see as free will is obviously not free and it's not even up for contention that we actually have the ability to decide, even though it feels as though we do.

AdieuToLogic• 11 years ago

This appears to be a bit of bubblegum neuroscience. First, the article states:

  He asked participants to report, using the clock,
  exactly the point when they made the decision to
  move.

Then, it goes on to say:

  There’s no reason to think that we are reliable
  reporters of every aspect of our minds.

Well, this is quite the conundrum. Which is it? Report exactly when you do something or accept that we are not "reliable reporters?" Yet the author goes on to further say:

  Even supporters of Libet have to admit that
  the situation used in the experiment may be too
  artificial to be a direct model of real everyday
  choices.

Given the delays in perception we all experience due to working at "chemical speed", as opposed to light speed, this experiment smacks of fast-twitch muscle measurement more so than exercising free will.

Even the original premise of Libet's experiment negates a free will choice:

  “let the urge [to move] appear on its own at any time without any pre-planning or concentration on when to act”

How is this free will? Free will is a conscious choice to do something, not a subconscious act.

raverbashing• 11 years ago

Really, the experiment seems to boil down to 'the decision happens before it is noticed'

It seems a big jump to extrapolate this to 'there is no such thing as free will'

varjag• 11 years ago

It can also be a case there's a cognitive delay associated with taking note of the time. The mechanism involved into decision making for muscular movements might be dramatically faster than cortical perception and verbalization of time.

eli_gottlieb• 11 years ago

The experiment seems to boil down to, "Processing and propagating signals encoding decisions, to report them to the executive centers where the conscious mind will notice them, takes time."

That's not really a wild statement about free will. It's just another paper showing that brains are made of atoms.

carapat_virulat• 11 years ago

Really I've seen so many terrible psychological experiments that spawned huge articles that I am not taking any of this seriously anymore. I don't doubt that we will get somewhere at some point, but it's very clear that at this point we are mostly bluffing.

Just because you set some kind of experiment that turns up some kind of data shouldn't mean that you compare your studies with science on the level of physics or astronomy. Unless your experiments provide a model that has any kind of predictive power you are probably getting nowhere.

The scientific method got where it got, because it started with people doing small experiments that turned up to generate incredible models with a great predictive power, and started building up from there. Just copying the method, disregarding the predictive aspects, and just bluffing from there to generate nice sounding headlines is hardly a similar path.

return06• 11 years ago

appears to be a bit of bubblegum neuroscience.

Libet's experiments were carefully planned and are highly regarded. That's why they have never been refuted, and in fact spawned a lot of newer studies in the field.

inscrutablemike• 11 years ago

They can be valid experiments that reveal important insights about the workings of human consciousness and still be utterly irrelevant to the question of free will. With 100% certainty, because this is the case.

cristianpascu2• 11 years ago

Much of the confusion about free will and how much free is our will comes from experiments that study limit cases like our body reactions to external stimuli or something of the sort.

But I don't think anyone can seriously state that we are absolutely free, specially when our bodies are involved. Extreme cases reveal that that there's much to say about how we (body+mind) work, and we're certainly not absolutely free.

The simplest and, perhaps, best argument for our partial freedom is that we can ask questions. I don't think there is any kind of property in the physical world that would yield questions and truth-intuition about answers to those questions. Electrons just bump into each other. They don't ask questions about it. They move on like there's no tomorrow.

Retric4• 11 years ago

Software which clearly does not have free will asked a question allowing you to enter that post.

So, if you exclude the language part of asking questions as cruft the type of experiments that human baby's do is the same type of experemnts birds do when learning to fly. But, biological examples of this get really simple and plenty of people have built non biological examples.

At a lower level you can argue that QM shows every time a wave function collapsed a question was asked and answered.

Which is really the issue of free will, most definions boil down to either, 'humans' or include simple systems that clearly lack choice.

rrss1122• 11 years ago

You could think of that software's question as an extension of the creator/owner of the software. Said person is asking for your opinion and delegating to the software the ability to receive your opinion and display it to other people.

StavrosK3• 11 years ago

I don't understand what free will means. Either all your decisions are pre-determined, or they're made at random. Is one better than the other?

coldtea9• 11 years ago

That's a simplistic dichotomy.

Random implies no agent (self) at all. It's just a roll of the dice.

Pre-determined, on the other hand, it's not incompatible with free will (precisely because free-will doesn't mean "randomness" -- it just means it's not imposed upon you by an external agent.

Let's imagine a totally deterministic universe. How could ever pre-determined play well with "not imposed upon you by an external agent"?

