Total Recall: Science Fiction or Science Fact?

    

This past weekend the reboot of Total Recall (titled after the 1990 film starring Arnold Schwarzenegger which was based on Philip K. Dick's "We Can Remember It For You Wholesale") hit theaters, and the timing is rather auspicious since just earlier this year, a paper came out in the journal Science that described a cutting edge attempt to implant artificial memories into the minds of some very special mice. So, the question is, did they succeed? and if so, could the implantation of false memories that are the basis of Total Recall soon be a reality?  This new research suggests that it just might be so... Recently, researchers at the Scripps Research Institute used some very interesting genetic manipulations to try and implant an artificial memory into the mind of a mouse.  To begin, we have to understand what memories are, which is possibly one of the biggest questions still to be answered in neuroscience today.  Yet, for our purposes here, let's agree to assume that a memory is a specific pattern of neurons communicating with one another in the brain.  As an example, when you walk into a place you have never been, let's say a restaurant you are trying for the first time, you sense all of sights, sounds, and smells that are particular to this restaurant.  If it is a Mexican themed restaurant, there are probably bright colors painted on the walls, and the smell of chili and cumin are likely in the air, and all of this activates a specific pattern of neurons in your brain.  Now, if you close your eyes and try to remember what this restaurant looks, smells and feels like, and picture it in your "mind's eye", research shows us that most of the neurons that were active when you were seeing the restaurant for real become active again when you recall the image from memory.  Using this as an analogy, what the researchers at Scripps tried to do was to artificially activate neurons associated with a room that had specific, memorable qualities, similar to the restaurant in the example, then artificially activate those same neurons while inducing fear in a completely different room, and then go back to see if the original room ("Mexican restaurant") would now be associated with the fearful memories.  They did this by genetically engineering mice so that their neurons would express a drug activatable receptor (called DREADD, or designer receptor exclusively activated by designer drug) whenever those neurons had been excited.  They then put the mice into a novel environment for a whole day, kind of like spending the whole day in our Mexican restaurant, and all day long, neurons in the mouse brains were being activated in a pattern that conveyed all of the sensory aspects of the room (all the sights, smells, etc.).  And while those neurons were being activated, they were making this DREADD receptor.  On the following day, the researchers put the mice into a new and different room, and carried out classical fear conditioning.  That is, they exposed the mice to something that would elicit fear, a small foot shock, and after repeated exposure, the mice would normally begin to associate the environment they were in, this new room, with the fear of being shocked.  BUT, what they did in this study was to give some of the mice the designer drug which activated the DREADD receptors, in turn activating the neurons that were active when the mice were in the original room (the "Mexican restaurant").  By doing this, the researchers hoped that they could make the mice think they were in the original room while receiving the foot shock, so that when they put the mice back in the original room on the third day, they would display behavior suggesting that they were afraid to be in the original room , even though they had never been shocked in that room.  Sadly, however, when the mice were put back into the original room on day 3, they did not show any more fear-like behavior than the mice that did not have their DREADD neurons artificially activated, suggesting that artificially activating a specific set of neurons does not completely convince the brain that it is in that environment. Now, if the researchers had stopped there, we would say that the experiment didn't work, and this attempt to implant a false memory had failed, relegating the memory implanting plot lines of Total Recall to the realm of science fiction for a long time to come.  BUT, the researchers wondered, what does the brain think is going on when you artificially activate a set pattern of neurons?  Maybe activating a bunch of neurons that wouldn't be activated naturally just messes things up, BUT, maybe what was actually happening is that the researchers were creating a hybrid memory, where the mice were not just experiencing the artificial sense memories, but also actively experiencing all the aspects of the new environment.  Following our analogy, this would be like moving them from the original Mexican themed restaurant to a McDonald's or Burger King and while in the fast food environment, activating the artificial memory of the Mexican restaurant made the mice feel like they were in a hybrid environment that had a lot of the aspect of both, say a Taco Bell.  If this was the case, then the researchers predicted, the mice would not have learned to fear the second environment UNLESS they were ALSO given the drug to activate the artificial memory.  This is exactly what happened.  When the mice were placed back into the fear conditioning room, without the drug, they did not display any fear behavior, suggesting that they did not associate that environment with the foot shock.  BUT, when they were given the drug again AND placed into the fear conditioning environment, they DID remember, and they exhibited fear behavior.  SO, while the experiment was not a complete success, it seems that the researchers were at least able to alter the mice's perceptions of the environment they were in to a certain degree and create a memory that was different from reality.

P.S. If you want to read the original Philip K. Dick short story, you can get a pdf of it here

Garner AR, Rowland DC, Hwang SY, Baumgaertel K, Roth BL, Kentros C, & Mayford M (2012). Generation of a synthetic memory trace. Science (New York, N.Y.), 335 (6075), 1513-6 PMID: 22442487