Interestingly, although the hippocampus plays a massive role in memory consolidation, memories are ultimately distributed throughout the cortex.

I'm curious whether this mechanism generalizes to all neurons or is specific to how the hippocampus can learn quickly, especially since the hippocampus is the one place where neurogenesis has been found in adults.

Interesting that this involves a response similar to an immune response to a pathogen. I've read a couple articles about alternate theories of Alzheimer's linking it to an increased immune response in the brain.

How does that work in visual memory, does it break and fix it too and that quickly? I have strong visual memory that I remember back in school I used to remember the page and it’s page number just by looking at it for few seconds, and if I saw a face for a second even randomly anywhere, I can recall when and where for a long period after.. I find it hard to imagine or rather scary all that is breaking/fixing the dna..

Fascinating that the apparatus for memory between generations may also be used for memory within a generation. "Breaking" and "fixing" DNA could also be taken as a description of meiosis or mitosis. Perhaps there is some code re-use there.

Although I have nothing substantial to contribute to the topic, I can't help but notice the beautiful mess of the neural field shown in the image; a reminder of the complexity of the real world and the challenges that still remain. Very far from our organized models arranged in layers of 'objects' and the didactic diagrams containing two neurons, or even convolutional network diagrams. Which brings to mind the good Professor “it must be made as simple as possible, but not simpler”.