Researchers have long known that memories are malleable. We reconstruct or rebuild our memories every time we recall them. There are also many different components to memories that appear to be encoded separately – the information itself, the source of the information, and its truth status, for example, are all stored separately and independently malleable.
Our memory of a location and the emotional memory of whether or not that location has a positive or negative association also appear to be separate.
Researchers publishing in Nature have recently demonstrated this with mice.
First they associated either a negative stimulus or a reward stimulus with a specific location. They then labeled two areas of the brain, the dorsal dentate gyrus (DG) of the hippocampus or the basolateral complex of the amygdala (BLA), that were seen to be active during the conditioning.
They then activated the DG in some mice, and the BLA in the others while exposing the mice to either positive or negative stimuli, but not exposing them again to the conditioned location. They found that they were able to flip the “valance” (positive or negative) of the association to the location by stimulating the DG during stimuli exposure, but not the BLA.
Stimulating the DG, therefore, associated the memory of the previous location with the new stimulus, even without re-exposure to the location.
It is unclear how this will apply to humans. However, this is yet another small piece of evidence that our memories are simply neural circuits in the brain, and we can mess with these memories without theoretical limit. This may not lead directly to the Matrix, zone implants, Total Recall, or your other favorite science-fiction analogy, but it certainly suggests the possibility.