The Fringes of Neurotechnology
“I think it’s time for some intracranial penetration.”
—WALTER BISHOP, “THE GHOST NETWORK” (1-3)
“The brain is a computer . . . It’s an organic computer; it can be hijacked like any other.”
—WALTER BISHOP, “OF HUMAN ACTION” (2-7)
Fans of Fringe won’t be surprised to hear that neurotechnology, an increasingly popular research field at the intersection of technology and the brain, has also become quite a popular topic in today’s sci-fi. The mysteries of the human brain and its relation to technology have provided some of the show’s most memorable moments, from the retrieval of Walter’s surgically stolen memories to Peter, Walter, and William Bell’s journey through Olivia’s brain. However, while we can all recognize when something involves the brain and technology, the term “neurotechnology” can actually prove pretty tricky to define.
There are many different definitions out there, but I’ll use the definition from the University of Frieburg, a well-established research group. Their definition divides neurotechnology into two types of technologies:
(I) technical and computational tools that measure and analyze chemical and electrical signals in the nervous system, be it the brain or nerves in the limbs. These may be used to identify the properties of nervous activity, understand how the brain works, diagnose pathological conditions, or control external devices (neuroprostheses, ‘‘brain machine interfaces’’).
(II) technical tools to interact with the nervous system to change its activity, for example to restore sensory input such as with cochlear implants to restore hearing or deep brain stimulation to stop tremor and treat other pathological conditions.1
In short, part (I) refers to devices that read from the brain (such as an EEG, MEG, or MRI), and part (II) describes devices that write to the brain (such as a cochlear implant or deep brain stimulator).
The area I work in, and in which Walter Bishop sometimes dabbles, is a specialized kind of neurotechnology called brain-computer interfaces, or BCIs. A BCI is explicitly defined as a device with four characteristics: it must (1) rely on direct measures of brain activity, (2) work in real time, (3) provide feedback to the user, and (4) rely on voluntary, intentional control.2 (Therefore, devices that write to the brain or rely on involuntary or unintentional signals, though they are still neurotechnology, are not BCIs.)
Neither BCIs nor other types of brain reading and writing devices are especially new. For example, deep brain implants have been used for decades and have helped many people with Parkinson’s disease and other movement disorders, chronic pain, and other conditions.
Stories about neurotechnology are not new either. They have been popular with sci-fi fans for decades; Neuromancer and the cyberpunk subgenre are grounded in the future of BCIs, and all five Star Trek series addressed neurotechnology many times. Technology used for mind reading and mind melding show up in The Matrix, Strange Days, Total Recall, Brainstorm, and X-Men, and mind control facilitated by technological innovation has been a staple of sci-fi for a long time. In fact, the fascination with brain-computer interactions is so prevalent in sci-fi that you could even name a subgenre after it: bci-fi. What is new–and sudden–is the way neurotechnology has seized popular attention, not just among the stereotypical bespectacled lonely males who can recite Monty Python and Maxwell’s equations in hexadecimal code, but through more mainstream shows like Fringe.
Fringe is well named. On the one hand, I doubt I’m surprising or offending many fans by noting that many of the show’s plots and devices are quite far from reality. On the other hand, like most good fiction, Fringe is obviously inspired by reality. Often, I’ve recognized similarities to real-world research studies that may have gotten the writers’ neurons firing. But the show usually adds some new twist involving neurotechnology or the people who use it that leads to a bizarre, violent, and harrowing inciting incident. Our task here will be to separate the truth from the fiction, the better to highlight Fringe’s creativity.
A Fringe Case Study
Let’s look briefly at the season-two episode “Of Human Action,” where a troubled teenager named Tyler developed the ability to make others do anything he wanted, just by popping some pills from Massive Dynamic’s version of a PEZ dispenser. These powerful drugs were part of a neurotechnological experiment: they were developed to amplify the brain waves of military pilots so that they could use a new hands-free navigation system to steer planes using only their thoughts (as picked up via electrodes in a helmet). Tyler, however, used the pills to fake his own kidnapping, hypnotizing random adults to commit criminal and even suicidal actions.
This might seem like pure fantasy–the sort of stuff you’d see in X-Men movies or The Lawnmower Man–but the ideas are actually based on some real-life developments. Which of these claims are true?
