**2. Therapy**

Deep brain stimulation is a procedure employed, in terms purely of therapy, for the treatment of Parkinson's disease, Tourette's syndrome, epilepsy and clinical depression. Electrical signals at a frequency of 150 to 190 Hz are applied into the thalamus or subthalamic nucleus of the brain. The effect of these signals is to counteract the original problem. The deep brain electrodes can also be connected bi-directionally with a computer such that electrical activity in the brain can be monitored. Using AI techniques, a better understanding of the nature of Parkinson's disease has been obtained [2]. It is also possible for the monitoring computer to be located remotely from the patient, e.g., in a different country. Hence, signals within the brain can be tracked in real time and fed into a computer. The computer can analyse these signals and generate alternative signals that are fed directly back into the brain. Effectively, in this example, part of that person's brain resides not only outside their body but physically in a different country.

In another vein, in the 1950s, Olds and Milner [3] implanted electrodes directly into the lateral hypothalamus of the brain in several rats. Each electrode was connected to a stimulator that was activated by the touch of a button. The electrical stimulation was aimed at causing feelings of pleasure. The rats were taught how to press the button themselves, giving them the feeling of pleasure. They were then given the choice between the pleasure button and a button that resulted in them being given food. The rats continually chose the pleasure button even when they were hungry and starving. Very similar results were obtained when rats were replaced by monkeys.

Consider that a brain implant can be employed to overcome clinical depression; when it is switched on, the patient can feel as though a black cloud is leaving them. The individual person is being given positive feelings. It is essentially acting as an electronic drug. Indeed, as the human brain operates electrochemically, drugs can potentially take on either form—electrical or chemical. It is merely a matter of conformity that thus far, almost all drugs are chemically based. In the years ahead, this could change dramatically. Possibly even headaches will be removed by electronic means in the future, as this would be more direct with potentially fewer side effects. Conversely, this could readily be extended for the computer to bring about positive feelings under certain circumstances and negative feelings in other cases.

Communication is vitally important for humans. Philip Kennedy developed an operable system which allowed an individual with paralysis to spell words by modulating their brain activity. Kennedy's device used two simple electrodes: The first was implanted in an intact motor cortical region and was used to move a cursor among a group of letters. The second was implanted in a different motor region and was used to indicate that a selection had been made [4]. As the patient thought about moving their fingers, these signals were translated into signals to move and stop a computer cursor. The patient could see where the cursor was on a computer screen. They could choose when to stop thinking about moving. In this way, words could slowly be spelt out letter by letter or heating and lighting could

be simply controlled. Although initiated as a therapeutic experiment, the recipient of the implant exhibited abilities that were beyond the human norm.

Another good example is that of Neil Harbisson, who is otherwise colour blind and has a camera which is attached to his skull. Di fferent colours cause the frequency of vibrations to his skull to vary. As a result, he has learned a very high degree of colour discrimination. The technology translates colour frequencies into sound frequencies [5] which are translated into vibrations via an actuator. Initially, Harbisson memorised the frequencies related to each colour, but subsequently he decided to permanently attach the set-up to his head. The project was developed further so that Harbisson was able to perceive colour saturation as well as colour hues. Software was then developed that enabled Harbisson to perceive up to 360 di fferent hues through microtones and saturation through di fferent volume levels. What is particularly interesting about Harbisson's experience is that his discrimination between di fferent colours has improved over time as his brain has adjusted to the di fferent vibrations experienced. Clearly, the extent of brain adaptability is a pointer to what can be expected in general with regard to either extending the present range of sensory input or rather inputting a complex range of new sensory input information into the human brain that until now has not been possible. It is another example of a therapeutic treatment that results in the recipient having abilities beyond the human norm.

Perhaps the most commonly encountered brain–machine interface is the cochlear implant, intended to repair an individual's hearing by connecting electrodes to directly stimulate the auditory nerve fibres. It is estimated that there are at least 600,000 recipients thus far. At first, the person realises electronic signals that may have little meaning, but gradually, their brain adapts to the signals input and learns to recognise them as specific sounds. In some cases, it may be that the individual can comprehend certain higher frequency sounds for the first time in their life. Ordinarily, the frequency range input is aimed at mimicking as much as possible the normal human frequency input. However, there is of course nothing stopping a di fferent frequency range from being input for the brain to learn and/or for the auditory nerve to be stimulated directly from a network rather than from sound waves.

Arguably the technology which has proven to be of the most practically successful in this area is the microelectrode array, currently referred to as the BrainGate. The array is made up of 100 silicon spikes which are 1.5 mm long with a platinum electrode on each tip. The electrodes are linked with platinum wires, and in this way, the array can be employed to both monitor neural activity and also apply stimulating currents. Several trials have been carried out using human studies. In experiments, the array has been fired into either the human brain or nervous system. In the first set of these experiments to be considered, the array has been employed in a purely recording role for therapeutic results.

