I'm not sure what got me started thinking about it--perhaps listening to Skeptics Guide podcasts--but lately I've been thinking a lot about how our brains store memory. I think the two biggest questions I have in life are what is the universe about and how do we think and remember. I doubt the first will be answered in my lifetime but the second likely will. 

Wikipedia is a wonderful resource for following the chain of topics to learn something new. I start with memory and move on to long-term memory. I learn the role of the hippocampus and about how long-term potentiation chemically creates memories in our brain. But I'm still left with the big question of how do we perceive. It's one thing to have stored pathways in the brain; it's another to be able to make sense of them. I'll have to keep thinking that one over.

But regarding memory, after much pondering, I wonder if it's all some sort of memory map, to borrow a concept from computer architecture. What if each pathway corresponds to a combination of all possible values of all sensory input? We know each sensory processing area of the brain projects the information from its input (an image, smell, sound, etc.) onto a map of corresponding nerve cells. I speculate that each of those areas then converts this combination into a chemical code and sends that to the hippocampus which in turn combines inputs into a combined code. That code is then projected into the memory storage area. The parts of the code that have been encountered and stored previously follow the same path but when a difference is found, a new path is etched. This would be an efficient system for retrieving memories based on related previous experiences. For example, hearing a certain sound often retrieves a set of related memories because the memories actually share a significant part of the pathway. Then, during thought, following a pathway mentally recreates the map of sensory input that created it in the first place by recreating its unique "code". Of course this model infers that the brain contains enough possible pathway capacity to encode everything we encounter in our lifetimes.

Something else occurred to me as I was writing above. When the sensory processing areas communicate with the hippocampus, it would be most efficient for them to only communicate changes (as most computer compression algorithms do). This would allow the sensory processors to shield the rest of the brain from a lot of "noise" when the environment is relatively stable. So, for example, sitting in a chair with eyes closed listening to music, after sensory processors have sent the initial message for touch, sight, smell, etc., all sensory input except hearing will remain relatively stable and only those sound changes need to be continuously transmitted. I wonder if that's how it works.

Comments (0)

Subscribe to this comment's feed

Show/hide comments

Show/hide comment form

Name

Email

Website

Title

Comment

smaller | bigger

Subscribe via email (registered users only)

I have read and agree to the Terms of Usage.

Add Comment