A) The air force has developed a noninvasive BCI that uses sensors placed in helmets to let pilots control aircraft.
B) Drugs can enhance some of the brain’s EEG patterns and make them more recognizable to a mind-reading device.
C) Hypnosis can change the alpha waves of the brain.
The answer? All of them–with some explanation. As far as option A is concerned, the USAF developed such a BCI over ten years ago.3 However, pilots could only use it to bank aircraft, not perform more complex tasks as implied in the Fringe episode. The device was noninvasive, and several groups funded by the US Military and others have worked on brain-wave sensors that could be placed in helmets or elsewhere.4 Those of us in BCI research are still very actively researching practical electrodes–by which I mean electrodes that do not require electrode gel, as conventional electrodes do. (Conventional electrodes require rubbing the scalp under each electrode and squirting some electrode gel to get a good connection. The process is safe and painless, but wastes time and tends to be messy.) Companies like Neurosky, Emotiv, g.tec, TMSi, and even Mattel and Hasbro have begun selling BCIs with practical electrodes, primarily to play simple games. Remarkably, Dr. Thomas Sullivan from Neurosky told me they have sold over a million of their BCI chips, and primarily in systems meant for the general public–a much wider distribution than just a few years ago.
You can gather evidence of option B just by drinking your morning coffee (at least if you happen to have an EEG recording device handy). When you give someone caffeine, or any stimulant, his EEG patterns might become more recognizable to a recording device–though so far, caffeine does not seem to affect BCI performance.5 Whether other drugs can enhance BCI performance, we don’t know yet, but it is of interest to us–check again in five years, and I suspect we’ll have more solid results.
Option C, which Walter commented on only in passing, is also true–in fact, it dates back to what would have been Walter’s early research days. We’ve known for decades that hypnosis can change alpha waves, as well as theta and some other brain-wave patterns.6
So far, everything we’ve explored in “Of Human Action” was based on real events. Of course, then came the Fringe twist. The pills didn’t just make Tyler’s brain waves easier to pick up; he could use them to control other people’s minds. Fortunately, this is quite impossible for many reasons (discussed in the mind control section below). Another bit of fiction? Dr. Bishop commented that resisting mind control can leave lesions on the surface of the brain. Though this helped make for engaging drama, it had no basis in reality.
(As a side note, about eight minutes into this episode, Dr. Bishop mentioned a neurobiology conference he once attended in Berlin. There actually was a large international BCI conference in Berlin in 2009, and one of the speakers was Dr. Niels Birbaumer, one of the giants of our BCI research community. His talk is available online at http://videolectures.net/bbci09_birbaumer_bip/. You can judge for yourself whether there are any similarities between the Doctors B.)
So the writers take a seemingly innocuous or even actively benevolent scientific advancement, and then imagine what would happen if something went wrong–if neurons didn’t have checks and balances, if sleep therapy induced instead of prevented nightmares, if a hands-free aircraft navigation became hands-free people navigation. This MO–a little bit of reality and a lot of fiction–makes for compelling stories. And knowing exactly which is which lets us appreciate what the Fringe writers do even more. So let’s look a little deeper at neurotechnology, both as it exists in the real world and as it’s reflected in Fringe.
Current brain computer technology is often compared to mind reading: we have many ways to find out what’s happening in the brain, from MRIs to EEGs to X-rays. But even the most advanced BCIs and other neurotechnology are not literal mind-reading devices. That is, you cannot simply think any word, image, or tune and have it appear on a monitor. We’ve tried. We keep trying. We are still not close.
“Downloading” your experiences onto a computer or someone else’s brain is a popular theme in Fringe. It was introduced in the very first episode with the prospect of “interrogating” John Scott’s dead brain (which was apparently feasible because he had been dead for five hours, not six–that six-hour limit doesn’t really occupy real BCI scientists much). In “Grey Matters” (2-10), an episode whose title warns you to expect some serious breaches of the Hippocratic oath, pieces of Bishop’s brain were removed and put into other people’s brains (though a brain reading a brain is arguably not neurotechnology, since the brain is not a device). The poor victims went crazy downloading just a smidgen of Bishop (perhaps because even that tiny amount had enough stored drugs to make three victims trip for fourteen years). At the end of “Os” (3-16), William Bell’s consciousness was “downloaded” into Olivia’s body after his death using “soul magnets.”
