Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Future Actions Predicted From Brain Activity

timothy posted more than 3 years ago | from the let-me-tell-you-about-my-mother dept.

Science 72

An anonymous reader writes "Bringing the real world into the brain scanner, researchers say they can now determine the action a person was planning, mere moments before that action is actually executed. In the study at the University of Western Ontario, human subjects had their brain activity scanned using functional magnetic resonance imaging (fMRI) while they performed one of three hand movements. By using the signals from many brain regions, the researchers could predict, better than chance, which of the actions the volunteer was merely intending to do, seconds later."

Sorry! There are no comments related to the filter you selected.

First! (-1)

Eponymous Hero (2090636) | more than 3 years ago | (#36627262)

Pre-Crime begins...

Re:First! (0, Funny)

Anonymous Coward | more than 3 years ago | (#36627274)

I knew you were going to post that...

Re:First! (2)

flaming error (1041742) | more than 3 years ago | (#36627546)

I have to admit I didn't know. I thought he was going to post about "thought crime."

"Do you know what that means? This damn thing doesn't work!"

Re:First! (1)

kvezach (1199717) | more than 3 years ago | (#36627672)

Yours is the minority report!

Re:First! (1)

tehcyder (746570) | more than 3 years ago | (#36631670)

Pre-Crime begins...

Good. Prevention is better than cure.

Re:First! (1)

Eponymous Hero (2090636) | more than 3 years ago | (#36634836)

ok, let's go with your plan. only safe bet is to eliminate the human species. let's get started, you first. wait, crap that was me. dammit!

"mere" moments? (1)

Anonymous Coward | more than 3 years ago | (#36627304)

Does anyone else interpret "mere" moments as implying that a shorter amount of time is better? Wouldn't being able to detect an action *longer* before it actually happens be better?

Re:"mere" moments? (0)

Anonymous Coward | more than 3 years ago | (#36627844)

Yeah, it does seem to embelish a little. But my own brain resolved that it's probably not possible to move beyond second into the minute range. (i wonder if they were able to predict that in test subjects, and edited the article on that basis)

Beyond conceptual planning, there are few physical actions you can anticipate in a person more than seconds into the future. I would say, only things like "going for a jog" or "taking a shit" would traverse beyond seconds of anticipation.

Based on the constant change of visual input and all of the implicit dependencies therein, it's probably really tough to make accurate predictions on real world movements.

Re:"mere" moments? (1)

jonahbron (2278074) | more than 3 years ago | (#36627856)

I would agree, but it probably means that they can do it "mere moments before", as opposed to "processing the data for hours and then checking the test subject's intentions".

Re:"mere" moments? (1)

milkmage (795746) | more than 3 years ago | (#36627860)

depends on where they want to implement the technology.

not sure i'd want my arms/legs/hands moving much before I actually intended

"Being able to predict a human’s desired movements using brain signals takes us one step closer to using those signals to control prosthetic limbs in movement-impaired patient populations, like those who suffer from spinal cord injuries or locked-in syndrome.”

Re:"mere" moments? (1)

tmosley (996283) | more than 3 years ago | (#36627932)

You're just jealous because I can bend spoons using only my hands.

Re:"mere" moments? (0)

Anonymous Coward | more than 3 years ago | (#36634268)

Does anyone else interpret "mere" moments as implying that a shorter amount of time is better? Wouldn't being able to detect an action *longer* before it actually happens be better?

They are intentionally glossing over the fact that they are not reading your "mind", as in your actual thoughts and intentions. They are reading the precursor triggers to muscle movements, so while they can tell what you "intend" to do it's not until you actually begin to do it. Pretty cool for input technology down the road, but not so great for "mind reading".

Art versus real life (2)

macraig (621737) | more than 3 years ago | (#36627306)

Once again art wins the contest by default, when real life plagiarizes it (Minority Report).

Re:Art versus real life (0)

Anonymous Coward | more than 3 years ago | (#36627836)

Are you calling Minority Report art?

Sweet Jesus...

Re:Art versus real life (1)

Anonymous Coward | more than 3 years ago | (#36627974)

It was an excellent short story before it was a mediocre movie.

Re:Art versus real life (1)

tehcyder (746570) | more than 3 years ago | (#36632296)

It was an excellent short story before it was a mediocre movie.

I think you're being a bit generous to the movie there.

