Heroes and Communicators
In Managing Knowledge I talk about scientists as heroes and I talk about engineers as communicators. Professor Kevin Warwick does nothing to dispel either of these ideas. He is both a scientist and an engineer and he is certainly a communicator. He has put his personal safety where his mouth (or nervous system) is in the name of Science.
When he starts to talk you listen because you know that you are hearing the ideas that progress is made of.
The Heart of the Brain
Lets put aside cybernetics, sensory enhancements, artificial intelligence, neural interfaces, sensory substitution, deep brain stimulation, Parkinsons, hebbian learning, cochleaar implants, visual neuroprosthetics and the more headline attracting aspects of his work and go straight to what I thought was the most tantalising idea he gave voice to tonight.
The brain is an amazing piece of equipment. It records, stores and processes the most esoteric data about the world around it and turns this information into ideas. In terms of processing, computing is only in the stone age in comparison to the brain.
The brain takes the amazing complexity of thought and translates it into trivial signals as it tries to communicate with other brains through the medium of the clumsy symbols of language. Although I would suggest that there is more to communication than language all the the sensory signals of emotion, body language, pitch, tone, volume, speed and all the rest are subject to the same problems as the spoken word. Communication is the issue. High bandwidth content, low bandwidth signal.
What if brains could bypass all that and we could plug directly into each other's nervous system? No more misunderstanding, high bandwidth data transfer and lots of learning.
What Warwick is after is no less than telepathy and he may be on the right track. He has connected to his nervous system through a chip in his arm and proved that electronics (hardware) can indeed be interfaced directly with the nervous system (software). He went one step further and when his wife agreed to have a chip in her arm he wired their nervous systems together.
Now maybe I am just an old romantic but it seems to me that this is a cyberpunk tale of true romance.
The talk ranged from EEG Analysis through to Neural Implants; from therapy to human enhancement.
EEG (electroencephalograph) analysis is the tool of choice for neuroscientists these days. You see pictures of people with electrodes attached to their scalps and you know that the brain is being mapped and that the future is coming. As exact a science as it is, it is still subject to subjectivity. 20 to 29 electrodes pick up millions of neurons but you still have to get your brain working a “certain way” to drive a small buggy about the lab and according to Professor Warwick we are still a long way from driving a Jag around the M25 by thought.
There is so much to pick apart in that idea and how it is stated and we are only minutes into the talk. If you are listening you notice something different here already. Most of the stuff you read about neuroscience is all about mapping which bits of the brain do what and understanding how to fix or avoid damage to the brain. What we are dealing with tonight is the idea of a bi directional interface. We are considering the ramifications of tapping into the signals the brain sends through the nervous system and allowing them to reach out beyond the confines of the body.
Sight and Sound
Sight and sound are among our best ways to interface with the world outside our skull and this leads us into the territory of cochlear implants and visual neuroprosthesis. We are following in the footseps of Count AlessandroVolta who invented the battery and promptly stuck electrodes in his ears and closed the circuit in order to hear bubbling. We are also scampering after Sir Isaac Newton who stuck wires behind his eyes in a series of anecdotal experiments he carried out on his own sight. Suddenly Professor Warwick seems to join a very salubrious group with his in-arm chip.
We watched a video of a woman who had received a cochlear implant and regained her hearing. At first it was all noise. Birdsong, which she had never heard before, slowly came into focus as her brain made sense of the signal over time. This is crucial information and starts heading us toward neuroplasticity. At first researchers thought that they would have to plug into millions of neurons to develop the cochlear implant but it turns out they only need about 28 electrodes. This is because the cochlear implant is not mimicking or reproducing sound. It is making use of sensory substitution, the ability of the brain to start interpreting something else, in this case electrical stimulation of the auditory nerve, as sound, even though it is not sound. As the eye fills in the blank pieces in some optical illusions the brain fills in the missing sounds and the brain hears.
Visual Neuroprosthesis is not as advanced. It is a more complex problem. At the moment it is grappling with the problem of powering the sorts of prosthesis required to stimulate the optic nerves in the retina required to make an image.
Bach Y Rita
It was here that I wanted to jump up and start asking questions about the sensory substitution work of Paul Bach Y Rita which took an entirely different approach. Read on because I got my chance later when I cornered Professor Warwick over an egg wrap and a glass of orange juice.
Back at the talk we were looking at a photo of the 1950 Ratio Club from which another hero looked down at us all. Alan Turing shared the photo with the club's founder Giles Brindley and with Professor Warwick's PhD supervisor. Neural links indeed!
