In other words, the future depicted in this weirdly charming and underrated anime: http://myanimelist.net/anime/3701/Kaiba
Good stuff, just be wary of spoilers.
I say that because there does not seem to be a non-interactive way to recall memories. Every time a person experiences a memory (remembers), they also alter that memory. Researchers have been able to fairly easily alter or implant memories simply by suggesting the changes to subjects while they are in the process of remembering.
It sort of makes sense--the moment of remembering is itself a memorable moment, and so the brain updates or replaces the original memory with what was recalled the most recent time. If the recall was messy, the old memory will now be messy too.
Just curious, in such experiments, how does the 'blue light' reach interior cells? Is it simply of such intensity, from an external source, that it penetrates deeply through tissue into the mice skulls? (A bit like a flashlight-through-a-hand?)
Or, is the light source somehow inserted to be close to the cells (even under the skull)?
Interesting; do you have any links to those papers with more information?
Like recovering data from a broken filesystem - the directory structure is lost and for a normal usecase the data might as well be gone, but a lot of the content is theoretically there and recoverable if some indexing system (even one different from the original) can be created.
That said, it is a great accomplishment for the field. Less so for the prospect of 'memory-restoring' technology in the near future, but more for what it says about how memory works. Essentially, this paper (and our work hopefully coming soon), proves that retrieval cues are the most superficial aspect of the brain's systems for memory. Thus, they fall first to the ravages of Alzheimer's and when they're gone, the pattern of information is still there, but the brain is unable to recall that specific memory when input into the network is noisy. There's been some other papers recently showing the persistence of memories even when synapses and spines are lost (for example, during hibernation). It's as if a record of learning is also stored somewhere in the cell's genes, but no one seems to really know what to make of it.
Maybe that stuff will also be of interest to neural network people...
All that said, there's now a large-scale study getting started in Canada that will be stimulating the brains of MCI and AD patients with tDCS to slow down Alzheimer's (pACT-MD). Maybe they'll be able to recover some those lost memories!