Re-Wiring the Brain: The Science of Nexus

My science fiction novel Nexus is out today. It's a story about the struggle over a technology that can get information in and out of the human brain – senses, thoughts, emotion, all sorts of information. (Read an excerpt here.) That's science fiction, right?

Yes and no. The technology I describe – a ‘drug' that's really a collection of nanodevices that attach to the neurons in your brain – is definitely fictional. But it's also based on real science that's been moving ahead faster than most of us have noticed – science that is already getting sight, sound, and touch in and out of the brain, and that's starting to go beyond that and into the realms of memory and even some types of intelligence.

The Brain is Electro-Chemical

We've known that the brain is electrical since 1870. That was the year that two radical German scientists, Fritsch and Hitzig, proposed an experiment to electrically stimulate part of the brain of an anesthetized dog. The University of Berlin, horrified at this idea, wouldn't allow the use of its facilities. So the pair used the dining room table of Fritsch's home. There they demonstrated that a very low current applied to part of the brain would cause the dog to reliably move one of its limbs. The brain was electrical.

Decades later, in 1930, a brilliant surgeon named Wilder Penfield took this a step farther. Penfield operated on epileptic patients, carefully cutting out small parts of the brain that caused their seizures. To find the responsible areas, and to minimize the damage as much as possible, Penfield took to using a small electrical current to map the different brain regions of patients as he operated on them. What he found was that stimulus to the right areas would cause patients to have vivid memories, or suddenly start speaking, or suddenly experience a sight or smell or sound. It wasn't just motion that was electrically driven. It was sensation. It was speech. It was, to use a loaded word, consciousness.


Those discoveries would lead to medical advances decades later. In the 1970s, Dr. William House introduced the cochlear implant. The cochlear implant looks like a hearing aid, and you'd be forgiven for thinking that it is one. But the way it works is profoundly different, and says something deep about our ability to interface with the brain.

We hear when vibrations in the air stimulate the hair cells of our inner ear. That stimulation, in turn, results in an electrical signal being sent along the 30,000 or so nerves of the auditory nerve bundle and into the auditory cortex of the brain. People with some hearing damage may have lost some of those hair cells. Normal hearing aids work by picking up sounds via a microphone, cleaning up and enhancing that sound, then playing the cleaned up version – potentially at a higher volume – into the ear. The remaining hair cells of the inner ear then pick up the sound and transmit it to the brain.

But what if you have absolutely no hair cells at all? For millions of people, that's the case. No hearing aid can help them, because no hair cells remain to pick up any vibrations in the air. That's where the cochlear implant comes in. Instead of cleaning up and replaying sounds, it uses electrodes to send an electrical signal straight into the auditory nerve bundle and from there into the brain. And while a typical cochlear implant has only 22 electrodes – less than 1/1000th the number of nerves in the auditory nerve bundle – it produces hearing that's good enough to hold conversations. Worldwide, more than 200,000 people already have cochlear implants sending data into their brains. And its impact is absolutely life changing.

Want to see? Here's a video of an 8 month old baby hearing for the very first time.

The first cyborgs are among us.


Progress on sending data into the brain didn't stop there. It's extended into ways to get visual data into a human mind.

In 2002, a man named Jens Naumann, who'd lost both eyes in a pair of unrelated accidents nearly 20 years earlier, had his vision restored. The system that restored it, designed by an eccentric scientist named William Dobelle, used a CCD camera worn on Jens' glasses to capture video. That video was sent to a small computer Jens wore, which translated the imagery into a series of electrical impulses that were sent, through a jack in the back of Jens' skull, Matrix style, to his primary visual cortex. And with those electrical impulses, carefully arranged in a pattern that matches his visual cortex, Jens can see.

Think about that. We can, today, take digital images and send them directly into a human brain.

