Can experiments settle the free will debate? | Julian Baggini, Peter Godfrey-Smith, Sarah Garfinkel

free will in a meaningful sense is having a degree of control and agency and that is always a matter of degree it's not you have it or you don't right [Music] I suppose for example we had an experiment um or rather some kind of finding whereby some scientists has developed the technology to predict with 100 certainty they scan somebody's brain they track every single position of

every atom in the brain and which direction neurons are firing and all of this kind of thing and they're able with some computer program to predict with 100 accuracy exactly what a human being will do over the next five minutes and every time they run it it predicts it with 100 accuracy is this the kind of thing that would make you skeptical that people are free

it would I think that that thought is always quite scary to people isn't it we don't like to think we're that predictable but predictability is in a sense a bit of a red herring there it's a bit like the idea you know it's a third theological argument that um if God God knows what you're going to do before you've even done it then how can your

actions be free right because the idea here seems to be that if it's predictable it's somehow you know fixed in the wrong kind of way but um sometimes I I make predictable free choices all the time right so if you see me going to order a coffee and you know what kind of coffee I like I don't literally order the same one every time but you'll

probably be able to predict what I'm going to order you're certainly going to predict I'm not going to order a caramel latte Grande Maximo whatever it's called um that's very very predictable it's entirely predictable but why does that mean it's not free in fact the reason it's freeze is because I mean Hume argued that it's the predictability which shows you it's free as if you like

that it's because my choices are driven by my my desires my personality and these things have a meaningful stability that's what makes the choice of the coffee meaningful and that in turn is what makes it somewhat predictable so I I don't I don't think we should get to sort of concern that predictability is in itself the big threat to Free Will Peter do you agree with

Julian on this if we have this computer that can predict I could I could I could write down on a bit of paper right now every word that you're about to say and answer this question and once you've said it I turned it around and said I predicted it with 100 accuracy would that be traveling to you [Laughter] a little disconcerting uh I agree but but

I agree more with Julian that predictability uh and freedom don't have a particularly antagonistic relationship between the coffee example I accept I think the sorts of things that are scientifically uh in principle showable that would really be destabilizing or undermining a free will would be work on the mechanics that showed that the kind of belief desire interaction that on the milder conceptions of free choice is

what has to underlie a free Behavior you know you want this you believe this this action makes sense in terms of your police and you desires so you do that if that's the model of an unconstrained reasonable free choice if it turns out that lots of what actually generates our behavior is quite different from that and the thing that we think of as the belief desire

story that might be told is a kind of confabulated rationalization we're just storytelling when we engage in those sorts of interpretations if that's showable then I'm bothered as we're talking here remembering experiment I I read about years ago I don't know if this has been followed up I think it's called the pantyhose experiment uh this was from some decades ago where the idea was there are

certain kinds of predictability in people's product choices if you line things up on a table I think it's a right hand preference or something like that people reliably go for a particular the object in a particular part of the display and you can do an experiment with the four items in the display are physically identical they'll go for a particular one pretty reliably and you ask

them why and they'll cook up a story they'll find it we'll say Oh I thought that one just had a nicer texture or it just the the particular color appealed to me even though the experimenter is quite confident that there's no physical difference between these things so that's a situation where there's a disconnect between the store the the rationalizing story that the person has in their

mind either before the fact or after the fact or both and the mechanics behind the actual thing they did now if there was tons of that if that was something that was all over Human Action where it's not predictability per se it's the fact that what predicts the choice is something that's far outside of what looks like the kind of rationalizing belief desire based story if

that was all over the place I would be more than disconcerted I think that really would be that would be trouble as scientists and we off we see self-report as notoriously unreliable like we as a neuroscientist I like including self-report variables in my experiments often because I find it fascinating how they don't necessarily align up with um how confidence doesn't severally line up with action and