Electrical activity from a few neurons monitored by the array electrodes, positioned in the motor cortex, has been decoded into a signal that enabled a severely paralysed individual to position a cursor on a computer screen using neural signals for control in combination with visual feedback. The same technique was later deployed to allow the individual recipient, who was paralysed, to operate a robot arm even to the extent of learning to feed themselves in a rudimentary fashion by maintaining su fficient control over the robot arm [6].

We later consider directly the use of this BrainGate implant in a set of experiments which were set up specifically to look at enhancement possibilities.

### **3. O** ff **the Shelf—External Electrodes**

In numerous research labs around the world, the concept of a brain–computer interface is actually realised by an interface between the experimentalist's scalp and a computer, see, e.g., [7]. In this method, weak signals from a brain (in terms of scalp-filtered averages from millions of neurons) are fed into a computer such that after significant AI processing, the individual concerned can learn to think in specific ways such that (typically) a connected truck will turn right maybe 8 times out of 10

when they want it to. This is of course purely a one-directional procedure, other than any feedback via the usual sensory route, e.g., their eyes.

It is not the author's intention to dwell on this area other than to acknowledge the research that is going on and to appreciate the possibility of monitoring a brain in this way to further understand what is happening inside with the hope of detecting neurological problems as encountered in cases such as epilepsy. Arguably several other scanning techniques provide similar or better such results. However, the method's big advantage is of course that external electrodes can be relatively easily attached to the outside of the scalp, which makes such research much more doable, thereby considerably reducing any potential problems of infection or rejection. It is interesting to note, however, that in many research papers on the topic, the same partial conclusion appears, which is [8]: "Instead of placing the electrodes on the scalp if they are placed in the cortex itself, it would provide a better result".

Rather, the focus of this article is fundamentally directed towards brain–computer connections involving an implant (or implants) in the brain or nervous system, thereby achieving a much higher resolution with the ability to extract signals from a handful of neurons. Importantly, such a technique is potentially bi-directional, meaning that both (e fferent) motor signals can be monitored and (a fferent) sensory signals can be applied.

### **4. Some Future Possibilities**

In this section, we look at two possibilities. The second of these is instantly more practical and is based on a set of successfully conducted experiments involving the author. The first, whilst also being based on the scientific experimentation of the author, is perhaps rather more speculative.

It is quite possible to culture networks of dissociated neurons grown in vitro in a chamber. The neurons are provided with suitable environmental conditions and nutrition. A flat microelectrode array is embedded in the base of the chamber, thereby providing a bi-directional electrical interface with the neuronal culture. The neurons in the culture rapidly reconnect, form a multitude of pathways and communicate with each other by both chemical and electrical means. Although for most research in the field thus far, the neurons are typically taken from rat embryos, it is quite possible to use human neurons instead once su fficient connections have been made between the neurons so that, in research, the cultured brain is given a robot body with the ability to sense the world and move around in it [9].

Humans are essentially our brains [10]. Our bodies keep our brains functioning, transport them, provide some sensory input and enable each brain to interact with the outside world. However, through evolution, our brains have largely become dependent on the bodies that carry them around. But apart from some limited transplants or artificial organs, when some physical feature malfunctions, then that may well mean the human dies, even though there is essentially nothing wrong with the brain, i.e., the essential self is well but it dies because a (possibly in the future) trivial physical element no longer functions appropriately.

Therefore, if we look to the future, theoretically for the moment, it seems sensible to consider directly keeping the brain alive, somewhat akin to the experimentation described, without its dependence on its physical body. As an example, if a person has liver cancer and dies from this, even though their brain was perfectly OK, it is an unnecessary death. Rather than considering further research into the treatment of such cancers and the like, surely it is better to consider ways to keep the brain alive outside of its present human body, as we presently do with culture experimentation. There is, though, the matter of scale—presently, cultures of typically 150 thousand neurons are supported, whereas with the human brain, we would need to support 100 billion neurons.

In this way, the body could be designed merely to fit around the brain. If something in the body functions incorrectly or stops working, then it can simply be replaced or upgraded. If a better body component becomes available, whether biological or technological, then the newer, more powerful option can be selected. The range of sensory input can be what you want, the abilities of the body can be designed to suit. Life expectancy would be much more enhanced as all of the body could be replaced as needed. A lifetime would be totally the brain's lifetime.

In a di fferent vein, the author also carried out an experiment in collaboration with surgeons at the Department of Neurosurgery and Neurosciences at the John Radcli ffe Hospital, Oxford [11]. During a two-hour procedure, a BrainGate array was surgically implanted into the median nerve fibres of the left arm of the author. The array measured 4 by 4 mm with each of the electrodes being 1.5 mm in length. With the median nerve fascicle estimated to be 4 mm in diameter; this meant that the electrodes penetrated well into the fascicle.

