Presented by http://cyberpunks.org
So, how do we communicate with our brain? What ways does our brain present us with information and how can we tap into that for greater understanding of how we think?
Through the years, scientists have tried many methods, from the downright barbaric to the more modern (but usually incredibly expensive) methods. I'll go through the different methods, explaining what brain signals (or lack therof) each method is attempting to utilise, providing a list of actual devices that use this method.
As explained in the previous section, the brain's signals travel via electro-chemical processes. Though no actual 'electricity' (as in a flow of free electrons) is involved, the movement of charged particles causes a similar effect. Thus it has been possible to devise methods of measuring and manipulating the elecric potentials that occur during brain activity.
Electrical signals (as measured directly) can be used to directly control the movements of a cursor onscreen.
Devices that detect and analyse the electrical potentials of the brain are:
Electrodes
EEG - Electroencephalograph
MEG - Magnetoencephalograph
This technique can be used by any device that measures the electrical potential of the brain. For example, it is often used in conjunction with an EEG. When the brain is presented with a stimulus, it will exhibit a response approximately 300 milliseconds afterward. If you present a subject with this stimulus, you can record their brain patterns uing an EEG and pinpoint where in their brain the response occured. You can thus determine where in the brain that stimulus is dealt with.
Evoked potentials can be used to determine whether a person is looking at a stimulus presented on a screen. it can therefore be used to create what has been dubbed the visual keyboard.
Devices that detect and analyse evoked potentials of the brain are:
EEG - Electroencephalograph
MEG - Magnetoencephalograph
The brain is not the only source of electrical potentials. An Electromyogram (EMG) is measured by placing electrodes over the muscles and reading the potentials created by their movements. If the signals recorded are processed by a computer, we can determine what sets of signals are related to certain movemens of those muscles and thus determine how the muscles (and corresponding limbs) have moved.
These signals have obvious usefulness for the control of prosthetic devices, and also for the determination of the position of limbs for VR.
An Electrooculogram is similar to an electromyogram but the sensors are placed on the muscles around the eyes. This way we can determine the direction of a person's gaze.
This technology could be used to fix many medical problems (see Tonneson et al for more as this falls outside the scope of this project) but also has potential benefits for the emerging technologies for VR. It can also be used in a similar manner (and certainly much more effectively) to the visual keyboard. A company called BioControl Systems inc has done just that with their device that they call the biomuse (see biocontrol below).
NMR stands for Nuclear Magnetic Resonance and is the same technique as used in MRI. Atoms have an inherent rotation that is usually in a random direction. When exposed to a magnetic field, they will align. When the magnetic field is turned off, they will release a little bit of energy that can be measured. See MRI for a more detailed description of the technique.
CBF stands for Cerebral Blood Flow and is an ingenious method for determining brain activity. How does it determine brain activity? Well, the neurons of the brain, like any other cells in the body, require nutrients to work. If they are active, they will therefore need blood to flow to them. If we can meausre the rate of blood flow through a particular region of the brain, we can tell if it is more active than surrounding areas.
The great benefit of using CBF is that you can tell what parts of the brain are active during the processing for a specific type of activity without having to directly touch or measure the brain. Through CBF methods you can test a person while doing various activities and plot the areas that dominate for certain types of related activities.
CBF is used by:
rCBF - regional Cerebral Blood Flow
MRI - Magnetic Resonance Imaging
Regional Cerebral Blood Flow is a further imporvement on the CBF method. It measures the blood flow in the brain by further utilising the way in which cells live. Brain cells require nutrients to operate and they will use glucose to respire when they are being very active. A radioactive glucose can be injected into the brain, then we must detect where the radiation is strongest.
rCBF is used by:
PET - Positron Emission Tomography
The methods below involve direct physical manipulation of the brain and often involve a lot of guesswork. They can usually only give us a vague idea of what is going on in the brain. If we want to be more specific, these techniques must be paired with more precise methods. However, they have been available to us for centuries and have contributed to all our early knowledge of brain function.
The problem with these methods is that due to the potential for irrecoverable nerve damage, ethically, they should only be used if absolutely necessary. Backyard practitioners are rare, but are known to exist (see wired article).
The first method is available to anyone, though I wouldn't recommend it - you tend to run into legal problems... something to do with human rights! :). This is the general class of "lesions and ablations". Though you can't really communicate via this method, people have been using it for centuries to work out which areas of the brain control which function of the body.
