Re: Whether electric shock therapy can create new memories

Murthy-

     You ask an interesting question. To address it, I think it's best if I describe the integral relation between the action potential and neurotransmitter release in the brain. Then we can consider the bigger issue of how memories are formed... Please pardon me, as this answer became somewhat lengthy - If you want the short and sweet answer, it's at the end!

     To start, I should state that the release of neurotransmitters and the electrical impulses which lead to an action potential are dependent on each other, not two separate processes that tend to occur together.
     I think you understand that the action potential is an electrical event. To read more about what an action potential is, and how it's formed, take a look at http://faculty.washington.edu/chudler/ap.html. When an action potential travels down an axon and reaches a synapse, it induces the release of neurotransmitter. (more on the specifics of this process can be found at http://www.madsci.org/posts/archives/dec2000/977272802.Ns.r.html ) As you pointed out, the release of neurotransmitters is the mode of communication between neurons. I should stress that the neurons make neurotransmitters all the time, but only release them when an action potential occurs; and only by releasing neurotransmitter does communication between the neurons occur.
     The next question to be asked is: 'how do neurotransmitters really mediate communication between cells?' The simple answer is that they induce an electrical current in the neuron they are 'talking to.' Thus, the end effect of a neurotransmitter is to put electrical charge in the 'postsynaptic' cell (the cell that received the neurotransmitter, while the one that released it is referred to as a 'presynaptic' cell). If enough charge collects in the postsynaptic cell, it will fire an action potential, which will cause it to release neurotransmitter onto yet another cell. I should say that a neuron receives input from 100-1000 cells, and makes contacts onto that many cells, too.

     Thus the process comes full circle: electrical activity causes neurotransmitter release, which causes electrical activity in the postsynaptic cell, which will cause neurotransmitter release... I think it's fair to say that a neurotransmitter is like a chemical continuation of an action potential, or vice versa. But they are different for a reason. I'll describe more about that point below.

     When considering the effects of 'externally applied' electrical charge, the basic principle doesn't change. Enough of a shock will cause an action potential to fire without the requisite input. Thus, the cell will release neurotransmitter onto postsynaptic cells and cause current, and the large amount neurotransmitter released in response to large shock is likely to cause an action potential in these cells. Excitation of these postsynaptic cells will start the whole series of events downstream in the network. I think it's also fair to say many of those downstream cells may be activated by the shock, such that the whole circuit is activated somewhat simultaneously, and then starts firing off in a rather unorganized fashion (as different neurons release large amounts of neurotransmitters onto different cells, which activate other cells, etc). When this happens spontaneously in people, we call it epilepsy.

     As you also pointed out, people have claimed to recall various memories after electro shock therapy (EST) or even selective brain stimulation. So you pose the question, "can EST actually cause the formation of new memories?" I think the answer is that EST definitely causes long-term changes in the brain, perhaps similar to the processes that occur when memories are formed. But I don't think it's likely to cause a true 'memory.' Let me explain a little bit about how we think memories are formed and stored: (you can skip these paragraphs if you already know about this stuff, but I've included it for completeness)
     While we don't have any smoking guns about what a memory looks like in the neuronal terms, we have some clues that let us hypothesize about what's going on. First, at the cellular level, there are processes called Long Term Potentiation (LTP) and the converse, LT Depression. LTP is a mechanism to increase the level of communication between one cell and another; LTD weakens the level of communication. The details are a bit involved, but the basic idea is this: If a presynaptic cell's release of neurotransmitter often causes action potentials in the postsynaptic cell, the connection between the two cells is strengthened. It seems like a handy cellular phenomenon that could lead to memory formation: Two cells that are active together in a particular experience could encode some aspect of the memory.
     These processes depend on neurotransmitter release - a modulation of the amounts of neurotransmitter released or the responsiveness of the postsynaptic cell to neurotransmitter occur. This is an example of a good reason for having neurotransmitters - they allow for fairly easy modification of information between cells.

     At a whole-brain level, there is a structure known as the hippocampus that seems integrally related to memory formation. Most people have two hippocampi, one on either side of your head. Several people have had them destroyed by stroke or surgery, with the end result being that the patients cannot form new memories. They can remember their childhood; but if you met them for the first time today and walked away for 10 minutes, when you returned they would have no idea who you were or that you had just been there. If you want read more about the hippocampus, there are several answers in the madsci archives, as well as many articles all over the web!
     We are not entirely sure where memories reside, but it appears that the cortex is the most likely place. But we don't know if memories are stuck in one place, or if they are distributed across the brain. I think it's likely to be the latter.

     So to finally get to the issue of whether EST can produce memories, I hypothesize that EST can produce a "memory," but not a true, specific one. As you said, it can certainly 'bring back' a memory - it's activating a circuit in your head somewhere. An EST-induced memory is not necessarily "true;" you may remember something that never happened. I also don't think EST could cause a person to have a specific memory - I don't think you could specifically shock someone into thinking that they had soup for dinner, or that they were raised in Turkey, or something like that. The memory that formed would probably be random. Plus, EST is a fairly traumatic experience, and people's memories around traumatic experiences tend to be kind of fuzzy.
     Why do I say a specific memory won't occur? What seems to be critical in memory formation is that a selective set of neurons is activated and likely to be changing. When you are going out to dinner or visiting friends, while your whole brain is active, particular neurons in various parts of the brain are encoding the experiences, and somehow marking them as being salient or not. You'll remember the salient ones. In EST, large portions of the brain are being stimulated, such that massive groups of neurons are being excited at the same time - this sort of stimulation lacks the neuronal specificity that is likely for specific memory formation; but random memories seem theoretically possible. That could beg the question, Could we invent a device to induce specific memories directly into the brain? My answer to that question can be found here: 1017588620.Cs

     Externally applied shocks will definitely electrically excite cells, which will cause a massive release in neurotransmitters. This excitation and neurotransmitter release will be propagated down many brain circuits. While this type of phenomenon could cause a permanent change in the brain that would lead to a 'memory,' one could not control what is going to be 'remembered', nor if what is remembered occurred in actuality. Plus, the trauma of such an intense shock is likely to prevent proper memory formation in the individual, such that really aberrant things are remembered!

     I hope this sufficiently answers the question! The question is a good one.

-Alex G
cgoddard@fas.harvard.edu