Re:Art versus real life (0)

Anonymous Coward | more than 3 years ago | (#36639790)

I suppose so. It's probably because I've never seen it, so therefore I can be generous without any emotional distress. If you asked me about Phantom Menace, on the other hand....

Re:Art versus real life (1)

smithmc (451373) | more than 3 years ago | (#36642184)

It was an excellent short story before it was a mediocre movie.

Indeed, and that isn't the first time. Let's not blame a brilliant guy like Philip K. Dick for the Hollywood atrocities committed in his name.

Re:Art versus real life (1)

dakameleon (1126377) | more than 3 years ago | (#36629376)

"Precognition"/predicting the future is prior art on "detecting electrical signals in the brain"?

"Better than chance"? (0)

Anonymous Coward | more than 3 years ago | (#36627380)

Anyone have a link to the actual study, so we can find out what "better than chance" really means? 40% accuracy? 35% accuracy? Either could be significantly better than chance (33%), but neither shows much promise on the "I'm going to read your mind" front.

Unrelated, but why did slashdot decide to change the UI again? It's even less responsive now than their last change!

Re:"Better than chance"? (1)

Naveen Gupta (970308) | more than 3 years ago | (#36627450)

Anyone have a link to the actual study, so we can find out what "better than chance" really means? 40% accuracy? 35% accuracy? Either could be significantly better than chance (33%), but neither shows much promise on the "I'm going to read your mind" front.

Unrelated, but why did slashdot decide to change the UI again? It's even less responsive now than their last change!

http://communications.uwo.ca/western_news/stories/2011/June/brain_research_predicts_premeditated_actions.pdf [communications.uwo.ca]

Re:"Better than chance"? (1)

PC and Sony Fanboy (1248258) | more than 3 years ago | (#36628160)

... better than chance is a pretty general science term that means "Yes, we can do it. Not perfectly, but better than chance."

In the world of science reporting, it means "Eh, not by accident, but not worth writing home about". If they could predict it at 95% accuracy, you bet they'd be reporting it. It's just the way science journalism works.

Re:"Better than chance"? (1)

tehcyder (746570) | more than 3 years ago | (#36632332)

Unrelated, but why did slashdot decide to change the UI again? It's even less responsive now than their last change!

I think they're trying to put into practice the programming philosophy of "if you take an infinite number of monkeys bashing at an infinite number of computer keyboards, eventually you will produce a Hello World program with a nice bevelled grey box round it."

If only... (3, Insightful)

N0Man74 (1620447) | more than 3 years ago | (#36627424)

If only they could have used this technology to predict them creating this technology.

Re:If only... (1)

FredFredrickson (1177871) | more than 3 years ago | (#36634768)

As we all know, this causes a temporal paradox, and since this wibbly wobbly timey wimey stuff isn't quite a straight line, this is ok.

Old news (1)

Anonymous Coward | more than 3 years ago | (#36627432)

This has been reported before, I forget what the research was, but "predicting mistakes before you make them" found that the decisions you make before proceeding against your better judgement (omg they found the little angel and devil shoulder imps via the MRI..) This is just the same research again, but focused on movements.

The logical outcome of this research is figuring out how to tap it for cybernetics.

Re:Old news (1)

Em Adespoton (792954) | more than 3 years ago | (#36628170)

Indeed... an interesting implant would be something that triggers the running reflex when a certain sound sequence (gunshot) is heard. This would give sprinters a definite edge.

I'm sure the same concept could be applied to many other fields where a certain reaction to specific stimulus is required to be as fast as possible.

Re:Old news (1)

Concerned Onlooker (473481) | more than 3 years ago | (#36631248)

Not only is this old news, they didn't even credit my mom, who when I was little always seemed to know what I was going to do even before I did.

stuipedo (1)

Anonymous Coward | more than 3 years ago | (#36627436)

Hilter used to claim this as well. BS Technology.

Paper Rock Scissors? (0)

Anonymous Coward | more than 3 years ago | (#36627464)

A little elaborate of a game setup, isn't it?

Re:Paper Rock Scissors? (1)

carpenoctem63141 (2266368) | more than 3 years ago | (#36627508)

They should have done the ESP card test from Ghostbusters.

Re:Paper Rock Scissors? (1)

flaming error (1041742) | more than 3 years ago | (#36628090)

Who says they didn' OUCH! Hey!

Re:Paper Rock Scissors? (1)

Jeng (926980) | more than 3 years ago | (#36627516)

Yes, but now with the right equipment we can cheat at it!