Basically, this method involves cutting a slice through axons (lesion) or cutting out a section of the brain (ablation) and watching the persons behaviour to see what they can no longer do. This is very damaging and very permanent and the results are often very ambiguous. For example, if a person now cannot recognise a person they once knew well - is it a problem with how they put the visual image together or did they lose a part of brain where the image of *this* person is stored?
One of the most famous applications of this method is the good old frontal lobotomy. This involved cutting the paths to the frontal lobe and was performed on people considered to be 'unmanagably insane'. Given that a lot of what we would consider personality is stored here, this operation usually turned the person into a walking vegetable. It may also be interesting to note that alchohol affects us by decreasing the activity of the frontal lobe, much like a temporary lobotomy. :) It has been often suggested that the effects of long-term alchoholism closely resemble the loss of personality and drive of the typical pre-frontal lobotomy patient.
However grotesque this method may have been in the past, be aware that it is still in use today, though in a slightly altered form. Whether or not there are secret laboratories of evil geniuses performing ablations on unwilling victims I don't know. I do know, however, that there are many strokes, brain cancers and accident victims every day of the year. Though it is unethical and illegal to be the cause of a brain lesion, you are quite allowed (with suitable permission) to study the effects on those who have suffered them through accident. Indeed there are many famous patients that have lent important information by happening to have an unusual difference of function after sustaining a head injury.
Some structures or cell types develop at a later age. By studying the abilities of a person as they grow, we can determine what these structures do. So too can we study people who have a natural problem in development. We can study their brain and see what structures are deficient or oversized.
An electrode usually consists of a very thin wire that can be precisely positioned in the brain at almost any depth or position. Either a small current can be applied to the very end or it can record the electrial potential present at that point. These have been the mainstay for brain research for many years now (before the development of the large brain scanners that I'll talk about later).
Electrodes have been used in innumerable experiments - usually involving animals, but there have been some with live human subjects. The benefit of electrodes is that they can directly stimulate a very precise area of neurons with a minimum of damage. Through vast amounts of selective stimulation, they have been used to fine-tune our knowledge of brain structures.
Most of the maps of brain activity have been formulated by people selectively test-stimulating each area of brain in turn and asking their subject (awake at the time) what they can sense, and observing any changes in behaviour. The main problem with using electrodes in humans is that the electrodes do cause some small amounts of nerve damage and thus it is unethical to use them unnecessarily.
Studies with humans are often conducted just prior to major brain surgery, when a patient has sustained damage to their brain (either cancer or a blood clot). Obviously, they will try to minimise the amount of brain tissue which has to be removed. The surgeon will try and determine which is live tissue and which is not. The patient has their scalp anaesthetised locally and then removed to bare the brain. The surgeon then inserts electrodes into the surrounding areas and runs a small current through them one at a time. The patient is awake and must tell the surgeon if they can tell that something is different when the current is on. This technique is really only useful in brain areas where an electrode can produce an obvious effect. Areas such as the visual cortex or any other major sense will do this as they will produce an obvious effect (eg a bright pinpoint of light in a specific area of the field of vision). If the damaged tissue is in an area that does not produce such an obvious effect (eg areas generally associated with memory storage) this technique is much less useful.
Electrodes are also used to passively measure the electrical potential of the neuron. This has been used to measure the effects of sub-threshold currents on a neuron and how currents can add or subtract from the electrical potential and how this effects the chances of an action potential. Electrodes can also be used to monitor whether or not a particular brain area is being used for a certain activity. For example, experiments have been performed on animals where they have had their heads fixed in place and had a tiny light shone onto a particular part of their field of vision. An electrode in their brain can determine which neurons are activated by this stimulus.
As a method of finely detailed research, electrodes can be very useful. As a method of stimulating a very specific neuron, they are the best. As a method of communicating with the brain, they are still not so good. We would have to be able to attach an electrode to a very large section of neruons. For example to every axon in a nerve bundle (like the 1000's of nerves in the optic nerve bundle)for this to be much use. Regardless of this, electrodes are the main possibility for the application of cyberware. It is the main avenue in which we are progressing, even though I consider us to be still in the barbarism stage (current methods requiring a severed nerve ending and a spiked plate covered in microelectrodes).
I believe that we must find a less invasive, but still very precise, method for reading the information coming from our nerves.
