There are two labile states of fear memories. The first occurs shortly after learning, before the consolidation process, during which short-term memories are vulnerable for disruption whereas long-term memories are impervious. After consolidation, with the exposure to the proper environmental cues, such a memory can be retrieved in the brain. Following this retrieval, the second labile stage occurs. During this stage, the aforementioned memory is susceptible to alterations once again. Evidence (Nader, et al. 2000; Kaang, et al. 2009) has illustrated that protein-synthesis dependent reconsolidation processes are required to maintain the original memory. Administering a protein synthesis inhibitor before or immediately after memory retrieval (not quite sure how they elicited the retrieval, or how they measured when exactly it occurred, which would be interesting to know) disrupted the original memory.
Kaang, et al. (2009) hypothesized that these reconstruction processes induced by memory retrieval provide opportunities for memory update or reorganization. Furthermore, they found that new information (stimuli) must be necessary to trigger the destabilization process after memory retrieval. When I first read this, I automatically figured this made sense evolutionarily. Wouldn't we want the opportunity to modify our memories if our environmental stimuli are changing around us? Especially as humans, who live for a much longer time than do rodents, it would be important for our memories to adapt to new technologies and experiences. No question about it. But then as I thought further - I thought about all the downsides this could potentially have.
Let's say as a child you develop a taste aversion to dairy products because you have gotten sick from them on multiple occasions. If your memory serves you right, hopefully you have learned to stay away from dairy. But what happens if one day you have dairy by accident and you don't happen to get sick? Does that mean that single experience should remodel the existing memory you already have of dairy-induced discomfort? This example could be applied throughout species - especially since taste aversions are much more imperative to survival of rodents and less complex animals than humans. Anyway...my point being - if memory reorganization occurs each time a long-term memory is retrieved, how drastic does the changed stimulus have to be to alter the existing memory? Does it have to occur more than once? Does it matter how long the memory has been encoded for, or how many times it has previously been retrieved (strength of memory)? It just doesn't seem like it would make evolutionary sense to have all long-term memories susceptible to disruption each time it is retrieved, does it? Would it make more sense to just form new memories in response to new stimuli rather than modify pre-existing ones? Or would that just require more (ugh, don't make me say it) neural plasticity/neurogenesis than our brains are capable of?
It's important to take into consideration that the studies i've read largely deal with fear-conditioned memories. It would be compelling to study and determine whether or not there are differences in the appearance of labile states in other types of memories (e.g. olfactory aversions, conditioned taste aversions, object/social recognition memory, spatial memory). Would it make sense for some of these to be more susceptible to disruptions more or less routinely?