Re:Paper Rock Scissors? (1)

RivenAleem (1590553) | more than 3 years ago | (#36632406)

I'll be interested when they can determine between 5 possible outcomes, so I can beat Sheldon.

You're Kidding? (0)

Anonymous Coward | more than 3 years ago | (#36627466)

By monitoring which parts of the brain are active they can use that knowledge to predict hand gestures. Why, it is almost like these people are making a conscious choice before moving their hands.

Re:You're Kidding? (2)

milkmage (795746) | more than 3 years ago | (#36627810)

The team found that by using the signals from many brain regions, they could predict, better than chance, which of the actions the volunteer was merely intending to do, seconds later.

key word there is intend - sounds like they're able to detect/predict before a conscious decision was made.

Re:You're Kidding? (2)

narcc (412956) | more than 3 years ago | (#36627870)

key word there is intend - sounds like they're able to detect/predict before a conscious decision was made.

The key word is indeed "intend" -- though that implies a conscious decision was made which was then predicted.

"Mere Moments" (4, Insightful)

Anonymous Coward | more than 3 years ago | (#36627518)

This is far from predicting future actions. I, myself, know what I am going to do moments before I do it. That is because I can think faster than I can act.

  It can only predict my actions before I do them, not before I think them. This isn't the start of pre-crime, because that would require planning out your actions before you do. This is not that, and it's not even a step towards that.

Re:"Mere Moments" (1)

Tyr07 (2300912) | more than 3 years ago | (#36629028)

This is far from predicting future actions. I, myself, know what I am going to do moments before I do it. That is because I can think faster than I can act.

It can only predict my actions before I do them, not before I think them. This isn't the start of pre-crime, because that would require planning out your actions before you do. This is not that, and it's not even a step towards that.

That's all someone needs to know you're planning on pulling that trigger and to shoot you first. Actually, that's all a COMPUTER needs to know if you're going to comply or not. Say hello to Robocop v2

Re:"Mere Moments" (1)

marcosdumay (620877) | more than 3 years ago | (#36629772)

The obvious application is to create a computer that reads your mind and moves your (artificial legs | exoskeleton | remote hight power actuator | watever) when you expect it to move.

Re:"Mere Moments" (1)

Purple Peril (110467) | more than 3 years ago | (#36629992)

But imagine a parolee forced to wear a device that incapacitates him if he intends to fire a gun, or a patient on suicide watch.

Or, imagine a remote-sensing device the police could use to sense whether a hostage-taker intends to pull the trigger.

Re:"Mere Moments" (0)

Anonymous Coward | more than 3 years ago | (#36630702)

While I didn't RTFA (and I strongly guess the summary is missing the point), experiments that determine what you think _before_ you do (yes, before you think it, not before you can act) are nothing new. It's not about predicting what you are going to do 10 hours in advance, but just before your consciousness makes up an explanation that conforms with your decision.

Re:"Mere Moments" (0)

Anonymous Coward | more than 3 years ago | (#36631854)

And how do you know this? How can you know for sure that your subconscious doesn't first make the decision and then send it to your conscious mind to rationalise it? For me it's impossible to know if I actually decide things when I think actively, or if it is like moving the hand away from something hot where the reflex moves the hand before you have time to notice the heat and make a decision to move your hand.

Rufus Riley (1)

king0lag (1243544) | more than 3 years ago | (#36627554)

Those thoughts about my cousin Elena, those were *just thoughts*!

This would be handy... (0)

Anonymous Coward | more than 3 years ago | (#36627584)

Since I don't ever plan anything, this could actually make my decisions easier by telling me what I'm going to wind up doing AHEAD of time - then I can just do THAT. I've found that for binary actions, a coin is able to predict what I'm going to do 100% of the time, but when I'm confronted with more than two options, I am paralyzed with indecision!

Re:This would be handy... (1)

JonySuede (1908576) | more than 3 years ago | (#36628320)

try the d100

How soon (0)

Anonymous Coward | more than 3 years ago | (#36627616)

I'm sure I saw in some BBC program (Horizon?) a claim that it was possible to use this technology to predict a person's action not only before they did it but before they even became *aware* of the decision they were going to make

Either bullshit or quite unsettling.

