Friday, May 18, 2012

Learning to Be a Borg

If using speech recognition as a user interface is cool take a look at this....

There’s nothing like a bit of brain control to make people excited. The recent letter in Nature describing the use of a chip implanted into the brain of a tetraplegic patient was a clear breakthrough. Here was a woman who had lost control of her limbs 15 years previously because of a brain stem stroke. A chip had been implanted into the motor cortex of her brain five years ago, which sensed the firing activity of neurons associated with hand movements. These signals were decoded by a computer and used to control a robot arm. After training and quite a bit of practice, the woman was able to use the robot arm to pick up a cup of coffee, much to the delight of herself and the researchers.

It’s a testament to our amazing computer technology today that we can achieve such marvels, although the technology remains unreliable and the chips are prone to degradation and failure most of the time. But it’s also a testament to our amazing brains that this kind of computer interface can work at all. Out of necessity the patient’s brain was being examined with the precision of Mr Magoo on a bad day. The chip is almost entirely blind to the activity in the brain. So the only way this kind of brain-computer-interface can ever work is if the brain itself adapts (cortical plasticity). The patient must learn to think differently. She must learn to make those neurons shine brightly enough to be seen under the shortsighted nose of the Mr. Magoo chip, even if their original purpose was nothing to do with movement. This is only something possible with a great deal of patience and dedication — feasible in a patient who has no other movement available to them.

It is wonderful to hear of such research, but my concerns grow when I hear that researchers into BCI (brain computer interfaces) wish to go further than this. They want to bring the benefits of mind-control to everyone. No more tapping on keyboards or moving mice. Speech recognition? That’s so passé. Just think what you want on the screen, and the computer reads your mind! There are already devices on the market that use non-invasive measurement of your faint brainwaves (EEG) to enable you to type letters or control a mouse pointer, (for example, the recent intendiX-SPELLER).

These ideas of BCI for able-bodied individuals remind me of the visions of Douglas Engelbart, the 1960s computer pioneer. Engelbart’s dream was — and still is — to “augment human intellect” through the use of appropriately designed human-computer interfaces and software. He realized that the computer keyboard was a clunky and inefficient way for us to perform all our interactions with computers. We needed better ways to express ourselves and expand our imaginations with these new digital machines. So through methodological testing, he invented and demonstrated many of the interfaces we now take for granted: the mouse, windows, hyperlinks, drop-down menus, video conferencing, easy to use word processors. These innovations were subsequently inherited and improved by Xerox Parc, and taken and improved further by Apple.

But Engelbart understood something that seems to have been forgotten by some BCI researchers. The human brain is a remarkable computing machine. It already has fabulously rich interfaces via nerves to muscles, bone and sinew, and it receives input from countless sensors. If these are still in working order, why would we forgo them and try to use direct mind control? Why would we train ourselves to use our neurons in bizarre ways rather than to control fully functioning limbs? The mind-control technology is so cumbersome and clunky — and will always remain so without horrifyingly invasive implants — that it’s like asking us to type by jumping on a pogo stick over letters painted on the ground. With enough practice we might manage a sentence, but it is not rather a waste of effort compared to using a keyboard?

We must applaud and continue to encourage all work that aims to restore function and improve quality of life for those patients who so desperately need it. BCI research is a lifeline for them, and every effort should be made to improve the technology for their benefit. But for those of us lucky enough to have full control of our limbs and senses — let’s use them. Let’s develop human-computer interfaces that exploit our inbuilt dexterities, our need to stay active, our liking of both stimulation and calm. Let’s have intuitive interfaces that learn and adapt to our needs, rather than the other way around. Let’s not wear silly hats that try to read brainwaves, and degenerate into frustrated computer-potatoes, capable of no more movement than a new curse at the inability of the machine to read our thoughts correctly.

 
FOLLOW SCIENCE
    google.com by Peter Bentley
    via http://www.speechtechnologygroup.com/speech-blog - If using speech recognition as a user interface is cool take a look at this.... There’s nothing like a bit of brain control to make people excited. The recent letter in Nature describing the use of a chip implanted into the brain of a tetraplegic patient was a clear breakthrough. Here was a woman wh ...

    No comments:

    Post a Comment