The array was pneumatically inserted into the median nerve such that the body of the array sat adjacent to the nerve fibres, with the electrodes penetrating into the fascicle. The array was positioned just below the wrist, following a 4 cm long incision. A further incision, 2 cm long, was made 16 cm proximal to the wrist. The two incisions were connected by a tunnelling procedure such that wires from the array ran up the inside of the left arm, where they exited and connected onto an electrical terminal pad which remained external. The arrangements described remained permanently in place for 96 days. During that time, a series of experiments was conducted.

The terminal pad was directly linked to a computer terminal either by hard wiring or preferably by means of a wireless connection, which enabled mobility. By this means, a ready connection was arranged with the internet. The link was bi-directional such that motor signals due to hand movements could be monitored and decoded whilst, via the same implant, sensory signals from a remote source could be applied to stimulate the nerve fascicle and thereby be fielded by my brain.

A number of di fferent trials relating to human enhancement were successfully realised. These were:

1. Extrasensory (ultrasonic) input was used to detect the distance to objects:

It took approximately two weeks for us to find a stimulating current that my brain could reliably recognise. All we had to go on initially was previous testing on chicken sciatic nerves [12]. Although the waveform was similar, we needed a stronger current at a voltage of 50 v. I wore a blindfold and heard a click; sometimes a stimulating pulse had been applied to my nervous system, sometimes not. After 1 week, I correctly detected 70% of pulses (50% would be the same as guessing); however, after 2 weeks, it was 100%. During that time, for about 2 h every day, we were doing the click testing in the lab. But then, we linked ultrasonic sensors on a baseball cap to cause the stimulation. The closer an object was, the more the frequency of pulses increased. My brain made sense of this immediately: Even with a blindfold on, I knew how far away objects were. So much so that when Iain, one of the researchers, suddenly moved a board quickly towards me, it was very scary—I could detect that something was coming towards me very quickly but I did not know what it was. It felt like a reactive response.

2. A wheelchair was driven solely by motor neural signals:

It was pretty straightforward from the early days of the experiment that we could detect neural signals by means of the implant. Neural signals are very di fferent to, for example, noise or muscular signals, so it is not such a di fficult signal processing problem. Understanding what those signals actually mean is a more di fficult proposition. However, it did not take a complex algorithm to link neural signals with hand-opening or hand-closing movements. We were working on the project with two hospitals and had obtained a wheelchair from the National Spinal Injuries Centre in Stoke Mandeville Hospital. We then used a simple menu device to link the neural signals with wheelchair operation. By opening and closing my hand, I could move down the menu to my selection—e.g., forwards, backwards, slow, fast—and the chair would follow my wishes. The point is that this would operate exactly the same way, with the same implant, for someone with a spinal injury but with the implant in their motor cortex. It could of course be a car, or any vehicle, rather than a wheelchair.

3. The behaviour of a group of small robots was altered by motor signals:

We were fortunate to have a lab full of little robots. These could exhibit di fferent "emotional" behaviours via ultrasonic sensors. Thus, on detecting another robot, they could act as though they were

scared and try and escape or conversely act aggressively and try and catch the detected robot. This was simply linked to my neural signals such that when my hand was closed, the robots acted in a friendly way, whereas when my hand was open, the robots were aggressive. It was a simple experiment and merely went to show how powerful such an implant can potentially make the recipient.


I have something of a communications background, and for me, this was the icing on the cake. A volunteer assisted by having microneurography. Essentially, two very thin needles were pushed into the nervous system in their left arm. With this in place, we set up in the lab with a group of people around the volunteer and another group around me—we had a variety of di fferent observers to oversee what we were doing. The volunteer and I were not able to see each other. We set the experiment up purely based on hand closures. When the volunteer closed their hand, I received a stimulating pulse on my nervous system, and the same happened vice versa. For me, it meant that my brain recognised the pulse. I shouted out "Yes" every time I felt a pulse, but only when I felt a pulse. Only the group around the volunteer could witness when they had closed their hand and when not. We achieved this with 100% success—the same being true in reverse. What I found exciting was that as the groups were splitting up, I felt a couple of quick pulses one after the other. Subsequently, the volunteer confirmed that they had done this. It was a "secret" message between the two of us, a new means of communication.

In every one of the experiments just described, the raison d' être could be heralded as being for therapeutic purposes, e.g., ultrasonic input for a person who is blind. However, in each case, the trial can rather be regarded as an enhancement beyond the human norm. Clearly, such enhancements throw up a multitude of intellectual and technological opportunities, but they also realise a range of ethical considerations. Indeed, there may be persons who do not wish others to exhibit such enhancements, whereas, on the other hand, individual freedom dictates that an individual can so upgrade themselves if they want. That said, individual freedom is not an absolute in many societies in the world today, and in any case, the freedom of one individual must be balanced with the e ffect of that individual on the freedom of others.
