Slashdot Mirror
Paycheck-Style Memory Erasure: How Close Are We?
Quirk writes "Scientific American takes a look at the movie Paycheck, based on Philip K. Dick's work of the same name. In the movie ...'a crack reverse engineer helps companies steal and improve upon the technology of their rivals, then has his memory of the time he spent working for them erased.' '...the main character gets several months' worth of his memories erased by having individual neurons zapped. Is that possible?'"
The brain is one of the least understood organs in our bodies, and tampering with it in any form is still quite tricky and dangerous. Sure, we might have a rough idea where your memory of your first day of school is, but erasing that and nothing else isn't something we're even close to. I'd say this is still at least 50 years away, and probably more like 100.
No. You don't just form/strengthen one new connection for every memory. If we knew enough to erase memories, we would know enough to back them up too.
"I would say that 99 per cent of what my father has written about his own life is false." - L. Ron Hubbard Jr.
If a company hires someone to steal technology, if it's done carefully (i.e. no email records, no obvious plagiarism), the only way to prove it would be to crak open the guy's skull and download his memories. Since it's not possible, why would there be a need to erase the person's memory in the first place? As far as I know, the best proof it's possible is Microsoft: nobody there has been forcedly lobotomized, and the strong company culture ensures that employees think technology theft as survival of the fittest, fair game, corporate smartness or other brutal but honest reasons that won't conflict with employees' sense of morality.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
This technology is here today. People conveniently forget about atrocities when the price is right.
Witness the mainstream press forgetting that Donald Rumsfeld, an official of the US govt, shook hands with Saddam during the 80s, or helped sell nuclear reactors to the North Koreans during his stint as a ABB executive. Or that George Bush II's father was VP when Saddam gassed Kurds.
Money is a good neural solvent. Sniff it up, America.
-Alex
Actually most of the evidence points to (long-term) memory being "holographic" (no lasers, mathematically so) - it is spread out. Damaging a neuron or 1000 decreases the intensity of memories, but the memory as a whole is only very vaguely localised, like the way you can cut holes in holograms, and have the bit be "filled in" weakly by the surrounding undamaged hologram.
Relax- this isn't the Crying Game.
Premise != plot twist.
The premise of the movie is no secret... how else do you expect to get audiences to go see it?
That's ok, Jesus likes me anyway.
Hate to be a nitpicker, but buying a company's product, taking it apart, and learning how it works is not stealing. It doesn't matter if you're the company's competitor, it still isn't stealing. You have a perfect right to do this, and employ the knowledge gained to your own advantage.
Now, if the technologies in the product are patented, and you built and sold your own products based on them, then you'd have a case of patent infringement. Which still isn't stealing.
Schwab
Editor, A1-AAA AmeriCaptions
There is a lot of controversy surrounding this.
For certain types of mental illness, electro-convulsive therapy is still considered an acceptable form of treatment by some physicians. I think the voltage has been lowered a bit and the duration, frequency and method of zapping is more tightly controlled, but it is still used quite regularly and has been since at least the 60s- maybe earlier.
One of the side effects of this treatment is a temporary loss of short-term memory. Suuposedly, it eventually returns, but patients lose short-term memory of events leading up to the treatment.
Having seen this sort of thing first-hand, i find it disturbing that anyone could support it. The brain, for the most part, is uncharted territory; and the fact that, without really knowing anything about it, we are willing to pump juice through someones brain because it 'seems to help' is insane to me.
To me, the concept is similar to patching a for loop that isnt working right by screwing with the counter in the test. It may get things working- but it also has the potential to break a lot of other things. It's the wrong way to go about doing things.
That's ok, Jesus likes me anyway.
... except maybe this. Computer neural networks are modelled on how we think the human brain works and so the following possibly applies to the human brain too.
Say you have a computer simulated neural network consisting of 10 neurons, and it can classify 20 different inputs into one of 3 different outputs. The network as a whole 'knows' how to do the classification, the combination of all neurons is responsible for the outcome. In order to adjust it so that it mis-classified one of the 20 inputs, you would most likely have to adjust the weighting (connection) of each neuron, or at least several.
Have you ever done a Rubix(sp?) Cube? Cheating aside, it's quite tricky to move only selected pieces around without mucking up the rest. Each single action you perform affects multiple pieces. You need to make numerous single gross movements to have a net fine movement. Tinkering with the human brain is probably a lot like that only much much trickier. And without the pretty colors. And you can't pull it apart and put it back together, or just move the labels around to do what you want. And if you tried to manipulate a brain like you do a cube you'd probably get your hands a lot dirtier. Okay... maybe it wasn't such a good analogy.
IANABD (Brain Doctor), but remember, the connections between neurons in the brain aren't electronic like you might think of computer memory as electronic. The interaction between them is, partly, but the actual physical connection isn't and as I understand it, the connection configuration is where the 'information' is stored. In order to get in and physically change connections you'd have to be tinkering with the actual neuron cells, requiring physical interaction which would be really hard for anything not on the surface.
I guess that leaves us with drugs, brainwashing, or tiny little robots, or something we haven't thought of yet. Far simpler to simply pay someone lots of money to pretend they've forgotten the thing you wanted to erase.
*snort*
I'm not a "low grade sci-fi writer", I'm so low class I'd have to improve to make no-class... and I hope I made that *perfectly* clear.
The point, mister anonymous, is that while the technology in "Paycheck" is vanishingly unlikely... the idea that we'd be able to untangle the changes in brain structure that represent specific memories and *reverse* them without changing anything else... well... it'd be easier to fix all the security holes in Windows armed with nothing but a bar magnet and a really good magnifying glass.