Really these should be labelled 'less-invasive' as all methods affect the brain in some way, but these are considered less damaging to a human and generally are performed outside the human body.
The problem with using these methods is that the equipment is usually prohibitively expensive for the average person. Research in this field is reduced to those who have access to the equipment and are willing to let them use it.
This device works by attaching several electrodes (not the pin-shaped ones described above, but ones that will lie flat against the skin) to the scalp. These measure the brain's electrical activity during thought processing. The output of the electrodes is amplified and recorded. The researcher can then analyse the data and can usually tell the overall state of the brain activity. A researcher can usually detect such things as whether the subject is asleep, dreaming, awake or whether they are problem solving or just daydreaming.
Abnormalities in the EEG can be detected when the subject has severe problems such as epilepsy or a tumour.
The EEG measures the average activity of a very large number of neurons under the electrode. Using a large number of electrodes can better pinpoint the location of the neural activity, but only to a certain extent. This is a large-scale, very generalised technique for measuring brain activity.
EEG's can be used to control the movements of a cursor onscreen. they have also been used for some very heavy experimentation on direct brain stimulation using the so-called montage amplifier (see article in wired).
MEG stands for Magnetoecephalograpyh and is very similar to the EEG. Instead of measuring the electrical potential, it measures the magnetic fields caused by the electrical potentials. Apparently, this method allows a more precise localisation of regional activity. [Coren et al, p645]
CAT stands for Computerised Axial Tomography. To take a CAT scan, the physician starts by injecting dye into the blood stream. The patients head is then inserted into a large xray machine. The machine takes an xray then rotates around the head 1 degree and takes another and so on until 180 xrays have been taken. A computer analyses the data recieved and creates a composite image of the brain.
This technique is useful for determining the brain structure without having to open up the skull. unfortunately, it is like taking a still picture. you get the idea of where everything is, but cannot see it in action.
PET is Positron Emission Tomography and it relies on the use of regional Cerebral Blood Flow to determine the active parts of the brain. The substance that is injected into the brain decays in a known manner, ejecting a positron at a statistically reliable rate. Glucose is often used as it will congregate in active neurons. When a positron hits an electron, the pair of particles annihilate one another, releasing some of their pent-up energy in the form of gamma rays. The gamma rays go in opposite directions simultaneously. A PET scanner is comprised of gamma ray detectors. When they detect two simultaneous gamma rays, they determine the position they originated from by finding where they intersect. When many of these are recorded, a computer can then plot the brain activity of that region.
MRI (Magnetic Resonance Imaging) is the same as NMR. Unlike a PET scan, it is capable of producing very detailed images of the brain without having to expose the brain to radioactivity of any sort.
MRI utilises a very interesting property of the atoms that make up our brain. Each atom has a certain rotation. Usually the axes of rotation are randomly oriented, but a strong magnetic field will align the axes in the same orientation. The atoms of hydrogen bound to the blood cells are usually pinpointed as they are easier to align. When a radio-frequency electromagnetic field is then applied to the aligned hydrogen atoms, they spin like tiny gyroscopes. When the field is turned off, they all relax into their previous positions, but simultaneously emit a very small amount of magnetic energy. When this energy is measured, we can deduce the concentration of hydrogen atoms in the region being monitored. This tells us which brain regions are active at present (see CBF for why).
MRI is slow, however, taking around 15 minutes per scan. A newer, faster form of MRI called Echo-planar MRI can form images in less than a tenth of a second. This is fast enough to watch blood flowing. This device will be able to greatly help us to study the structure of activity in the brain.
References for this article:
BioControl Systems Inc, Neural interface
technology-The future of Human Computer Interaction.
In: WebPages belonging to BioControl Systems Inc
Branwyn, Gareth The desire to be wired.
In Wired 1.04, October 1993
Coren S, Ward L, Enns J Sensation and Perception
[4th Ed],
Harcourt Brace college publishers, Fort Worth, 1994
Kalat, JW Biological psychology [5th Ed]
Brookes/Cole publishing Co., Pacific Grove California, USA, 1995
Lusted, HS and Knapp, RB Controlling
Computers with Neural Signals.
In Scientific American, October 1996
Tonneson, Cindy and Withrow, Gary BioSensors
(c) Copyright Taryn East 1996-1999 (All Rights Reserved).
Last Updated: 5th August 1999