Because this pre-determined you is not an external agent imposing its will on a third subject, it's exactly who you are as a "self".

That is, you are your history (and JUST that), and the only decision your self could make is a decision based on its precise history, which is exactly what it makes every time.

Anything else wouldn't be YOU or YOUR free will, it will be arbitrary (random).

So, I think people just have this misconception that "free will" should have the emphasis on "free" -- to mean "whatever".

But the emphasis should be that it's a "will", i.e. the expression of the subject. And such an expression can only be understood in the sense of a subject that has history (you weren't just born this instant) and that makes decisions precisely drawing upon that history -- not just as a source of inspiration of his final decision, but as the total embodiment of his past (with all its aspects) in his decision.

In other words, free will is not you being free to will anything. It's just you being free to be you -- an entity of a concrete past.

stefanix• 11 years ago

About the experiment, how can Libet conclusively derive the decision came after the initial EEG measurement. I mean how did he know the duration it takes to report the decision. The origin of the decision to being able to reflect on it may take a variable amount of time therefore invalidating the claim.

Also the strict separation of conscious/unconsciousness seems simplistic. Most actions may very well be processed subconsciously with very little decision making bubbling up to consciousness. Driving a car comes to mind. This also means when asking a subject to process something impulsively the vast majority of their actions may be subconscious.

tl;dr People spacing out does not invalidate free will.

return06• 11 years ago

The subjects reported the position of the dot, that's how he knew the exact time point in which they were 'aware' of their impending action. There was a comfortable time difference between the two.

More recent experiments have been able to "foresee" actions ~ 10 seconds earlier: https://www.ncbi.nlm.nih.gov/pubmed/18408715

These studies do not invalidate free will, they just show that free will is not something ethereal, but a product of the specific makeup of ourselves, which is encoded in our brains.

nota_bene10• 11 years ago

There is no need for these experiments: There is no free will, and you can understand this by simply applying pure logic/rational thinking:

When you're born (the "1st second" of your life) you're given 2 variables: (1) Your genome and (2) the environment you're born into. You have absolutely no control whatsoever over these 2 variables (with their infinite number of sub-variables they "contain"). And everything else that follows (the "2nd, 3rd, nth... second" of your life) is a function of these 2 initial variables: And this includes your brain that you will use to make the infinite number of choices in your life (both consciously, such as "I'm going to buy milk" and subconsciously, such as "firing neuron X at second Y", "moving atom A to location B", etc.).

EDIT: Downvoters - please state why this would be wrong.

mattmanser• 11 years ago

Sigh. Arm chair philosophy at its worst.

nota_bene10• 11 years ago

What's worst is when people leave worthless comments like these.

interpol_p• 11 years ago

Even if your simplified argument were correct, you still could not conclude that there is no free will.

We all have the ability to act at our own discretion. Does it really matter if it is within a deterministic framework or not?

nota_bene10• 11 years ago

Where you're correct is: We can never prove the absence of free will, because we can never understand then entirety of our environment (how the universe "really works", etc.). But when you take what we know today and apply rational thinking, it certainly looks like there is no free will.

We all have the ability to act at our own discretion.

That is exactly what my argument would deny. There seems to be not even an inch of free will.

Does it really matter

100% percent yes, for example: The absence of free will forces us to stop judging people, simply because the concept of "guilt" (on the negative side) and the concept of "merit" (on the positive side) are proven to be entirely baseless. If we applied this to our culture, we'd live in a totally different world.

return06• 11 years ago

The determinism kind of removes the "free" from free will. Determinism means that, if someone else knows your internal state, they can predict, or worse, manipulate your free will. People are naturally extremely personal and possessive about their free will, and these experiments are "stealing" it away. That's why the knee jerk reaction to these experiments is always extremely negative.

return06• 11 years ago

In 2015 this should be considered a no-brainer, i'm baffled at the blatantly unscientific responses in almost all the comments in these threads.

ucaetano3• 11 years ago

Depends on how you define free will:

1) Free will is the capacity to act at our own discretion, independent of our current surrounding and context, including the physical state of our body and brain

2) Free will is the capacity to act at our own discretion, independent of our current surroundings and context, not including the physical state of our body and brain

Note: In both cases I'm assuming there is no soul, mind beyond the body, etc.

If you define free will as (1), then there is no free will, since any decision will always be subject to the physical state of my body and brain (except, of course, if there's something not physical in play as described in my note).

If you define free will as (2), then there is free will, since despite the context surrounding you, you can still make decisions independent of it.