Re:How soon (3, Informative)

Daetrin (576516) | more than 3 years ago | (#36628214)

I'm sure I saw in some BBC program (Horizon?) a claim that it was possible to use this technology to predict a person's action not only before they did it but before they even became *aware* of the decision they were going to make

Either bullshit or quite unsettling.

There is a middle ground in this case between "untrue" and "bad." I've seen the same studies, showing (theoretically) that what we're going to do is already decided before we actually carry out the action, but it doesn't necessarily negate the idea of freewill or anything like that. The mind is not a simple machine, it is incredibly complex and your consciousness only represents, and is only aware of, a fraction of all the activity going on. There are subconscious feelings you don't fully realize you have or don't realize how they will impact certain decisions, there are autonomous responses that will kick in before you get the chance to think about an action, there are trained reflexes that short-circuit the regular decision making process. At least one such study claims that many actions are started "automatically" but higher levels of consciousness then have the option to veto it before it's actually carried out. This is probably exactly how people can literally have second thoughts about actions (although for a lot of people there are unfortunate cases where the conscious mind is a little late and tries to veto the action after it's already being carried out.)

So no, neither our thoughts nor our actions are under 100% conscious control all the time, but that's not anything that anyone with even a basic understanding of human nature didn't know already.

Ro-Sham-Bo (1)

MDMurphy (208495) | more than 3 years ago | (#36627620)

They predicted one of three hand movements and it wasn't Rock, Paper, Scissors? They missed a huge opportunity to upend the RPS wagering market

Re:Ro-Sham-Bo (1)

bistromath007 (1253428) | more than 3 years ago | (#36627734)

The article doesn't say it wasn't. The upshot of all this is that input prediction isn't just for arcade fighter bosses anymore.

Old news? (0)

Anonymous Coward | more than 3 years ago | (#36627730)

I remember this being posted a long time ago. I don't know if it was on here or not, but I definitely remember it. In fact, I remember something that wasn't mentioned here: that the researchers couldn't account for last-minute mind changing.

Well, yeah (0)

Anonymous Coward | more than 3 years ago | (#36627838)

If you have no brain activity, you are unlikely to take actions in the future.

Re:Well, yeah (1)

PC and Sony Fanboy (1248258) | more than 3 years ago | (#36628184)

If you have no brain activity, you are unlikely to take actions in the future.

Someone tell the republicans.

Rock Paper Scissors (1)

jacerm (1787262) | more than 3 years ago | (#36628048)

Will never be the same.

Pre crime? (0)

Anonymous Coward | more than 3 years ago | (#36628108)

Bunch of lunetic scientists on the loose. What benefit does this give the human race? Oh I forgot the pre crime god wanna be's will soon be in power

I *am* a neuro-scientist, and... (1)

Anonymous Coward | more than 3 years ago | (#36628312)

I *am* a neuro-scientist, though I focus more on the visual areas at the back of the brain, rather than the planning area at the front.

This stuff is kinda shiny, but nothing all that new. We can already pick up the brain activity relating to motor actions, both with fMRI and the decidedly more portable eeg. Heck, we can pull this classification trick on the visual system and determine what people are seeing.

The real story here is not being able to determine what people are doing - or even that we can do so before they do it - but rather that we can identify which regions of the brain are involved in planning tasks.

Yes, this allows us neuroscientists to then go on to do more interesting things like design systems to control robotic limbs, but it also enables us to ask interesting questions such as how does experience/learning effect the behaviour of these bits of brain.

One end point of this (amongst others) is to figure out how the brain works enough that we can duplicate useful techniques for use with artificial intelligence.

Re:I *am* a neuro-scientist, and... (0)

Anonymous Coward | more than 3 years ago | (#36629774)

Yes I agree that knowing what brain regions is responsible is good but we already have a good idea of what regions are responsible for deciding motor movements. There is much to learn about the process but we need to delve a little deeper than what an fMRI is going to tell us. Especially if it is only better than chance. If I was trying to model how the brain accomplishes motor decision then I would toss this paper aside and move onto another that was more informative.

I am hoping that there was more to the paper than the article suggested.

Re:I *am* a neuro-scientist, and... (1)

ahavatar (1672510) | more than 3 years ago | (#36629876)

Yes, this allows us neuroscientists to then go on to do more interesting things like design systems to control robotic limbs, but it also enables us to ask interesting questions such as how does experience/learning effect the behaviour of these bits of brain.

One end point of this (amongst others) is to figure out how the brain works enough that we can duplicate useful techniques for use with artificial intelligence.

