Scientists Predicting Intentions

An anonymous reader writes to tell us German scientists claim to have the means of predicting decisions of high level mental activity. "In the past, experts had been able to detect decisions about making physical movements in advance. But researchers at Berlin's Bernstein Center for Computational Neuroscience claim they have now, for the first time, identified people's decisions about how they would later do a high-level mental activity _ in this case, adding versus subtracting."

Suspicion

[BWJones]

My first reaction is suspicion.... suspicion of a whole lot of possibilities regardless of whether or not this work has any validity. For instance, I've talked with more than one DOD general who was interested in military applications of electroencephalograms for "mind reading" and such. Certainly there are some applications for lie detection such as the P300, but one has to be very careful about the structure of the interview so as to not attempt to extract non-meaningful information from an evoked potential. My concern is that a whole bunch of additional DARPA type money will suddenly be thrown at the problem and claims will be made that will further impinge upon individual rights and freedoms waaaaaay before even the science is understood (not that understanding science is an excuse to stomp on civil liberties).

My more immediate concern is of the claims that are being made. The fundamental problem of course is developing a global signature for mind reading that is clean enough to derive robust statistics, keeping in mind that individuals brains are far from uniform in their anatomy, physiology or wiring. Work I performed more than a decade ago revealed similar cortical mapping patterns on subjects who performed tasks and then imagined performing those tasks. Certainly it is possible to determine volitional movements based upon our knowledge of neuroanatomy and statistical averages of wiring, but predicting "intentions" is a whole other ball game. The article is light on details and I've tried a search on more in-depth content, but if they are labeling "intentions" as complex behaviors, my eyebrows will be raised. For instance, determining which of two buttons to press invokes a whole series of kinesthetic volitional programming that should be able to be determined by mapping pre-motor cortex. However, if "intentions" are whether or not to engage in complex behaviors are what they are talking about, there is much more complex circuitry to consider including the possibility of imagery or imagining an action versus actually volitionally engaging in that activity.

devil's advocate

[User 956]

researchers at Berlin's Bernstein Center for Computational Neuroscience claim they have now, for the first time, identified people's decisions about how they would later do a high-level mental activity _ in this case, adding versus subtracting."

A big portion of the work of prosecution in this country is spent proving intent. For example, the funny-looking guy that hangs out at the playground. Is he a creep, or is he just a birdwatcher? Obviously, a scanning device would figure that out pretty quick.

(... And I guarantee you that's the same kind of argument they'll make when pushing this thing, too. Because it's all about protecting the children. even at the expense of your fourth amendment rights.)

Re:devil's advocate

[Ed Avis]

You can prove intent, but intent is not enough to get a conviction: you need the act to have been committed or attempted too. There is no crime of having intent to rob, but there is one of robbery (theft) or of entering a house with intent to rob (burglary). If people start being prosecuted for mere intentions, then you need to fix the law, not worry about mind-reading devices (which after all are just the messenger).

neat but...

[symes]

Then they studied which type of patterns were associated with different intentions.

"If you knew which thought signatures to look for, you could theoretically predict in more detail what people were going to do in the future," said Haynes.

Which isn't a million miles from... "we observed that just before our participant scratched their nose they raised thier hand". Using this observation we were able to predict when participants were about to scratch thier nose. And did so with an accuracy rate of 70%."

Don't get me wrong - I think this research is very interesting - but a little over egged at this moment in time.

Re:devil's advocate

[exp(pi*sqrt(163))]

A big portion of the work of prosecution in this country is spent proving intent.

And I've always complained about it. People make a lot of noise about freedom of speech but we don't even have freedom of thought. If you unlawfully kill someone while intending to do it you get a longer sentence than if you didn't intend it. Punish someone for killing, but to punish them additionally because of what they were thinking at the time seems like the grossest kind of human rights abuse to me.