The same could be applied to a state-machine (finite or not): 1) Free will is the capacity of a machine to produce outputs independent of it's current internal states and the external inputs 2) Free will is the capacity of a machine to produce outputs independent of it's external inputs

(1) means no state-machines have free will, (2) means they can have free will.

coldtea9• 11 years ago

If you define free will as (2), then there is free will, since despite the context surrounding you, you can still make decisions independent of it.

How do you come to that? Obviously the state of our current surroundings and past histories determines the state of our brain and body (down to the firing of the very last neuron).

nota_bene10• 11 years ago

The issue with (2) is that the physical state of our body and brain is a function of all past and current surroundings and context (including our genome). Which means it can never be independent, but is instead fully defined.

chm• 11 years ago

This assumes that you were capable of conscious decision making, or had free will, at birth. The assumption is the problem, I think. I would think that "free will" is developed, not de facto available from birth. In fact, I think free will in a baby could be a pretty bad evolutionary disadvantage, whereas it would be a great advantage for an adult.

nota_bene10• 11 years ago

I would think that "free will" is developed

If you see life as a 4-dimensional network of variables linked through "cause and effect", then there is no place for a "free" variable which could introduce free will (or its development). Unless you introduce invented concepts such as "soul" or "god", etc., which means that you quit the rational discussion altogether.

Someone• 11 years ago

"And everything else that follows (the "2nd, 3rd, nth... second" of your life) is a function of these 2 initial variables"

That's an opinion sprouting from a view on the world that states free will doesn't exist. It may be true, but goes counter to many people's intuition that there is a free will.

A good argument would explain how, from the assumption of 'no free will', we can arrive at such things as 'property' and 'crime'. Why would societies sprout up in which people get imprisoned for using stuff that they didn't make themselves, or that they didn't get permission to use for from those who made it?

My argument would be that this is just evolution: local optimization of the non-physical world, in the same way evolution gave us all kind of complex physical structures. On the one hand, I can't see much fault in that, but on the other, giving up the idea of free will means reintroducing predestination, accepting that no criminal can be blamed for their crimes, no Isaac Newton or Olympic champion can be praised for their accomplishments, that capital punishment and war are just things that happen, etc, and even accepting that there is no such thing as 'me'.

So, yes, if physics as we know it is correct, logically, I think you are right, but emotionally, I don't believe it. That might have to mean that somewhere in our bodies there is a part that can bend the laws of quantum physics as we know them.

lawtguy• 11 years ago

This world view is a very 19th century clockwork universe one: if we know all the initial conditions in sufficient detail, we can predict the outcome. There are two problems with this: at a fundamental level, quantum mechanics says you can't know the initial conditions with infinite precision[0]. Second, even at a classical physics level non-linear systems with very similar starting points can have very different final states. As a practical matter that means even if the rules are deterministic, you cannot predict with 100% accuracy what the outcome will be.

So we have a complex non-linear system (your genes) and a very very complex non-linear system (the environment) that make a very complex non-linear system (your brain). Because we have all of these complex non-linear system interacting together, we cannot determine what will happen in any practical sense. We can probably make some observations that will generally hold about some decisions a human brain will make (e.g. it's been a few hours, it will be hungry and decide to eat), but these are more probabilistic rather than 100% concrete deterministic rules.

0: https://en.wikipedia.org/wiki/Bell%27s_theorem

wbillingsley2• 11 years ago

Libet's study is terrible -- as with many of these studies it's famous for its controversial conclusion, but its flaws are very obvious.

He specifically asked his subjects to listen for an urge rather than applying free will ("let the urge [to move] appear on its own at any time without any pre-planning or concentration on when to act”). That the urge is measurable before the decision, then, just means his subjects did what he asked them to -- to wait for a feeling of an urge.

And frankly, it's most likely the urge was the tension of having sat there a while, knowing someone's expecting you to feel an urge to move your arm...

1. Ask subjects to wait for an urge to react to (and therefore not apply free will)

2. (Nice-sounding but actually irrelevant measurement set-up with dots, clocks and electrodes)

3. Observe subjects decided after a measurable urge

4. Conclude free will doesn't exist, rather than that your subjects did what you asked them to in step 1.

return06• 11 years ago

Thats a simplistic refutation. There is a long line of studies on the subject summarized in wikipedia here: https://en.wikipedia.org/wiki/Neuroscience_of_free_will

wbillingsley2• 11 years ago

Yes, indeed it is a simple flaw. One of my criticisms of the paper, given how popular it is to write about it in the media, is quite how obvious the flaws are.

coldtea9• 11 years ago

As stefanix writes:

Also the strict separation of conscious/unconsciousness seems simplistic. Most actions may very well be processed subconsciously with very little decision making bubbling up to consciousness.

