View Full Version : Memories, neuroscience, and neuroimaging

12-14-2013, 11:34 PM
In my WIP, I have a lab that does memory removal and transfer procedures. Obviously, I'm taking a ton of liberties with the science, since it's not currently possible, but I just want to make sure it's not totally off the wall.

Here's how it currently works:

1. The subject is put a state of hypnosis and brought back to the memory that's being removed.

2. The scientist maps the specific neuron network that is active during that memory. Right now I have them using EEG, which is probably not specific enough. I may change this to MRI, but that involves a much bigger machine/operation and the lab is pretty small and clandestine.

3. The scientist inactivates those specific neurons, making it inaccessible to the subject's conscious memory. I'd love some ideas on how this could feasibly work.

For transferring memories, a similar process is done with hypnosis to bring up the memory being transferred. The lab maps the memory and then implants it into somebody else's head. I wanted this to be as non-invasive as possible, so they're using transcranial magnetic stimulation. Again, a huge leap in terms of what the technology can currently do. I know those MIT people who implanted a false memory in mice used a laser, but I'm pretty sure they also had to open up the mouse's skull and shoot it directly inside, which is not an option for the outpatient procedure in my book.

Thoughts on this process? I'm sure it'll royally piss of anybody who actually knows about neuroscience, but I'm hoping to make it mostly believable to a general YA audience.

12-15-2013, 12:41 AM
Depending on the EEG it can be fairly specific (http://science.howstuffworks.com/science-vs-myth/extrasensory-perceptions/hypnosis4.htm). EEG always struck me as a monitoring system, while MRI is more like an x-ray. Have you looked up different EEG machines online? There are some with explanations on eBay last I checked, you could also look up videos about like Ken Wilber (http://www.youtube.com/watch?v=LFFMtq5g8N4) and other people that use machines to monitor and track their consciousness/neurons.

Towards inactivating neurons using EEG, MRI, fMRi... I think the most plausible way to do that (though I really have no idea) is to destroy either the neural pathway and the brain cells, themselves, through a burning/laser method.

Implanting a false memory is a lot different than finding a specific memory and extracting it... I don't know if that'd even be possible because current neuroscience has only vaguely linked physical manifestation of memory storage in numerous locations, but regardless of where they might be stored, recalling them is a brain-wide process distributed through-out using synapses (http://en.wikipedia.org/wiki/Synapse). Not only that, but memories are differentiated (http://curiosity.discovery.com/question/where-brain-store-our-memories) by practicality such as procedural memories, semantic memories, working memory, etc. causing a variety of different places where they are potentially encoded, if not causing a potential of overlapping storage and a complex web of recall.

So, for specific removal of a single memory would be near-impossible using current neuroscience technology as it would be highly likely the removal of a storage facility would remove /all/ memories involved there, and removal of synapses would affect not just the specific memory, but a number of memories that might use that that pathway during the complex recall (http://www.scientificamerican.com/article.cfm?id=the-memory-trace) process.

Right now, the best bet for memory damage is lesions on the hippocampus or frontal cortex, but again, this doesn't gain specific control over memories, but rather losing the ability to form new memories and mostly for short-term, however... plus such findings are only from a minor study with an extremely small sample size of males.

My personal guess is that memories interact with the pineal gland, ahurr hurr, since the pineal gland and the hippocampus are linked (http://thebrainlabs.com/Images/hippocampus.gif) together within the brain.

I think perhaps looking at studies involving neuroscience and neurodegeneration (http://en.wikipedia.org/wiki/Neurodegenerative_disease) could lead you down a potential path of advancing the current technology into something that you could use for your purposes.

Also, I think it's a lot more likely that you could suppress a specific memory using an advanced form of hypnosis which has very little to do with neuroscience and everything to do with unlocking the hypno-ee's subconscious regardless (http://www.webmd.com/brain/news/20050627/what-hypnosis-does-to-brain) of where it might be linked in the brain since human beings have a lot of power in our Understanding (i.e. placebo effect). So yeah, if you could figure out a heightened, technological-based hypnosis (http://www.scientificamerican.com/article.cfm?id=hypnosis-memory-brain) (bordering on the mind control programs of 40s-German, 50-60s-USA), it'd probably be the most plausible route.

King Neptune
12-15-2013, 02:42 AM
NQuinn has good points.

It is my opinion that the alleged "false memories" implanted in mice were actually a type of learning, but it might be that the article was so poorly written that it didn't say what had actually been done.

It hasn't been used in higher animals, but the transference of memory RNA that was done in some nematodes (I think) is another way to go, but the lack of specific locations for specific memories makes that difficult, but you are writing fiction, so you can slide past that.

12-15-2013, 06:57 PM
Neither MRI nor EEG is specific enough to map out individual groups of neurons that may be holding a memory. An EEG signal is a summation of very large groups of neurons, and source localization in EEG is notoriously unreliable.

MRI is used only for imaging the brain. If you want to look at brain function, you use fMRI, which measures blood flow to certain areas. This has good spatial specificity, but nowhere near the scale you'd need to observe specific neurons/synaptic junctions that may be responsible for memory.

I think you're better off not using those techniques, and making up a new one, or just being very vague about specifics.

Other things to consider: Intracortical recordings are possible in humans (it's very invasive, though) and recently there has been research into Transcranial Magnetic Stimulation (TMS), which temporarily inactivates areas of the brain. A combination of these techniques (advanced up for sci fi, of course) might be better.

12-15-2013, 10:31 PM
I think you're better off not using those techniques, and making up a new one, or just being very vague about specifics.

I think I'll end up going this route, probably with some kind of "machine" you stick your head inside. And I like the idea of using TMS to inactivate brain parts. That way I've got somewhat technical lingo in there.

Thank you, everyone!

12-15-2013, 11:11 PM
Have you heard of this experiment?


It seems like it would be relevant to your idea, and the video is jaw-dropping, IMO. It makes science-fiction tinkering-with-your-brain stuff seem a lot closer to developing in real life.

12-15-2013, 11:18 PM
Ooooooh, I hadn't seen that, and it's perfect. Thank you so much.

12-16-2013, 12:23 AM
So I think what I'm going to do is have the scientist explaining the procedure to the MC -- who is 17 -- in an oversimplified, patronizing way. I'll have them using "a technique similar to functional MRI" to record the parts of her brain that activate as she re-experiences the memory, hypnotic suggestion to block access to memories, and a form of transcranial magnetic stimulation to add new memories. Still probably quite cringe-inducing to a neuroscientist, but hopefully it'll work for my purposes.

Russell Secord
12-16-2013, 04:51 AM
I am not a neuropsychologist, but I seem to remember from TLC (when it was The Learning Channel and not reality hell) that memory is more holographic than anything else. In other words, you can't go to a specific location and erase a specific memory the way you can zero out an address in a computer memory chip. One type of memory is encoded in one type of neuron, but one memory doesn't necessarily go into one localized area of the brain--more likely, it's spread throughout a network that holds similar memories.

For the purposes of your story, I would have the subject's brain read into a buffer, perhaps another brain grown for the purpose, that is effectively blank. Any memories that need to come out are copied into another brain. The buffer is altered to reach the desired result and copied over the original--in fact, the buffer could be physically transplanted into the original subject, but that seems a little sledgehammery. Potentials for skullduggery (pun intended) are left as an exercise.

12-16-2013, 10:59 PM
False memories can be implanted quite easily through hypnosis and suggestion.