This page describes some potential future technologies that could be used to retrieve or reconstruct memories. It's organized with the easier technologies toward the top, and time spans are given in an effort to compare complexities. But, of course, any predictions are guaranteed to be spectacularly wrong. The technologies below won't happen as described, but similar technologies might develop, and they could do so without violating any laws of physics.
Stem cell researchers are currently making slow progress in regrowing new tissues and organs. Scientists could eventually figure out how to routinely guide cells with high precision by using physical manipulation, chemical signaling, scaffolding, nutrient supply, stimulation, reprogramming, etc. Mature tissue engineering technology would be capable of replacing organs, limbs, bones, skin, teeth, etc. A society that had mature tissue engineering would also be capable of building artificial wombs, growing meat for food in factories, and many other fantastic technologies. Most diseases could be cured by the replacement of the malfunctioning organs. Cancer could be curable in most cases by removing the cancerous tissue with wide margins and replacing it with newly grown tissue. Diseased brain tissue could be replaceable to some degree, although it would result in some memory loss and having to relearn certain skills. Incremental replacement of brain tissue might mitigate some of these issues, and would still be preferable to dying. Tissue engineering is easily envisioned and eagerly anticipated by many scientists and the public.
The application for cryonics is that this technology could grow an entire new body around an existing repaired brain. It might also be able to replace damaged sections of the brain, although without the original memories intact.
A brain could be scanned with electron microscopy, layer by layer. Each scanned layer would be removed to expose the next layer. At some point, it could be possible to do this on certain cryopreserved brains with enough resolution to capture all of the neural pathways and synapses. This scanning would largely be a structural scan, yielding little chemical information, but that might be adequate as long as memory is not encoded in smaller structures than is currently believed.
The application for cryonics is that this could facilitate mind uploading, described below.
aka Whole Brain Emulation (WBE)
aka Mind Emulation
Once the neural pathways have been scanned and mapped, powerful computer hardware and software could emulate the mind. The mind could be woken up and a virtual body could be used to provide input and output. The main objections to mind uploading tend to be philosophical, with some people feeling that this would be a cheap copy and not really "them". This leads to long debates about "identity" and "self". This angst exists simply because it's hard to envision software so complex; we have no daily experience with that sort of technology. But this is just a form of vitalism, a belief that there is something fundamentally special about the brain that could not be replicated by some other computational machine. If accurate emulation technology is ever developed, the philosophical objections will quickly wither in favor of pragmatism.
Uploading the 302 neuron mind of a worm has been an ongoing project of scientists for the last 40 years. See openworm.org. It would still take many more decades of additional work before they might actually succeed. The virtual environment for the upload must be richly detailed and accurate, the entire body must be accurately emulated, and scientists must demonstrate in blind studies that learned behavior survives the upload process. After a worm, the next upload target might be the 135,000 neuron mind of a fruit fly. This is certainly many more decades away, but if they eventually succeed, uploading larger animals would just be a matter of scaling up, and steady progress would be likely. Early uploads would think slower than real-time and would require large and expensive hardware. With engineering advances, the speed could improve and the cost and size could drop.
This technology should be much easier to develop than any of the physical repair technologies listed below. It is, therefore, the most plausible revival scenario.
Scanning closer to molecular resolution could be accomplished by a number of different methods. In one scenario, electron microscopy could be supplemented with chemical analysis of each layer. In another scenario, highly parallel arrays of probes on articulated arms could feel their way across each layer of the brain prior to milling. In a third scenario, grippers could remove individual molecules from the surface, disassembling the brain layer by layer in a highly controlled manner. Any of these proposed methods would allow a superior scan compared to the connectome scan. Revival would still depend on mind uploading.
The brain could be scanned by a method described above. It could be "repaired" in software, and then a new brain could be rebuilt from scratch using new molecules. This seems absurdly complex to us because it would involve the controlled movement of about 10,000,000,000,000,000,000 molecules, but it's just an engineering problem that's well within the limits of physics, and it's very plausible that this could eventually be accomplished. This is the lowest form of technology that would not require mind uploading.
Again, each layer could be scanned. But instead of destroying the layer, manipulating arms could grab the atoms and move them over to another site where the brain was being rebuilt. In this fashion, the repairs could be made by actually fixing the original molecules.
But... there would be no scientific advantage to this, as compared to the simpler Molecular Scan and Rebuild. The only advantage would be to appease those with philosophical issues who insist on the reuse of the original molecules. This scenario is here solely to point out that none of the less complex technologies would be capable of retaining the original molecules that made up the brain. In other words, it's very likely that those doing the "repairs" would not go to the extra effort and expense of reusing the original molecules even though they could if they waited a few decades for higher technology.
A frequently described cryonics repair scenario is to use nanorobots that swim through the body and make the repairs with manipulating arms. While devices such as this may certainly be part of future medicine, it's very unlikely that they would ever be used to repair a cryonics patient. The manipulating arms and working tips on such a device would be massive compared to the molecules they were trying to repair. The tips would be operating by feel, requiring a tremendous amount of manipulation and tunneling just to characterize the damage in the first place, let alone make the delicate repairs. These basic physical limitations might eventually be overcome to some degree by very complex engineering, but this engineering would take decades. In the meantime, the scan and rebuild technology described above could be getting faster and more sophisticated. Scan and rebuild would have the huge advantage of operating on a flat surface. The manipulator arms could be arbitrarily large and could make use of a much lower level of technology. In the end, there would simply be no need for nanorobotic swimmers in cryonics repair. Even if they were eventually feasible -- a very big if -- their cost and performance could never be competitive with scan and rebuild technology.
Repair and memory reconstruction could take place by a number of different technologies without violating any laws of physics. Regrowing a new body would be by far the easiest of any of the technologies listed.
Mainstream scientists will usually, after some consideration, agree with every single statement on this page as being reasonable. This is especially true if they are reminded that there is no time limit for developing these technologies. At the same time, they will discount cryonics for a number of reasons unrelated to its technical feasibility. A common irrational reason given is that these technologies would be so far in the future as to not be relevant.