Unlocking the locked-in

Locked-In Syndrome is about to steal the limelight again, with the release in the UK of the film of Jean-Dominique Bauby’s memoir, The Diving Bell and the Butterfly. The condition is rarely out of the news these days. Last November, for example, another locked-in patient made the headlines.

Eric Ramsey hasn’t spoken since he was in a car crash at the age of 19, eight years ago, but scientists think they have devised a way to give him back his voice.

Their approach is so close to the one I describe in my novel, The Quick (out in paperback in March), that reading Ramsey’s story made my spine tingle. Jonathan Brumberg of Boston University and colleagues have surgically implanted wireless electrodes just beneath the surface of this brain, in a region which is responsible for generating speech. The idea is that when he imagines speaking, the electrodes detect the pattern of activity of brain cells in this area, and turn it into synthesised speech that matches his thoughts.

After Bauby suffered his stroke in 1995, he was left with the ability to blink his left eye. A secretary sat by his bed, reading out an alphabet that had been rearranged according to the letters’ frequency of use, and Bauby would blink at the letter he wanted. By this painstaking process, they constructed a book. How technology has advanced in a decade.

But wait: Ramsey hasn’t spoken yet. In November Brumberg announced that he was confident his team could recognise the sound he was thinking of 80 per cent of the time. The next step was to let Ramsey hear for himself the sound their device generated. By giving him feedback in real time, they hope he will be able to tune his thoughts to produce a sound more closely resembling the one he wants. Once he has built up a big enough repertoire (back in November, he had only a few vowels), he’ll be ready to practise the art of conversation.

The concept of feedback is interesting. When a baby is born, most of its actions are reflex, but with time it learns to control and direct those actions. It learns to link cause and effect. To take an early example, if it sucks while its mouth is clamped on its mother’s breast, it is likely to receive milk. Thus the baby learns to suck when it is hungry. Feedback can be positive or negative, reinforcing an action or not, but without it there is no learning. The reason congenitally deaf people speak differently to hearing people is that they have never heard their own voices (or anyone else’s). They have no notion of what speech sounds like, only what it looks like, and they produce an approximation based mainly on that visual information.

Now, consider a locked-in patient who, by virtue of being completely paralysed, has received no feedback from the world since they entered that state. In the case of my character, we’re talking a decade. Is it conceivable that such a person might revert to the newborn condition, and forget how to control their own actions? In 2006, a group of German researchers led by psychologist Niels Birbaumer described the case of a 46-year-old German woman who might have done just that.

She had been diagnosed with the degenerative disease amyotrophic lateral sclerosis, and by the time Birbaumer saw her, she had been locked-in for a year. The researchers started to train her to use their own version of a brain-computer interface, which involved sticking electrodes to her scalp, but she was not able to shape her brain signals in the way the device required. It occurred to the researchers that she might have more success if the electrodes were implanted into her brain, closer to the source of the signals, but to perform that operation they needed her consent. Since she couldn’t give it, their hands were tied.

One of Birbaumer’s colleagues came to the rescue. Passing an electronics store one day, she spied a medical device for measuring the pH of saliva, and had an idea. They trained the patient to change the acidity of her spit by imagining either the taste of lemon or the taste of milk. She learned to push the pH one way to say “yes”, the other to say “no”, and using this simple method gave her consent to the operation. Three hours later, she lost the ability to change her salivary pH at will. The operation went ahead, but she was still unable to gain control over her brain activity.

Birbaumer fears that, after a certain time, locked-in patients may lose the capacity to control anything voluntarily. Bauby was spared this ultimate humiliation, as was my fictional locked-in patient, and both were therefore able to invite outsiders into their rich, and much altered, mental worlds. The early indications are that Ramsey might have been lucky too. Fingers crossed that he regains the power of speech before it is too late.