We know what you’re thinking: Scientists find a way to read minds

Now isn't this just something to grab your attention! The title describes the story in a case where I really can't call the headline sensationalized, despite how it first looks.

They really do appear to be on the path of the Holy Grail of many different fields of effort. The ability to SEE what the brain sees, and as it sees it in relative comparison to others as controls to refine the technology.

In the study, scientists hooked participants up to an fMRI brain scanner – which determines activity in different parts of the brain by measuring blood flow – and showed them images of faces. Then, using only the brain scans, the scientists were able to create images of the faces the people were looking at.

“It is mind reading,” said Alan S. Cowen, a graduate student at the University of California Berkeley who co-authored the study with professor Marvin M. Chun from Yale and Brice A. Kuhl from New York University.

Source

It is what it is, and that is what it looks like. No pun intended. The article has images of original pictures compared to overlays of different scans that, collectively, rebuild the original photo or something quite close. It's amazing.

In it's current form, it's nothing to necessarily halt the presses over, but this is brand new and in a way, one could say they are just learning what calibration means for getting it working. Crime and court applications or other things down that path are real benefits and perhaps both ways for how it goes. However, something like this would also apply strongly to the disabled in both physical sight issues and mental perception issues like nothing else ever could. As a starting point, it's quite a thing to watch for developments with!

Hmmm, but can we trust what people see? For example those with eating dissorders observe themselves to be grossly overweight, even though they are just skin and bone.

ETA: I mysellf was in the kitchen one day washing the dishes. Children outside were running about playing. I watched quite a large dog running about with them. about 30 seconds later I realised it was MY dog (he'd escaped), but the point is, I'd always seen him as a much smaller dog, it was only when I didn't know it was my dog that I saw him as being a big dog, so can we really trust these brain scans?

Oh!..............What a wonderful interrogation tool!
Surely DARPA must be funding the research eh?

Pretty cool.

People were shown various faces superimposed on synthetic faces and their MRI patterns were observed. When shown other faces the patterns were compared to the "training face" patterns and a computer program produced the reconstruction. More of a composite, really.

www.sciencedirect.com...


The thing is, without the computer being "trained" for an individual observer, it wouldn't work.

“There’s a wide variation in how people’s brains work under a scanner – some people have better brains for fMRI – and so if you were to pick a participant at random it might be that their reconstructions are really good, or it might be that their reconstructions are really poor, which is why we averaged across all the participants,” Cowen said.

www.foxnews.com...

It won't do them any good if they try it on a woman, it's impossible to know what a woman is thinking.

reply to post by stirling


I'm not sure what they are talking about here could be used that way. Not so much for interrogation. It takes clear and natural brain patterns to read this and someone fighting could really screw everything up. Imagine just shifting images in your mind through things ranging from Old Faithful to Saturn's Rings, to a Whale, to a pretty red rose.

(imagines) Oh...that would get A lot. DHS would be chasing down a new biological space monster hiding in the ocean and colored kinda red, with the whole US Navy headed out. (/imagines)

What I can see it being useful for first is a true and useful sense of what people with mental or physical issues see. There are countless brain injuries that produce regular, really bizarre but relatively benign visual problems. (one of my instructors on a study course is a Neuro-Psych by profession and talked a lot about that).

If they get a thousand people to look at a dozen very structured patterns for a baseline, I can see where computing power today could really draw useful data for showing what a disabled or brain injured person has looking at those same patterns. it's not a flashy or sexy use of technology, but for what this sounds capable of in time? It'd be far more life changing for people who need it.

reply to post by Wrabbit2000


The Superbowl with the black eyes peas featured was exactly the baseline test being performed on a scaled up application.

This is cool, but is it just me that think it is scary that we might potentially be on the verge of loosing our last privacy?