you get these lovely divergences and and yes we do create narratives after the fact and there is a reason why um self-report measures deviate from uh actual mapping of a neural and physiological measures and you see that more for self-related Concepts than you do for extra receptive so we have the X receptor system which is how to navigate and understand the world we have the Interceptor

system which is how to understand ourselves um and our internal signals and you can do experiments where you look at people's ability to be confident that they've seen felt heard an extra sepsis signal and that nicely corresponds to confidence you get a nice confidence accuracy correspondence but with Interceptor signals about what's happening in our cells you get this massive Divergence between confidence and accuracy for self-related

signals and that's potentially because there's been a Divergence in our extra sector signals have helped us navigate the world whereas we're constantly bombarded by signals in our body to our brains all the time and we don't necessarily want to be aware of them influence them by them unless they've reached a threshold of hunger and other types of homeostatic needs so you do get this potential Divergence

between the the Interceptor signal and conscious access it's my recollection of that older experiment with the display is that accurate yeah yeah I've heard I've heard that vague one which is absolutely people will narrate things afterwards yeah but I want to say about that was that yeah sure we we do confabulate we do create these stories absolutely and and we underestimate how much of our choices

are like that but the thing about these experiments right is that people go oh my God this is extraordinary just because it's to the right these experiments are usually around things which are like trivial or unimportant or marginal right so you know even the libit thing you know press a button for no good reason whatsoever at any time right how and and that experiment is meant

to tell you about whether you have free will about you know the person you choose to have your children with or or stuff like that pantyhose oh so people will choose the pantyhose to the right rather than this do people go into the shops and we need to buy pantyhose and walk out with socks because the display no they don't right so I I think that

so often the problem here one of the problems here is that it's very very easy to show that we're swayable and persuadable by things we're not aware of but the experimenters tend to be several things are more important things so let's take something more serious voting right I think there is some evidence that um the order of the candidates on the ballot paper has that has

a statistically significant effect on how many votes they have but but that does not mean that 99 90 people are going there know who they want to vote for they find the candidate and they cross it what's happening is the sort of like 10 of people who are just undecided or not very thoughtful go in and a percentage of them are influenced by the position on

the ballot paper and if you statistically analyze it that turns out to be significant it tells you it doesn't tell you that most people going into vote aren't voting for someone who they've chosen to vote for because that candidate they believe most of lines with their values so you know generalizing from these very very contrived laboratory experiments to the big important decisions on life is is

um is a huge loop as an experimentalist I completely agree with that like uh we need to have these things in the lab in order to study them but I absolutely believe they're contrived as well and we can exercise there's different circuitry that underlies this automatic type processing and more controlled processing like the prefrontal cortex sizes inhibitory control over these more automatic signals so absolutely it's

hard to investigate consequential choices in an experiment because you don't want too much to depend on the downstream consequences of what you do in an experiment yeah we're limited yeah this kind of uh Insight shows us you know the idea that there's a somewhat statistically reliable indicator of voting based on the order of candidates on on a form and this isn't enough to determine that somebody

will always vote for the first person but it's it's one of a great deal of influencing factors but let's say we just started to consider what the other influencing factors are so the fact that the first name on the on the ballot uh that's not enough to determine somebody voting but it has a small influence in that direction so why someone who does tick that name

why else might they take that name well maybe because of their upbringing and the kind of political uh culture that they're born into they have no control over that that's quite a big influence over who they vote over maybe they watched a television debate and they and someone did particularly well and and maybe like they didn't control whether the TV was on or they walked home

and the show happened to be on maybe their parents were watching it or something like this you know if you consider all of these kinds of factors and we we discovered that the majority or even all of the factors when you consider all of them are the kinds of things like the order of the the ballot that we just don't have control over but it turns

out that it's things of that kind which overall do in fact determine ultimately why the person uh ticked that box there might be lots of different factors but if we're equally not in control of any of those factors can we really say that we're in control of who we take to continue watching this video click the link in the top left or in the description below

or visit iai.tv for more debates and talks from the world's leading thinkers on today's biggest ideas