But I suspect we're not far from being able to induce things like temporary anterograde amnesia. If you could actually do useful work in that state it would make a heck of a security protocol. For some skills that would be enough: sightreading and playing a score may be possible, if a mob boss in hiding wants live entertainment. For others, well, you'd need to be able to replace long term memory with something external to the brain. How far off is that? I don't know, but I'll bet it's closer than "memory erasure".
Yes, I'm sure it's likely the information is stored in a distributed manner. The terms holographic and fractal are certainly very relevant.
Since large portions of the cerebrum can be surgically removed without a patient losing memory, certainly zapping individual neurons will be even less feasible for the erasure of memory.
In an artificial neural network, even simple information is typically stored as patterns of connection weights between many neurons.
Just where in that quote did either the doctor or interviewer imply that it was good for you? If anything, I would have interpreted And that's rather gross stimulation as implying the opposite.
a.
You've hit upon the fundamental difference between scientists and doctors here - which, incidentally, is why most people of either profession refuse to take the other seriously. Scientists think the way you do: if something's going wrong and you don't understand what's happening, then figure it out before you do anything that could screw things up even worse. Doctors think in a different way: do whatever is necessary, and whatever you can, to keep the patient alive and as healthy as possible; it doesn't matter if you understand how the treatment works or not (for example, we have no idea how most drugs have their effect, which is a large part of the reason why drug development is so expensive and time-consuming and requires clinical trials). The difference stems from the fact that scientists want to understand (or at least predict) the behavior of the universe, whereas doctors want to keep people alive.
To bring this back to the discussion at hand, there are two competing theories of how our minds work. In the first, we have specific cells devoted to specific memories - e.g., you have a "grandmother cell" that remembers your grandmother, and if that cell were to die, you'd lose the memory. In the second, our brain is a state machine, so the memory of you grandmother is spread throughout the activity of the entire brain. There's evidence to support both ideas, which suggests that the truth is somewhere in the middle. From the standpoint of believable movie science, do we understand enough about the brain to be able to erase someone's memory precisely, accurately, and repeatably, knowing exactly what we're doing? No. That's the scientist's point of view. Do we have enough tools at our command to be able to erase part of someone's memory if it were really, really important and we had plenty of time and money to spend on the problem? Maybe. That's the doctor's point of view (not that a doctor would do this necessarily, but it illustrates the solve-a-practical-problem vs. understand-the-fundamental-principles mentality that separates the two cultures).
(and, once again, five mod points go unused.)
On stereophonic equipment, the monaural sound obtained through multiple channels will enhance your listening pleasure.
I don't believe anyone in neurobiology believes in the grandmother cell. It's still used to describe how memory might work, but everything we know about the brain indicates distributed storage.
There are cells dedicated to specific purposes more general than grandmother recognition. These functional areas are dedicated to things like speaking or understanding speech (seperate areas of the brain.) For another example, everything you see is pretty much projected onto the neurons on the surface your occipital lobe.
A person with brain injury can lose specific skills or abilities. My grandmother lost the ability to speak after a stroke. She relearned to speak.
They can lose types of memory. People with Korsakoff's syndrome live with no intermediate or long term memory. Loss of short term memory preceding a traumatic event is more common. After an accident it is common for the injured party to not remember the moments leading up to the accident, because that information essentially never got written to intermediate or long term memory.
But the current view is that memory is highly distributed. If you use a neural net as a trivial model of how the brain might work, you will realize that for a large and complex neural net with diverse purposes, there isn't a single cell devoted to anything. All the information is contained in the strength connections between cells.
Karl Pribram used the phrase "holographic brain." The image on a hologram is distributed, so if you break it in two, you have two complete images, although each is less detailed. If you scratch a hologram, you don't lose part of the image, you lose detail overall.
There are drugs that prevent short term memory from being retained. Those drugs also keep you from being very alert or useful for anything, and the only people who use them to that purpose are rapists.
So, to answer the poster's question: No way.
Crude manipulation of the mind is hard. Hypnosis can't make you do something you'd be unwilling to do otherwise. Truth serums ain't. Lie detectors don't. I'd suggest that truth serums & lie detectors are far simpler tasks than erasing human memory based on content.
The brain is just too vast & complex for such a trivial approach. You need to use something subtle and powerful to manipulate the mind, like advertising or religion.
Assembly is the reverse of disassembly.
He'd be obsolete after his first job. He'd be the perpetual low-paid intern, fresh out of college, for his entire career.
PHEM - party like it's 1997-2003!
Really nice writing. Just one thing that strikes me as flawed.
(2) Even without understanding something, we can manipulate it in such a way as to achieve the desired affect.
The problem is that neither me, you nor anybody else have so far been able to reliably predict the effect of treatment in a specific patient. Sure there are statistics, but that does not help predicting reactions in each specific case. Maybe you should say
(2) Even without understanding something, we can manipulate it and hope we achieve the desired affect.
I agree that medicine is not science, but an art. The problem is that most doctors view themselves as scientists.
Infinite minus one.
The brain just doesn't work that way. The "storing" of information in the brain is an emergent property that comes from a vast array of correlated neuron activations.
Think of a massive Hebb-rule neural net.
So how do we "forget" things?
Well, as in a neural net, old patterns, if not reinforced, are slowly lost as the "writing" of new patterns degrades them. But every time we think about something, it reinforces that pattern.
So how can these patterns be directly and quickly erased?
They can't. You'd have to be able to identify all the minute changes in neuron co-activation that represented *this* particular memeory and no other, and directly (physically) reverse them.
This would require a technology so advanced that it would be indistinguishable from magic.
NOTICE: This notice will appear at the bottom of all my slashdot posts.