In other examples of using technology to read human thoughts and memories, Dr. Bishop “read” the last image a dead woman saw by scanning her optic nerve (which, he said, would have been absurd but for the fact that she had relaxants in her system that “froze” the image) in “The Same Old Story” (1-2); Dr. Nayak could experience other people’s dreams by eavesdropping with an evil BCI in “Dream Logic” (2-5); and in “Unearthed” (2-11), Dr. Bishop noticed that a murder victim, Andrew Rusk, had taken over a teenage girl by looking at what seemed to be only one very noisy channel of EEG data. These are exciting, haunting ideas, and so it is understandable that they led to good stories. But they aren’t possible.
Here is about where we are today: BCIs work by detecting certain brain activity patterns associated with voluntary mental activities. Unfortunately, most of the things people think, feel, want, or perceive do not produce patterns of brain activity that we can discriminate. Different thoughts, feelings, desires, and perceptions do reflect different brain states, but we lack the neuroimaging and signal processing technology to distinguish nearly all of them. Therefore, BCI research has had to proceed so far based on the relatively few distinct brain states that we can identify. The most common BCI approaches rely on imagined movement or visual attention, not because these are the most natural or obvious activities to use for communication or control, but because brain states associated with some movement and visual attention tasks are the ones we know how to recognize, at least under very specific conditions.7
There are certainly other intentions that BCIs can read. For example, the first BCI paper published in Nature relied on a different approach called Slow Cortical Potential, or SCP. SCP works by reading different emotional or cognitive tasks the user can learn to perform.8 Other BCIs rely on conventional tasks like rotating an object, imagined singing, or math.9 But most BCIs today are based on motor imagery and visual attention.
Imagined Movement BCIs
BCIs are often based on some type of imagined movement, but they can only recognize a few different signals. For example, if you think about moving your left hand, right hand, or feet, you get reliably different activity over somatomotor areas of the brain responsible for movement and touch (specifically, the areas around the central sulcus, which is located on the top of the brain and divides the frontal and parietal lobes). Imagining (or performing) left-hand movement will show up in the EEG as a reduction in “power” somewhere between 8 to 12 Hz over these areas on the right side of the brain.10 (For reasons we only partly understand,11 many different mental activities are correlated with 8 to 12 Hz power changes. For example, relaxing or closing your eyes, a common way to control simple toy BCIs, can cause an increase of 8 to 12 Hz power.) Right-hand movement produces changes over the left side, and foot movement produces activity that is most distinct over the center.
Hence, you could–by reading a subject’s intended hand and foot movement–allow them to navigate a virtual environment, spell, or perform other tasks just by thinking.12 A recent paper described a landmark achievement: people who could move a cursor in all three dimensions (left/right, up/down, and forward/backward) with a noninvasive BCI.13 However, this feat required dozens of hours of training, and even with training (pending further research) seems to be possible only for a small percentage of people.
Visual Attention BCIs
We can also tell if you are “paying attention” to–i.e., looking at–one item out of several items, at least if the items flicker or flash in a specific pattern (say, if the items are LEDs or images on a monitor). Specific patterns are necessary because each pattern elicits a different pattern in brain activity, and hence each pattern creates a unique signal that a BCI can distinguish.
There are two types of BCIs based on this kind of visual attention. They are named after the type of visual stimulation used, which results in different EEG patterns. The first, the P300 BCI, relies on flashing stimuli, such as a matrix of letters and numbers that each flash one at a time or in rows and columns.14 The second, the SSVEP BCI, relies on oscillating stimuli, such as characters that each oscillate at a different frequency.15
There are certain limitations that restrict what’s possible with all mind-reading BCIs.
First, you need a device on or in your brain. Most BCIs rely on noninvasive sensors–EEG electrodes outside the scalp. However, electrodes can also be implanted under the scalp, on the surface of the brain, or even inside the brain. Both types are referenced in the episode with the aforementioned evil doctor, Nayak: he seemed to wear an EEG-based system but also described implanting the electrodes in the thalamus. The system worn in “The Equation” also looks much like a jury-rigged field EEG system, and the lines on the monitor look very much like real EEG patterns. Before Agent Dunham went into the sensory deprivation tank in season one, she got an implant near her brain stem, though reading complex, specific actions would probably require electrodes elsewhere. (It is unclear why the brain stem is such a popular spot for futuristic neurotechnology implants–it seems to be in about the same place as the evil BCIs in The Matrix, Space: 1999, X-Men II, Johnny Mnemonic, and others.)
