A password will be emailed to you.

Scientists have hijacked the memories of mice as they slept.  That’s the headline they’re going with, and it is rightly quite sensational, as mainstream media is wont to be.  It may not be entirely accurate though – to our great surprise.

I’m talking about a paper that was published yesterday in the scientific journal Nature (Neuroscience).  It expounds on a breakthrough experiment whereby researchers from the National Center for Scientific Research (CRNS) in France claim to have implanted conscious memories into mice as they slept.[1]  That requires a bit of explanation though.

The group, led by Karim Benchenan of the CRNS Brain Plasticity Unit’s MOBs Team, achieved this astounding breakthrough by implanting electrodes into the brains of mice, specifically targeting a type of brain cell called place cell neurons.  Place cells, which were discovered in 1971, are a specialised type of neuron that are key to remembering where one is and where one has been.  They act as a sort of map inside your head, with individual cells firing when you’re at a certain location, creating a memory of that spot, so that you can find your way back there in the future, should you need to.

Now, when you – or any of us – sleep our brains undertake to review the memories we’ve made during our waking day.  That process entails a firing of the neurons involved in those memories, just as when the original experience occurred, and scientists can monitor this process.

By comparing neural scans of the mice from a period of exploration in a maze, to their later subconscious review of the associated memories, paying attention to place cell neurons, they were able to identify which neurons were associated with memories of which places inside the maze.  Once identified, they used the electrodes implanted in the mouse’s hippocampus, to stimulate a pleasant feeling at the same time as targeted place cell neurons fired during recall.  They effectively created an association between the memory of whatever location was involved and a pleasant feeling, or a reward so to speak.

The interesting thing is, once the mice woke from their short nap, they immediately headed straight for the location now associated to the pleasant feeling, as though looking to recreate the experience.

This research has obvious potential to transform treatments for post-traumatic-stress-disorder, obsessive-compulsive disorders, and even schizophrenia and depression, by allowing clinicians to alter memories, building positive associations where traumatic or problematic ones once existed.

Benchenane insists, and it seems Nature’s peer-reviewers agree, that the mice’s behaviour following the procedure is proof that they have artificially created a conscious memory.  They qualify it by calling it an “association between a particular place and a reward [that] can be consciously access by the mouse.”  But is adding a sensory association to an already existing memory really creating a conscious memory?

The same effect has been sought, and variably achieved, through behavioural conditioning by various means, but the product in that case is always a subconscious association.  As seen with smokers trained to associate cigarettes with the smell of rotten eggs.  The difference seems apparent, but it might be an illusion.

The sole reason Benchenane and his colleagues believe this is the first example of an artificially induced conscious memory is the fact that the mice actively sought out the location associated with the memory and reward upon waking.  Benchenane states, had the mice wandered randomly and simply stopped and focused on that key location once they stumbled upon it, that would have represented a subconscious memory.

This seems like a little semantic word play, as I see it.  I don’t really question the notion of conscious recognition, so much as the idea that a memory was implanted in the first place.  Here’s my reasoning: the researchers simply added an association to a memory that already existed.  The mouse’s natural cognitive machinery created the memory the same way all of its memories are created, by experiencing stimulus.  The question is, is an association the same as a memory?  Clearly, connecting bits of sensory information is a key part of the memory making process, but isn’t a memory a little bit more than just the sum of its parts?  If we are to properly call this the artificial creation of a conscious memory, then I would expect a more direct influence on the origin of the memory, not simply the introduction of another sensory input to be associated to an already existing memory.

I suppose you could reduce this to the argument between dualism and reductionist materialism (no pun intended).  If consciousness, and therefore memory, are nothing more than emergent properties of the biomechanical processes of the brain, then perhaps Benchenan et al are correct.  It certainly appears that their ability to manipulate memories and sensory input supports the notion of emergent consciousness, but reductionist materialism is by no means a foregone conclusion, not yet at least.

But where does that leave us in the dualism camp?  I don’t have the answer.

No matter your philosophical bent in this case, or even in the case of dualism vs. reductionist materialism, Benchenan’s research is indeed a valuable step forward, and has real potential to drastically change the lives of people suffering with debilitating mental illness.  Even if we can’t agree on exactly how or why it works.



[1] Karim Benchenane, Gaetan de Lavilléon, Marie Masako Lacroix, Laure Rondi-Reig. Explicit memory creation during sleep demonstrates a causal role of place cells in navigation. Nature – Neuroscience March 9, 2015. doi:10.1038/nn.3970