X-Message-Number: 5186 From: (Thomas Donaldson) Subject: Re: CryoNet #5174 - #5181 Date: Wed, 15 Nov 1995 12:51:15 -0800 (PST) Hi again: Mr Lynch makes several statements about how our brains work, some of which need amplification and others are leaning a bit in the wrong direction. First, LTP seems to be a stage in the formation of long term memory. In that sense, it is important, but the REAL long term memory persists for other reasons entirely. What apparently happens is that after a time (one thing LTP does is to hold a memory while that time passes) relevant synapses can increase in number, form dendritic spines where they formerly did not, and new dendrites can be grown, with synapses, to make more contacts between the two communicating neurons. These changes take place on a scale much larger than that of molecules. Just what the effect of releasing NO here may be remains unknown. A lot more experiments need to be done to find out (I am referring to his reference to the SCIENCE paper in which this was found to happen, at least for one part of our brain). Since we are dealing with LTP rather than the changes that lead to long term memory, just what the effect will be on long term memory needs to be examined specifically. (Yes, one reason there is so much attention to LTP is that these other changes are much harder to follow). Although I feel sure that one way or another we will learn someday how to make neural nets which form new connections rather than simply strengthen old ones, that isn't presently the case. If we have such a neural net (which is our brain) it becomes impossible to work out memory capacity simply by counting the number of synapses. Given the fact that our brains are enclosed by bone, and that limits the size to which they would grow, there must be an upper limit ... but even so, any calculation would have to include the fact that brain connectivity can change slowly over time, and even increase. Furthermore, the problem very much does NOT stop just with the information we need to reconstruct a HEALTHY brain. One reason we will want to be able to at least identify the locations of critical molecules is that for cryonics we will ultimately (hopefully not for everyone, but always for some) have to put together a DAMAGED brain. The distribution of various brain chemicals gives us important clues as to what is connected to what. We are just beginning to get some handle on what chemicals these may be. AT least 2 classes, together with the different neurotransmitters and their receptors (most neurotransmitters have more than one chemical receptor) may provide such clues. (The classes are neurexins and connexins). It's very important to have and use such clues --- otherwise the problem of reconstruction becomes virtually impossible. (If you tried to put together a puzzle with pieces which were identical in every way that you could see, you'd have a severe combinatorial problem. Each piece might even have a label, and somewhere inaccessible to you the correct arrangement is stored, but that would not help you at all). I will finally say that while analyzing a brain on a molecular level will prove helpful, analyzing one on an atomic level will NOT. It is important to know how those atoms connect to one another. Simply knowing their position is not enough. At a maximum, if we only knew about the atoms, we might be able to work out how they were connected ---- at the cost of a great deal of totally unnecessary computer processing. At a minimum, we might not even be able to do that. Long long life, Thomas Donaldson Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=5186