We're a little crazy, about science!

Day 254: Review – A Brain to Spine Interface for Transferring Artificial Sensory Information

Fig.1 of paper showing drawing of implanted electrodes and the two experimental setups for the rat.

Experimental setup for artificial sensory discrimination using DCS and brain-to-spine interface. (a) Rats were implanted with recording electrodes in motor cortex (M1), somatosensory cortex (S1) and striatum (STR) and dorsal column stimulating electrodes in the thoracic epidural space.  (b) Behavioral setup for artificial sensory discrimination using DCS (c) Setup for the brain-to-spine interface consisted of two modified aperture width tactile discrimination boxes.

If you ever were to read one of my review papers, this one’s for you. It’s so awesome and falls in line fairly closely to the things I want to accomplish, albeit going a different route to get there. I’m super excited to share this with all of you and I hope I did the study justice in my summation and while I admit, I had far too much enthusiasm with this one, it shouldn’t take away from just how amazing this is, see for yourself! The study is open access too, so if you want to know more details, you can go take a look!

Brain machine interfaces (BMI) have made considerable advancements in recent years and have proven to be a valuable tool as the basis for assistive and restorative technologies for people with neurological disabilities. While traditional BMI are used to control the real-time movements of artificial limbs or cursors, somatosensory feedback is often overlooked. By combining BMI and spinal cord stimulation (SCS), this study explores the possibility of creating a brain-to-spine interface. The immediate benefit for technology like this would be for people who are living with spinal cord injury, but this has added benefit for returning sensory feedback to people who use prosthetics.

In this study, researchers used Long-Evens rats (n = 12) and the study was divided into two experiments. In the first was if dorsal column SCS using different frequency and pulse patterns could induce distinguishable artificial sensations. This was tested by training the rat to perform one of two different tasks depending on the stimulus; either a single pulse or 100 pulses at 333 Hz. The second experiment was broken into two parts, the first was to see if tactile whisker response recorded from the neocortex of one “encoder” rat (n = 6) could be transferred to a second “decoder” rat (n = 6) using SCS. If the rat performed the correct response to the stimuli, the trial was considered a success. The second part used artificial stimuli similar to the first experiment on the “encoder” rat and the “decoder” rat, once again if the “decoder” rat performed the correct response the trial was considered a success.

The first experiment using artificially induced tactile feedback had 20 trials to train the decoder and 100 trials total. The rat success levels started at chance levels and ended at a 91.75% accuracy. Two rats were then trained on a third condition using a third stimulus condition of 100 pulses at 100 Hz and had an 80% accuracy in 15 sessions. One rat learned a fourth stimulus pattern, 5 busts at 20 pulses each with a between pulse frequency of 2 Hz and an inner pulse frequency of 333 Hz and also had an 80.33% accuracy. The first part of the second experiment the decoder rats had to perform a “virtual narrow” or “virtual wide” response based on the encoder rat whisker stimulus. The encoder rats performed the appropriate response with an accuracy of 93.21% and the decoder rats performed it with an accuracy of 71.3% and a best session performance of 82% in 100 trials split over 11 sessions. For part two, the encoder rats performed slightly better at 87.88% accuracy than the decoder rats at 84.03% accuracy with a best session performance of 92.22% accuracy. In both cases, when the amplifier was turned off the accuracy returned to chance levels.

This study demonstrates several different technological paradigms, namely that we can generate sensory feedback using spinal cord stimulation, but also that a real-time brain-to-spine interface can be used to restore or augment function. This study is interesting for a myriad of reasons, but what I find most interesting is that they managed to decode neural signals into something that can be decoded via the spinal cord. This is important because the brain, spine, and muscles all appear to use different styles of communication. The spinal cord is ostensibly the Rosetta stone of the body, being able to translate inputs between the muscles and the brain. By gaining an understanding of how the spinal cord can do this, we not only learn about how the spinal cord itself functions, but we also gain a clearer understanding of how the brain communicates. This knowledge has the potential to greatly advance BMI research and could one day mitigate the effects of spinal cord injury and other neurologically based diseases.


Yadav, A.P., Li, D. & Nicolelis, M.A.L. A Brain to Spine Interface for Transferring Artificial Sensory Information. Sci Rep 10, 900 (2020). DOI: 10.1038/s41598-020-57617-3

But enough about us, what about you?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.