Second, you usually get a vocabulary of only a few signals, and you have to tell the user what to do to produce them. You can’t just hook up a person (or corpse, like John Scott in several first-season episodes) and expect the BCI to read everything, or much of anything.
Third, an involuntary BCI is not really possible. Even if the user is somehow prevented from removing the electrode cap you’ve attached to his or her head, he or she could just choose not to think about moving or not pay attention to visual stimuli. In other words, it’s no use on dead people–or sleeping ones. You cannot really “sneak” a brain imaging device on someone, although you could of course trick them into using one and then use it to read more than intended. For example, it’s conceivable that a person might use a BCI to play a game, but an unscrupulous eavesdropper could determine his sleep cycle (which doesn’t necessarily require a BCI; for some gamers, you need only look at when they aren’t playing).
Moreover, even with the full cooperation of the user, you wouldn’t really be able to determine anything useful to Fringe’s stories, only that the user was alert or performing one of a few simple tasks. Right now, you can only pick up one signal at a time, though this could change with “hybrid” BCIs, which are one of my main research interests.16 Under ideal circumstances, typical modern BCIs only allow their users to convey about twenty bits (or a few letters) per minute. Rich, detailed information that requires a high bandwidth is not imminent. For the moment, even the most active people in BCI research still write articles primarily with keyboards, not BCIs, and I doubt we’ll change that in the next ten years, at minimum.
Is a Mind-Reading BCI Even Possible?
Will we ever develop technology that can read anything and everything people experience? Definitely not in our lifetimes, and it may not be possible with any technology;17 this has been a philosophical debate for thousands of years. One of the best-known inquiries comes from Descartes, who mused about whether our experiences are real, or whether we might just be fooled by an evil demon who makes us think the outside world really exists. Descartes concluded that we cannot be sure of anything except our own thoughts–hence his famous phrase “I think, therefore I am.” Neurophilosophers and others still grapple with the discussion of whether we can really trust reality. Unless our very perception of what is possible is also fed to us (through something like the BCI in The Matrix), we can very safely say that reading a complete reality is impossible. (So is writing one, but we’ll get to that in a minute.) On the other hand, this is not really necessary for developing helpful technology.
If a mind-reading revolution does occur, there is some work that might foreshadow possible mechanisms. In a recent study, subjects looked at the word “neuron” while lying in an fMRI scanner. If the subject stared for many seconds, the device could generally determine that they were observing the word “neuron” by decoding the activity in visual areas.18 But an fMRI requires a huge superconducting magnet: the machine costs millions, and scanner time typically costs around $300 per hour. And its “mind reading” was not a real-time system. Moreover, the term “mind reading” seems rather grandiose when describing a system that can only detect one word, which is already known to the experimenters. Other work19 asked subjects to listen to one of seven three-second segments of music. Based on the EEG, they could tell which of these seven pieces a listener was hearing with up to 70 percent accuracy in real time, and the article noted that it could be increased to 100 percent accuracy. Another very recent paper showed that an invasive BCI could distinguish between pairs of sounds heard by the user.
Perhaps the most promising recent story about where we are now comes from the work of old friends of mine in Albany. Gerwin Schalk, Theresa Vaughan, Jon Wolpaw, and others described new BCI systems, including research into an invasive BCI to distinguish between thirty-five different imagined words.20 This is remarkable progress, but a long way from a “universal” BCI that can decode any thought.
Mind-Writing and Mind Control
We’ve established that mind reading, where possible, is a much more limited process than the one we see in Fringe. But there’s another aspect to neurotechnology, and that’s mind-writing–or, as often seen in bci-fi, mind “control.”
