During World War II, mathematical phenom Alan Turing cracked the German Enigma code with help from fellow mathematicians. In doing so, he transformed gibberish into German.
Inspired by his incredible feat of cryptology, a team of biomedical engineers has developed a code-breaking method of their own. Their cryptographic technique, however, decodes the activity of motor neurons in the brains of monkeys.
Now, before you question the practicality of this research, the team say their work was used to predict arm movement based on the monkeys' brain data. They hope the code-breaking technique can be used as a stepping stone for future work that decodes more complex patterns of muscle activation. This ability could prove useful for those with prosthetic devices.
The study, published in Nature Biomedical Engineering, was led by Konrad Kording, a Penn Integrates Knowledge Professor, and Eva Dyer, previously a postdoctoral researcher in Kording's lab and now an assistant professor at the Georgia Institute of Technology.
For their research, the team analyzed the neuronal data of three rhesus macaques as they accomplished certain tasks, such as reaching for various targets. This allowed the scientists to monitor the spikes of electrical activity associated with each specific movement.
While there are current brain-computer interfaces that also use data for robotic prosthetics, they do so via “supervised learning”. This means they use patterns found in the activity of neurons to reconstruct movement.
"In cryptography, 'supervised learning' would be called a 'known plaintext attack,'" Kording said in a statement. "That is, we have both the encrypted and unencrypted message and just need to figure out the rules that turn one into the other. What we wanted to do in this study was to be able to decode the brain, using a movement model, from the encrypted message alone."
The team essentially wanted to find a mathematical approach to map the patterns they found in the monkeys’ brains.
“The algorithm tries a range of possible decoders until we get something where the output looks like typical movements," Kording said. "There are issues scaling this up – it's a hard computer science problem – but this is a proof-of-concept that cryptanalysis can work in the context of neural activity.”
In terms of the utility of this technique for prosthetics, the team says that a robotic limb that can interpret a user’s thoughts without calibration may improve the quality-of-life of patients. Currently, patients must be trained to use their robotic limbs, which can be complicated and time-consuming. The team hopes to bypass this present necessity.
Clearly, their work is at a preliminary stage. However, Kording notes that ”[W]e should be able to do this within the next decade.” Whether that is a tad optimistic is up for review.
“The Germans were actively working against decryption and modern ciphertexts are basically impossible to break," Kording added. "We have it easier. The brain ended up with this encryption system through natural selection, so it's essentially making the same kind of 'mistakes' that allowed us to crack Enigma in the first place."