More recently, physicians have switched to an approach more similar to the cochlear implant. Instead of opening up the skull to put an implant in the visual cortex of the brain, most research now uses a retinal implant – a chip placed at the back of the eye – to send data along the visual nerve and into the brain. And those implants, now tested on dozens of patients, are moving along the path towards full clinical approval. A decade from now we may have tens of thousands or even hundreds of thousands of formerly blind men and women wearing retinal implants, essentially the ‘bionic eyes' of fiction.


We've also gotten data out of the human brain. In 1997 a Georgia drywall contractor, Vietnam veteran, and occasional blues guitarist named Johnny Ray suffered a massive stroke in his brainstem. One minute he was on the phone, having a perfectly normal conversation. Then nothing. When he awoke, he was in the Veterans Affairs Medical Center in Atlanta. He was paralyzed from the neck down. An emergency tracheotomy that had saved his life had also taken away his voice. He could think, but his interaction with the outside world had shrunk dramatically. The only way he could communicate at all was to blink his eyes – once for no, twice for yes.

There are around a quarter million quadriplegics in the United States. Worldwide there are around half a million "locked in" patients like Johnny Ray who are both paralyzed from the neck down and unable to speak. For all of those, there may be hope.

That hope started with a neuroscientist named Phillip Kennedy. Kennedy, working in monkeys, had shown that he could implant electrodes in the motor cortex of a monkey brain – the part that controls motion – and teach them to use that electrode to control a computer cursor or a robot arm.

That's exactly what he did with Johnny Ray. With the help of a surgeon, Kennedy placed a single, wireless electrode in the part of Jonny Ray's motor cortex that controlled his right hand. Then, over painstaking months, they trained Johnny Ray to use that electrode to control a computer cursor, and use that to type out messages to his doctors, family, and friends. He went from being able to communicate only by blinking to being able to type out whole messages.

Today those systems are even better. Using 32 or 64 electrodes and more advanced computer algorithms, they can be implanted, calibrated, and working well in minutes rather than the months it took Johnny Ray. And now those systems, too, are moving through human trials. Recently io9 posted this video of a patient using a more recent version of this technology to control a robot arm.

Making Us Smarter?

We've also gone beyond sensory data – at least in animals.

Two provocative studies have shown that we can improve memory and decision making in other mammals. In 2011, researchers working on ways to repair damage to the hippocampus – part of the human brain that's critical for forming new memories – demonstrated that their artificial ‘hippocampus chip' could actually improve memory in rats.

And this year, in 2012, a team at Wake Forest University went further. They trained rhesus monkeys on a task that was, in a crude sense, a monkey IQ test. As the monkeys learned, a brain implant in their frontal cortex – the part of the brain involved in decision making and attention – watched how the monkey's brains worked and learned those patterns. Then the researchers impaired the monkeys' performance on those tests by giving them doses of cocaine. What the researchers found was that, as they hoped, turning the implant on could undo the temporary damage done by the cocaine. But more than that, it could improve the monkeys' performance on the test, beyond their baseline scores.

Planet of the cyborg apes, here we come.

Data In, Data Out

The frontier I'm most excited about is communication.

I've described two systems that get data into the brain – sound and sight – and one that gets data out – in motor control. But in both of these areas we can go in either direction. In fact, there's a general rule in neuroscience that, for the most part, sensation, imagination, and recollection are the same in the brain. Roughly the same neurons light up when you see a red triangle, when you remember a red triangle you've seen before, or when you imagine a red triangle anew. What that means is that if we can get data in by stimulating a certain set of neurons, we should be able to get data out by observing those same neurons, and vice versa.

And we can see some of this happening already.

Last year a group at Berkeley showed that they could use an fMRI brain scanner to reconstruct what a person was seeing.

Also last year, a group of scientists at Duke University demonstrated that they could send touch data into a monkey brain. They equipped the monkey with a two-way implant that allowed it to control a virtual robot arm (in a video game) and feel something when the robot arm touched objects.