His own students surgically implant magnets into their fingers to experiment with sensory substitution. He admits that many of the ideas have been around in the arts for a while but that this body modification is about scientific research. They are developing extra senses that react to distance, infra red, temperature and music. Then he mentioned Bach Y Rita and his brain port. A sensor on the tongue conveys all sorts of information to the brain. They are picking up on it and exploring the idea because the tongue is a shortcut to the brain. Bach Y Rita used it to convey balance information to the vestibular system and images to the visual cortex by associating sensations created on the tongue with that kind of information and letting the brain adapt.
What the team at Reading have discovered is that the same signals will manifest differently to different people leading them to believe that some of the signals are hardwired. For instance a triangle broadcast through the tongue electrode can be identified by everyone as a triangle with practice; but the orientation of the triangle will be different for different people. Practice may improve the definition of the shape but the orientation remains personal and immutable. If your triangle is upside down it stays upside down.
Rat Brains and Robots
The next section was kicked off with a video about rat brained robots and a video of Jeremy Clarkson interviewing the professor for “Inventions that changed the world”.
I have had a bit much of Clarkson in the media over the past few days so I focussed on the robots who had 50 to 100 brain cells. This apparently gives them the intelligence of a slug.
Notice how I am avoiding all sorts of temptation here?
The more advanced robots have tens of thousands of brain cells and have the intelligence of wasps or bees.
The brain cells are grown on a multi electrode array (MEA) from rat embryo cells. Within seconds they start to try to make connections. This is the fascinating stuff. Making connections is what neurons do, it's their nature. This rat brain is telling us a lot about human nature. They put the “brain” in a robot body (a small being on wheels) and “teach” it to avoid obstacles. When they take it out and examine it they can see the growth of connections happening under a microscope. This is mind boggling.
The robots learn by habit. This is called hebbian learning. I am particularly interested here because this is the theory that "neurons that fire together wire together" and is close to my heart and my thesis in the Trousers of Reality books.
The point of these experiments is to discover what memory is and and how memories are formed in the brain. This understanding is crucial if we are to tackle dementia and other debilitating brain disorders.
The next stage is using human neurons. While the rat neurons are arranged in a two dimensional array the human neurons are grown in the US and delivered as three dimensional arrays. This puts the number of neurons in the cybernetic brain up to thirty million.
Ethics are starting to march with purpose into this research. Given that Professor Warwick has been selected by the Institute of Physics as one of only seven eminent scientists to illustrate the ethical impact of their work (along with Galileo, Einstein, Curie, Nobel, Oppenheimer and Rotblat), it is not surprising that he is asking some difficult questions. One of the questions he considers and which we will all soon be considering is:
When that number starts to approach and overtake the 100 billion neurons in a human brain what sort of rights will it have? What sort of rights will we have?
Deep Brain Stimulation
As we considered this we watched a video about Deep Brain Stimulation. Electrodes were inserted into the subthalamic nucleus of a patient with Parkinsons disease. Before the procedure the patient had very little control and had devastating tremors and muscle seizure. The drugs had stopped working. When the electrodes were powered on, the tremors disappeared and the patient not only walked but twirled about much to the delight of the audience at the filming of the video and the audience in the University of Coventry.
These electrodes use a lot of power. Batteries have to be replaced every 18 months and that involves surgery which is both expensive and runs the risk of all surgery. Enter stage left the cybernetic brains we were looking at earlier. They can be trained to monitor electrical activity of the nervous system. They can be trained to predict tremors 15 to 20 seconds before they happen. In this feedback loop the AI kicks in to power the electrodes only when they are needed. This would significantly increase the life of the batteries (which are implanted in the chest cavity).
Professor Warwick then talked about his famous cybernetic arm. While describing the gorey details of the experiment would be interesting, his thoughts about the results of the experiment are more interesting.
Having a bi directional interface via a Utah Array implanted in his arm for three months he was able to connect wires to his nervous system. This extended his nervous system into those wires and the wires became part of his nervous system. This leads to some interesting thoughts:
The brain and body do not have to be in the same place – he operated a mechanical hand in Reading from New York by connecting his nervous system to the internet. He sent signals to the arm and interpreted feedback from the hand to pick up things without breaking them. Could we use this technology to allow our consciousness to go places unimagined?
If we can use these cybernetics to connect to machines can we use them to connect brains to brains and extend the ways we can communicate. He connected his nervous system to his wife's and overheard the message she sent to her own hand.
The talk finished with an experiment. A lucky member of the audience wore a cap with a sensory device that warned of proximity to objects with a vibrating noise. She quickly learned to use this feedback to negotiate around a stage full of people.
After the lecture the Professor joined us for a bite to eat and I was delighted to be able to congratulate him on both his lecture and his pioneering work. I asked him about Bach Y Rita and to my immense relief he agrees with me that Bach Y Rita never got the recognition he deserved and that his work is an outstanding achievement of science, creativity and lateral thinking.
Blog Home Page