In science fiction, mind-writing varies in sophistication from simply tweaking emotions or sleep cycles to completely replicating someone else’s experience and/or dominating every action. It can be therapeutic or insidious, healing or controlling, voluntary or involuntary. The only real common thread is that it usually looks pretty effortless on the part of the controller. We see this in Fringe with Tyler, but also with Bell’s control over Olivia’s body, Dr. Nayak’s dream manipulation, Rusk’s control of the teenage girl in “Unearthed,” and others. This is very common elsewhere in science fiction, as well. Borg technology in Star Trek, for instance, relies in invasive BCIs that read and write to the brain.
However, tweaking brain activity in ways that produce specific and complex perceptions, emotions, experiences, or actions is not nearly that simple. And mind control like Tyler’s, especially, is way, way beyond modern technology. But why?
Even beyond the question of how information was getting from one brain to another, there are limits on how specifically we can control neural activity. There are many ways to do this, ranging from basically turning neurons off or making them fire haphazardly to making an established specific network of neurons fire in a totally novel way. If you want to simply prevent an area of the mind from working, you have lots of options. Simply cooling a region of the brain can shut down function, at least for a while, without any serious or permanent effects. Damaging brain tissue would also prevent it from working, but that’s of course more permanent. To cause specific behaviors (such as shooting other policemen) you need more than simple deactivation. You need to make the brain do something different. To accomplish this–to actually alter brain activity–you need to make neurons fire in specific patterns.
Neurons fire on their own every day; every conscious act you perform (and many other functions you’re totally unaware of) is triggered by the firing of neurons. Just making neurons fire, or making them fire more, is also relatively easy. In “The No-Brainer” (1-12), a computer virus made home PCs play a movie clip designed to kill people by liquefying their brains. Dr. Bishop explained that it involved “a complex combination of visual and subsonic aural stimuli ingeniously designed to amplify the electrical impulses of the brain, trapping it in an endless loop.” The “electrical impulses” Walter was referring to are the signals between neurons, and they can in fact be made to fire more than they otherwise would, even from the outside of the brain. Television shows have produced seizures in the past, through a mechanism somewhat like the phenomenon Bishop describes–excessive neural activity.
Luckily, though, the “endless loop” from that episode was just a cute metaphor. American culture loves the notion that bad programming can rot your brain–and maybe it can. Not literally and instantly, though. The brain has numerous mechanisms to prevent this kind of wildfire explosion of neural activity. Many neurons have an inhibitory feedback system that releases inhibitory neurotransmitters if the neurons themselves, or other neurons around them, become too active. This helps to prevent seizures and other problems. (Also, even if a seizure is induced, even serious seizures are not nearly as bad as complete brain liquification.)
Brain meltdown still won’t trigger any specific actions, however, such as shooting a gun or speaking a sentence. Simply overstimulating the brain’s neurons could lead to seizure, brain damage, or death. While these might be appealing goals to a mad scientist or a soulless executive, if you want a specific reaction, you need to send a specific signal, either via electrical, magnetic, or chemical means.
There are multiple ways to send such signals, but the most direct method is something invasive, such as implants or neurostimulators, and invasive stimulation methods are prominent in Fringe. In the real world, deep brain stimulation (DBS), for example, stimulates the brain through electrical means and is a common and effective technique to reduce symptoms of Parkinson’s disease. Doctors implant a neurostimulator into the basal ganglia, a region heavily involved in movement and production of dopamine, a neurotransmitter that is depleted in Parkinson’s disease. A deep brain stimulator can stimulate neurons in some basal ganglia regions like the globus pallidus and subthalamic nucleus, which does not cure the disease but, by encouraging the production of additional dopamine, can reduce symptoms and dependence on medication.
In one famous paper in Nature, researchers used invasive electric stimulators to “control” rats’ behavior.21 They implanted electrodes into three regions of a rat’s brain: the right and left primary somatosensory areas, which are responsible for the sensation of touch on the left and right whiskers, and the medial forebrain bundle (MFB), which is critical in reinforcement learning. The researchers stimulated the left or right whisker areas, then (if the rat turned left or right) provided a “reward” by stimulating the MFB. The rats quickly learned to turn left or right and would also move forward if you stimulated the MFB alone. After some training, they could be guided through a series of tasks and even moved through bright open areas, which rats typically avoid. Humans have more or less the same areas; our “whisker cortex” is just the part of the primary somatosensory cortex that responds to touching the face. But the effects are limited to just a few simple behaviors and must be developed over time. The control isn’t instantaneous.