And fascinatingly, a team at Washington University has shown that they can tap into the speech centers of the brain. They used an implant in those centers to allow patients to control a computer cursor, but their real goal is to allow paralyzed patients to directly ‘think' words out of their brains and into computers. Which of course opens up the possibility of thinking those words directly into the brain of another person, equipped with a similar implant. Digital telepathy.

All of this fits with how we use technology. Once upon a time, ‘computers' were things we used to do solitary work. Now they and their offspring – smart phones, tablets, and the like – are primarily about communication. We use our digital gadgets to access information and to interact with each other. It's inevitable that brain-hacking technology will follow the same path. We have a deep innate desire to connect with others. And what could be more intimate than actually sending signals directly between our brains?

Huge Challenges

All of this said, there are incredibly huge challenges to realizing the full potential of this technology in humans.

First, the systems are extremely crude. Jens Naumann, the man who's vision was restored, achieved 256 pixel vision. He could see in a grid 16 pixels by 16 pixels. It's utterly amazing that he could see at all, given that his biological eyes were both destroyed. But it's not the super-enhanced vision of the bionic man. It's not something we'd voluntarily switch to. The same is true of the cochlear implant. Yes, these deaf people can hear, but not as well as a normal hearing person. Cochlear implants are just now getting to the point where music appreciation may be possible. And they still don't produce word recognition as well as most hearing people have.

Second, brain computer interfaces today generally require brain surgery. Yes, there are some systems like the fMRI brain scanner or EEG based devices that don't need to be implanted inside the brain. But those are the minority, and they're fairly limited in how accurate they can become. If you want a really high fidelity system, it's probably going to have to be placed within your skull. And how many of us are going to be willing to go through with that? To motivate brain surgery, an implant has to bring huge advantages.

Both of those problems will eventually be overcome. We'll develop implants that can tap into more neurons, allowing more data – and higher fidelity – to be sent back and forth. We'll develop algorithms that can decode that data in real-time. And we'll develop safer and easier ways to get these systems into the brain.

Rodolfo Llinas, one of the elder statesmen of neuroscience, has proposed one of the most intriguing ways to get a system into the brain. Instead of cutting open the skull, he proposes, we could insert a bundle of carbon nanotubes in through the blood stream, and allow them to branch throughout the brain. Even a bundle a million carbon nanotubes thick would take up only about 1% of the area of the smallest capillary in the brain, and it would allow us to electrically listen in on and stimulate neurons in the brain through the capillary wall, without the need for surgery or the risk of infection.

Researchers seem to be moving forward on technology that could make Llinas's dream true. In 2012, a team demonstrated a carbon microthread that could listen in on or stimulate a single neuron, and which is 1/100th the width of current electrodes used in brain implants. It's still many times larger than a single carbon nanotube, but it's a huge step in that direction. And even when used in conventional implants that are placed inside the brian, it will enable researchers to tap into many more neurons, with less impact on the brain, allowing them to send more data in and out, and moving brain implants one step closer to reality for the blind, deaf, paralyzed, and one day perhaps for all of us.

Side Effects

Of course, the prospect of implants in our brains leads to all sorts of questions. What happens if they crash? Can they get viruses? Can we hack on our own implants? Can someone else hack on them?

How will society deal with these sorts of technologies? Will their complexity keep them expensive and only available to the rich? Or will the same economics that have governed other electronics come into play, and drive their prices down to the point where all of us can afford them?

And the biggest questions, I think: How will society react to this technology? Will we embrace it or ban it? Will control rest in the hands of individuals? Or will this be a tool for control? Would a future full of people with wired brains look more like the crazy, chaotic, diverse, bottoms-up nature of the internet? Or would it look more like Big Brother from Orwell's 1984?

Those are the questions I try to tackle in Nexus. If you're intrigued, I hope you'll check it out.

And if you want to know more about the science behind all this, you can read my H.G. Wells Award-winning non-fiction book, More Than Human: Embracing the Promise of Biological Enhancement. In fact, if you buy Nexus before the end of the year, I'll send you a free e-copy of More Than Human.