Technology that steals memories

Question

There is a device in my story that »steals« memories from a person (I had some sort of transmitter in mind, but this can easily be modified, as long as it uses powerful long-range signals to achieve its function) and uploads it into its server, where it can be accessed for further use.

A person completely loses his memories when they are stolen. I have no idea how to explain the process of that happening, I was intending to qualify a memory simply as files like pictures and videos on a computer (this can be modified) all of which the device then somehow steals, but I need it to sound believable and at least to some extent consistent with basic laws of computer sciences and neuroscience since such a technology is probably mindboggingly far fetched.

I am no expert on both of these topics so I humbly turn to you for help.

Answer 1

We don't know how to digitize, copy, delete, edit, or record memories.

The technology to do so is so far advanced from what we currently understand as to be indistinguishable from magic. Any attempt to provide a science-based explanation for how machines are remotely stealing them will come across as awkward for anyone with even a minimal understanding of neurology.

What will probably be best for the believability of your story is to skip the explanation entirely and leave the inner workings of the dream stealing machine a mystery. If you really need an explanation, any cool sounding technobabble should suffice.

Answer 2

Make it so that each and every person on earth wears an implant that enhances brain functions: increase memory capacity, faster thought processing, and such.

That way, memory can be read and written, and stolen, by remotely accessing the device.

Artificial enhancements to organs, including brain, should be pretty common in the future.

Answer 3

As the answer by sphennings pointed out, there is a lot which is still unknown about the human brain. But a plausible theory is that long-term memories are stored in form of the configuration of synapses and neurons in the brain. If the process to scan them would have the side-effect of destroying (or at least reconfiguring) them, the result could plausibly be loss of long-term memory.

Answer 4

Memories are stored in the brain as neurons, which work by conducting electrical impulses. One could imagine that in order to steal a memory, you'd first force the neurons to fire, and then use your (immensely-sensitive) device to capture at long range the pattern of impulses that results.

Let's say the act of forcing the neurons to fire requires a powerful signal directed at the subject's brain. Perhaps the signal is so powerful it actually fries them in the process. Or perhaps the capture process requires the memory to be fired many, many times in quick succession in order to correctly capture the pattern, causing the neurons to degrade rapidly.

Answer 5

You can copy (or copy-then-delete to "steal") files from a computer by sending it a signal only because it has hardware to receive the signal, and software that interprets the signal as commands to send/delete things. The target computer cooperates in the process of stealing the file.

Even "hacking" to copy files that the computer isn't intended to give you works by figuring out that a defect in the software means it interprets certain signals as commands in unintended ways. It still basically depends on the target computer's cooperation, it's just that the computer cooperates in ways/cases that it wasn't intended to.

You cannot steal files from a computer just by beaming a signal at it if it doesn't have hardware for receiving that signal and sending responses (and almost certainly not if it isn't running software intended to listen to that hardware and send responses back). So I find it hard to imagine how you could simply beam a signal at a brain and have memories get transferred or deleted, if it's any kind of signal that science currently knows about.

The brain does "listen" to sensory inputs of course, and "send responses" in the form of speech and motor control. So you could posit that certain patterns of light and/or sound could exploit a defect in the human brain that cause it to delete memories, but it seems you'd need large quantities of handwavium to explain how you "target" memories, why the feature being exploited is universal in all human brains when it serves no evolutionary purpose, and especially how you actually record a memory (does your pattern make them go into a trance and write a detailed account, then wake up without the original memory?).

But, you could posit other technology changes that would allow the victim's brain to "cooperate" with the memory thief. What if in your world it's common to have neural implants that have access to memories?

The intended benefits of this technology could be improved (even perfect) recall, storing more information, making learning new skills (or at least rote memorisation) easier and faster. Perhaps it's even intended to allow memories to be transmitted, as the new form of sending postcards from holiday or posting your meals to social media. And certainly we can imagine them being networked, so that the device could receive and store messages and play them directly into your brain as desired.

Then all the inventors of your memory thieving device would need to do is find a flaw in the implants that causes the device to delete memories and transmit them, in response to specially crafted signals sent along the normal networking channel. More realistically they would need to find separate flaws in every specific model of the implant, but perhaps these are the smartphones of the future and one or two companies have captured a large fraction of the market at the time your story is set, so that the flaw could be in a part of the software that is common to all or most devices.

If the neural networking infrastructure works somewhat similarly to current mobile phones, then the devices would be intended to send signals to towers kilometres away, and you could probably boost that range significantly using an extra-sensitive receiver in your memory-hacker (to pick up the more attenuated return signal from further away). If the neural implants are intended to communicate long range by talking to satellites then you can probably hack from almost any distance provided you've got a moderately clear line of sight (going through buildings is probably fine, going through the Earth probably not). If the nature of your exploit is that the hacking signals are compatible with the "routing" used normally in the network you might not even need line of sight at all; you just inject your signal into the normal stream and the ordinary network will carry it to your target. But if you don't want it to work like that you could posit that the exploit requires a signal that isn't ordinary network packets (which might explain why the possibility was never identified in testing), so you have to beam your pirate signal directly at the victim because the network wouldn't forward it for you.

Answer 6

There are two ways I can currently think of :

• You mentioned a powerful long-range transmitter to steal someones memories. You could say that the radiation emitted from the transmitter would damage the brain of the victim while extracting the data. The brain damage results in destroyed memories (Memories are stored as neutrons and connection between them) . The only problem I see here is that scientists as for today think that memories are distributed redundantly, meaning that loosing some neurons doesn't result directly in loss of memories. Neurons die all the time and new ones are created. Although if you destroy all neurons associated with the memory you would loose it.