Invasive chemical means have also been employed in animal research to change the brain and affect behavior. By putting a device called a cannula–essentially a tube–into the brain and sending chemicals to specific brain regions, researchers can directly affect neuron behavior. This technique can be very effective for studying precisely how neurotransmitters work or how different chemicals affect different regions, but using it to control specific actions would be quite difficult.
Not every mind-writing strategy requires invasive means. There are ways to directly influence the chemical composition of the brain noninvasively as well, such as giving people drugs. This technique is, of course, Dr. Bishop’s favorite, especially when it comes to affecting his own brain chemistry. However, he is far from the first person to explore the fringes of drug effects. And if people could control others’ minds or perform similarly fantastic feats by taking lots of LSD, then some shivering old hippie on Haight would rule the world by now.
Pyschoactive drugs were put to a more insidious use in “The Dreamscape” (1-9). Victims were tricked by a hallucinogen produced by frogs into becoming so scared, or so convinced of some horrific fate, that they actually produced physical symptoms, such as trauma from hallucinated thugs or cuts from a thousand butterflies. This seems to be based on a couple real stories. Many people have heard examples of spontaneous injury, such as stigmata, or even spontaneous combustion. “Voodoo death,” in which people become so afraid that they die, is nothing new. Similarly, there has often been enthusiasm for hypnosis, drugs, or other techniques that can help people instantly heal themselves or produce other dramatic physical changes. But these examples are fairly far removed from people suddenly developing massive and fatal physical trauma because they think they are being attacked by ghost thugs or evil mechanical butterflies. Drugs are also notoriously nonspecific, since they travel all over the brain, in different concentrations, and have different effects on different areas.
Electroshock treatment, a very old technique with some benefits for persons with mental disorders, is a fairly dramatic noninvasive way to alter the brain, but the effects are too broad to be truly useful. Newer (and rapidly developing) methods like transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (TCDS) can also make neurons fire, or inhibit them, or change how they fire to influence behavior. Both of these approaches are noninvasive and involve electromagnetic fields outside of the head. These approaches have gained a lot of attention recently within the neuroscience community, producing some promising results for treating depression, obsessive-compulsive disorder, and other conditions.22 (Disclaimer: I invented a range of neurotechnologies based on direct electrical stimulation of the brain [both invasive and noninvasive], which could influence depression, fatigue, and many other functions, so I’m pretty biased here.23)
However, even new and relatively precise noninvasive methods still cannot target individual neurons, or even small groups of them, which is necessary if you want to produce specific complex actions. Even invasive techniques, which can produce much more precise changes, are of limited value because we often do not know which neurons we need to influence, and how. One other drawback of invasive technologies is flexibility. Even if you know how to implant a network of stimulating electrodes that gives good control over all possible movements of a single finger, what if you want to control another finger? To produce a range of complex movements, you’d need numerous stimulating electrodes, in specific areas, capable of stimulating neurons in specific patterns.
It would probably be easier to trigger complex actions that the user does frequently. Walking is quite complex, but is also so well established that a simple command can trigger a lot of complex movements. If you want to make someone play a complex piece on the piano, it is probably much easier if you find a victim who already has spent hundreds of hours playing it.
Let’s look at the complexity involved in stimulating specific neurons to get an intended result by taking the example of vision. Say you want to make someone see a tree that isn’t really there. The primary visual cortex is a horribly serpentine mess, with dozens of overlapping regions that all perform different functions. These microstructures are generally smaller than a millimeter across, and twist and tangle across three dimensions. Which neurons would you stimulate to replicate a picture of a tree? And how exactly would you stimulate them, and nothing else?
There has been some work toward writing visual images directly on the eye or brain, primarily to help people who have trouble seeing.24 Emerging technology could write directly to target neurons that otherwise would get no input at all, essentially creating a (pretty fuzzy) image in the brain based on an artificial eye. But, there’s a lot of work between that and replicating healthy natural sight. Cochlear implants also directly contact neurons, and they’ve been around for decades. But they can only amplify certain sounds, not create a rich and complex auditory scene.