This. The "unconscious" processing an answer/reaction belongs to the same entity that the conscious part belongs too.

Implying we're some kind of automata because reactions can come pre-processed by the subconscious part misses the point.

The subconscious part is the same "self" that the conscious part is, it's not just what we can think "out loud" in our head that counts as us taking a decision.

evanpw• 11 years ago

This is scientific evidence, not mathematical proof.

One way to frame the concept of free will is that some subjective experiences cause rather than are caused by or are identical with, physical changes in the brain.

One way to test for causation is to see which one happens first. If free will (at least as defined above) is true, we expect that some subjective experiences should precede the corresponding physical change in the brain. If free will is false, then we should expect subjective experiences to be simultaneous with or come after the corresponding brain change.

Since the actual result of the experiment (brain change comes before subjective experience) is less likely under the "free will" hypothesis than it is under the alternative, by Bayes' rule, it should make you less confident of the existence of free will. If you believe that this result would be only slightly less likely even if free will were true (as you apparently do), then you should update your estimate only a little bit, but this is still the right kind of experiment to run if you want to test for free will.

coldtea9• 11 years ago

One way to test for causation is to see which one happens first. If free will (at least as defined above) is true, we expect that some subjective experiences should precede the corresponding physical change in the brain. If free will is false, then we should expect subjective experiences to be simultaneous with or come after the corresponding brain change.

Only if you assume that "free will" implies only conscious decisions.

Your subconscious brain is part of the same self as "you". So whatever you think as your "free will" is also part of that, just that part of it plays as "voice inside your brain" (conscious thoughts) and some of it does not.

It's not like the subscoscious processing is done by some other entity and imposed upon you.

danbruc3• 11 years ago

In my opinion it is almost trivial to see that something is fishy with the idea of free will - just try to formalize it.

Assume you are sitting in a restaurant facing the decision to either order a steak or some pasta. One possibility is that your decision is a deterministic choice, a function of the current state of your body (What nutrients do you need?), your brain (What are your past experiences with steak and pasta?) and your environment (What does the steak on the neighboring table look like?). This does not resemble what I would call free will.

The other extreme is that your decision is completely random and not influenced by anything. This assumes that there is real randomness in our universe which is, as far as I know, an open question. There is no real randomness in classical mechanics, only apparent randomness due to ignorance of microscopic degrees of freedom. Quantum mechanics seems to have probabilistic features but they are, as far as I can tell, at odds with the unitary evolution of quantum systems and it remains to be seen whether there is real randomness in quantum mechanics or not.

But lets just assume there is real randomness, at worst, if there is only apparent randomness, this option becomes deterministic and degenerates into the first option. There is still some freedom in this option, namely the probability distribution over the different choices. This probability distribution may just be what it is for no deeper reason, a fundamental property of the source of randomness. In this case I wouldn't call it free will, too, because the choice is entirely random.

It may also be the case that the probability distribution gets shaped in a deterministic way. The steak on the neighboring table looks really good making it more likely that you choose it but there is still some probability that you will choose pasta. This is kind of a middle ground between the first two options, the final choice is random but the probabilities reflect your current and past states and the state of the environment in a deterministic way. But again I would not call this free will.

So what would free will have to look like? The choice must not entirely depend on the current state of you and your environment but it must also not be completely independent of it, i.e. be completely random. I spent quite some time thinking about this but I am completely unable to come up with something that is in some sense between deterministic and random (including deterministically shaped randomness). Am I - or even everyone - missing a (fundamental) third option? Is thinking about free will in terms of systems and states and state changes in some way inappropriate? For the moment I will side with the people denying the existence of free will, if someone can formalize what free will really means I will reconsider things.

vixen99• 11 years ago

Which leaves criminals where?

robgibbons• 11 years ago

The law of causality is ultimately incompatible with free will. You cannot have truly autonomous agency within a physical framework ruled by cause and effect. To truly express freedom of will, without influence from any prior cause, you would literally need to exist in your own isolated vacuum of space and time (no outside influence from environment), having also created yourself (no outside influence from genetics). That is to say, if you believe you have free will, the burden is on you to prove that your body, and your brain itself, are not bound by the laws of the physically causal universe.

skwosh• 11 years ago

Our perception of time is an also at odds with our scientific understanding of the universe, but that doesn't mean that it isn't real...

Free will experiments reveal how little we know about our minds

Related Stories