• Another way of explaining the memory loss is that the device does this on purpose. Like a computer can copy files and then delete the originals. The device would function like a virus stealing data and deleting it afterwords, depending on your story this might be a good choice.

Answer 7

Each person's brain develops slightly differently. Memory is stored not like files on a computer but as a giant associative net where each memory is connected to a related one. Because of this it is to today's knowledge not possible to extract memories or 'implant' them in someone else.

However what might become possible is that you can reconstruct a dream, a daydream, one's mind's eye or even someones thoughts by the activity in a brain. But this requires a labeled training set of brain activity (for each person) + what that person was thinking about at that time. For example here's a video, in which a video has been reconstructed from brain activity. And scientist have also been able to 'read' parts of mice's dreams after observing their brain patterns during the day: https://news.mit.edu/2002/dreams

Answer 8

We currently don't really know much about the workings of the brain on the level of memories. We can use that to our advantage by working backwards.

Let's say we take a large (but not mind-boggling) leap, and say we find a way to selectively destroy memories by scrambling neural connections, probably with a variation on TMS. Let's say researchers then measure the distortion in the magnetic field or something. They figure out it's not random. They then manage to decode the "deleted" memory from that signal. The researchers then realise their machine can be used to read memories from anyone, albeit destructively. What started out as a project to find a way to wipe memories (for whatever reason, nefarious or otherwise), suddenly turns in a way to actually steal memories (for nefarious reasons, let's be real). Those researchers will probably get very rich or very dead very quickly.

Or just skip over the explanation completely. It's fine if the audience doesn't know how the Magic Box of Memory Theft works, they just need to know that it works and it's because of Science, and they'll happily suspend disbelief.

Answer 9

There's talk of memory storage occurring in dna upon methyl groups being added to dna. Remove the methylation process and short term memory vanishes. Add the methyl group process back in and memory production recurs. In this way, even though proteins are used up in neural synapse firing, there is still something left behind that holds the info--the dna. https://www.google.com/amp/s/www.newscientist.com/article/mg20026845-000-memories-may-be-stored-on-your-dna/amp/

Your device will need to extract the dna, and chemically treat it to access the code (remove the methyl groups).

Most importantly, the device will then also need to somehow get the key to the code the person used to store the memory into the dna. That's the tricky part since without the key, the memory is just a bunch of gibberish. So, then the device must also trick the dna into believing it is within the code-generating organism so it will allow access to the key. How to do that?

Since the purpose of the memory storage is to hold information until needed at recall, the dna is alerted to transfer the coded info upon stimulation by...another dna molecule. So, keep some of the dna methylated, start the synapse firing sequence same pattern as observed within organism prior to the extraction, and the extracted demethylated dna will receive the keyed info and overlay it onto the coded memory. It cannot transfer the keyed info back however due to the missing methyl group, so it sits there now with the info decoded and available to be read by yet another part of the amazing memory extracting device.

Since all memories in the original organic brain must be destroyed in your story, upon completion of memory extraction, a huge bolus of a cocktail of every stress hormone ever in existence is released into the blood stream, and the memories are magically gone. Well, not gone, exactly. They will instead be replaced by one overpowering memory of the horror of the extraction process, since the stress hormone cocktail basically imprinted that current memory as the most important thing that has ever occurred to this brain and therefore it must be retained.

Answer 10

In the future, we might have completely decoded and understood the brain. I think we already can tell now, that the brain generally works like a (quantum-)computer. This means, if we do understand exactly how the brain is working and built up, we would be able to develop technologies, to manipulate the brain the same way, as we manipulate data on a hard drive. Meaning, we could read memories, manipulate them and delete them.

Answer 11

You will have to make some assumption of sorts about how memory is stored inside the human brain. At the moment, we are far from having a scientific agreement on the matter: some hypotesis involve synapses, certain memory-specific brain areas and even the single neurons themselves.

As a writer, though, you don't have to be 100% scientifically accurate. As many other suggested, you could just hint that the scientific community inside your world has a perfect understanding of the brain's inner workings and leave it at that - at least for what concerns memories.

If you'd like, you could choose one hypotesis of your liking and give few further explanations on that (e.g. : if memories are stored as the quantity of synapses between specific neurons, your machine would be able to decode and destroy those connections).

I would be more concerned about how the transmitter works. It seems reasonable enough to have an "antenna" inside our brain able to read its content (maybe it monitors signaling around neurons?) but how would it be able to erase anything?

I'd rule out bio-chemical ways of doing that - since they would be too imprecise. Maybe you can erase certain "memory-storage-units" (them being synapses, neurons, or whatever) with high voltage, precise electrical shocks - burning them, in practice - but you would need some sort of system able to specifically target only the items you want to destroy. Imho, the only good option here would be highly specific nanomachines. It's technobabble, allright, but it's also one of the sensible options.

Also, if I were you, I'd make the transmitter work at night, or whenever the subject sleep. It makes sense and reduces the chances of being discovered by the conscious mind of the subject - plus the erasing part may include some pain.

Answer 12

Memories and dreams get their substance on another level of reality, the very level where Universal Mind lies.

On that level of reality, memories can be handled just like the marbles depicted in Pixar's "Inside Out".

I've been told that bee-like animals exist on that level which are able to forage memories and dreams: they steal and process them to make their honey-like food. Unfortunately memories are destroyed in that process.

I've also been told that that process is the root cause of Alzeimer's disease.