As some scholar already pointed out, I doubt that your approach works at all. For example, if I give you an excellent scope that detects every electro-magnetic activities of a personal computer, can you build(or design) a personal computer from such data of electro-magnetic activities? A human brain is some orders of magnitude more complex than a personal computer. I bet that you can't figure out a personal computer with such an approach. Your neuro-science approach can't work.

Re:I *am* a neuro-scientist, and... (0)

Anonymous Coward | more than 3 years ago | (#36631794)

For example, if I give you an excellent scope that detects every electro-magnetic activities of a personal computer, can you build(or design) a personal computer from such data of electro-magnetic activities?

Stop using straw man arguments. This isn't about building a new brain, and was never anything so ambitious. This is about getting an idea of the processes; a much lower goal and quite achievable.

For example: with the hypothetical scope I could detect which parts of the pc were active during particular types of processing. From this I could determine which bit the graphics card was by running some graphics intensive program, I could determine which bit of the CPU had the floating point unit by firing off some particularly floating point intensive code, etc.

To further extend it I could watch these measures while some process was going on and see when things were done on the CPU and when they were farmed out to the GPU to, for example, see if a process made use of parallelisation with GPGPU techniques.

The same applies to neuro-science. We can replace the "hypothetical scope" with EEG or fMRI (as appropriate) and get an idea of which bits of brain to what sort of processing, and then look at what order they get involved in a task. This can show us things like how the brain gets around object recognition being a nasty NP type problem by using attention to narrow the possibilities.

Neuroscience and Free Will (0)

Anonymous Coward | more than 3 years ago | (#36628462)

Reminds me of Neuroscience and Free Will [youtube.com] [Youtube]
Also Neuropath [amazon.com] [Amazon] is a fun book about this taken to the extreme.

I've seen something spookier than this (0)

Anonymous Coward | more than 3 years ago | (#36628806)

I saw something about a study where subjects were wired up, then asked to press a button at a random moment.

It turned out the machine could detect a build up of activity prior to the button push before the subject was consciously aware that they had decided to push the button.

uh oh (1)

ideaz (1981092) | more than 3 years ago | (#36628940)

and all this while I didn't realize that my plans and contemplations were hidden from by brain.

Not that it works, but how well (1)

yet another coward (510) | more than 3 years ago | (#36629552)

That these signals exist in our brain is no marvel. It is obvious. The news coverage focuses on the wrong aspect of the research. The question is how well their technique works. A device to control a prosthesis better have very high accuracy.

Better than chance doesn't mean good (0)

Anonymous Coward | more than 3 years ago | (#36629734)

Being able to predict a human’s desired movements using brain signals takes us one step closer to using those signals to control prosthetic limbs

An arm has six degrees of freedom. Better than chance only means that it has to be correct more than 17% of the time. What does it do the other 80% of the time? "Pardon me, I meant to shake your hand not grab your breast."

The action is already in motion... (0)

Anonymous Coward | more than 3 years ago | (#36630172)

It's like looking at a house being constructed and then declare: I predict there will be a new house here.

Pre-crime is already here, we just need to..... (1)

threeseas (2245516) | more than 3 years ago | (#36630226)

.....enforce it. I mean liked who doesn't yet know politicians will lie? And lawyers... well them lawyer jokes are not just pulled out of the air....

Advances in Input (1)

Reteo Varala (743) | more than 3 years ago | (#36630642)

So far, I haven't seen anyone mention one very useful application of said technology: Advanced input devices. Think about it. If a computer could predict your actions even a second before you do them, then the system can use this data to keep pace with your actions. Who needs a tablet interface, when one can draw on a piece of paper, and the desktop reacts accordingly? What's the point of a touchpad or mouse when one can just move a finger or two over the tabletop next to the computer? Any monitor can be used as a touchscreen, as there is no need for a sensitive layer.

Interesting Article (1)

fatimasb (1995762) | more than 3 years ago | (#36634010)

What an article Can't wait until they actually start predicting when people are going to commit a crime before someone gets robbed or killed. http://www.bbcleaningservice.com/ [bbcleaningservice.com] [bbcleaningservice.com] [bbcleaningservice.com]

Three Hand Actions? (1)

DrChandra (82180) | more than 3 years ago | (#36634594)

So, what they invented here is an extremely expensive way to cheat at Rock, Paper, Scissors?

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?