Interestingly, Fringe proposed controlling the mind by writing to the thalamus rather than the primary sensory cortex. In “Dream Logic” (2-5), patients being treated for sleep disorders began experiencing terrifying dream states while awake, leading them to attack coworkers and friends whom they believed were monsters or ghouls. Inspecting the body of one of these patients, Dr. Bishop found a brain implant in the thalamus, which he said was the main relay center of all sensory information. This is largely true–all sensory information except smell goes through the thalamus before it goes on to higher processing in the brain. (Unfortunately, Peter commented that the thalamus is a midbrain structure, and his father replied, “Very good, Peter.” Sorry, Bishops–it isn’t.) When the implant is activated, it stimulates the thalamus in order to provoke the sensory experiences of a dream.
Are the writers on to something here? Is it possible that the thalamus might be the best region to target to produce sensory experiences? It might be, with a lot of caveats. The thalamus does not receive the detailed representation of sight, sound, taste, or feeling found elsewhere in the cortex. For example, nearly half the brain is devoted to vision, with specialized regions that process basic lines, shapes, colors, form, movement, and other features. The thalamus, in contrast, is a fairly small structure in the middle of the brain, and very early in the processing chain, meaning that the complex details of sensory perception haven’t yet been decoded when they reach it. So, writing to the thalamus could produce some false sensations, but with limited fidelity. There are a few other potential issues, as well. The thalamus is involved in a number of different functions, and tweaking the thalamus could have many side effects. Also, the thalamus is not near the surface of the brain–unlike many of the other cortical regions involved in sensory functions. Therefore, surgery that involves the thalamus is more complicated, expensive, risky, and ethically thorny than sticking some electrodes into the cortex (though, to be fair, the writers of Fringe have handled this potential challenge within their plots by establishing that neither Walter nor Massive Dynamic are especially concerned with these four factors). Even if the thalamus did prove to be a useful place to write to the brain, the challenges are such that research into other methods is likely to prove more fruitful, at least for the time being.
Back to Reality
So, are mind reading and mind control possible? Not really–at least not as usually portrayed in bci-fi. It’s humbling working with real-world BCI systems, which are generally quite mundane compared with the expectations of many science-fiction fans or (worse) patients and their families.25 Still, real-world stories have clearly influenced Fringe’s writers, and electrode caps off to them for parlaying so much neurotechnology into a popular show. With all of the rapid progress in BCI research, and the increasing attention among so many different people, there are countless emerging opportunities for new bci-fi and neurotechnology plot devices. A front-page story one day might be an episode of Fringe the next. It may take somewhat longer for what we see on Fringe to translate into new neurotechnology.
- “Neurotechnology: A Definition,” www.neurotechnology.uni-freiburg.de.
- J. R. Wolpaw, N. Birbaumer, D. J. McFarland, G. Pfurtscheller, and T. M. Vaughan, “Brain—Computer Interfaces for Communication and Control,” Clinical Neurophysiology 113 (2002).
G. Pfurtscheller, B. Z. Allison, C. Brunner, G. Bauernfeind, T. Solis Escalante, R. Scherer, T. O. Zander, G. M¼ller-Putz, C. Neuper, and N. Birbaumer, “The Hybrid BCI,” Frontiers in Neuroscience 4 (2010).
- M. Middendorf, G. McMillan, G. Calhoun, and K. S. Jones, “Brain—Computer Interfaces Based on Steady-State Visual Evoked Response,” IEEE Transactions on Rehabilitation Engineering 8 (2000).
- F. Popescu, S. Fazli, Y. Badower, B. Blankertz, and K. R. M¼ller, “Single Trial Classification of Motor Imagination Using 6 Dry EEG Electrodes,” PLoS One 2 (2007).
L. J. Trejo, N. J. McDonald, R. Matthews, and B. Z. Allison, “Experimental Design and Testing of a Multimodal Cognitive Overload Classifier,” Automated Cognition International Conference (2007).
C. T. Lin, L. D. Liao, Y. H. Liu, I. J. Wang, B. S. Lin, and J. Y. Chang, “Novel Dry Polymer Foam Electrodes for Long-Term EEG Measurement,” IEEE Transactions on Biomedical Engineering 5 (2011).
A. Luo, and T. J. Sullivan, “A User-Friendly SSVEP-Based Brain-Computer Interface Using a Time-Domain Classifier,” Journal of Neural Engineering 2 (2010).
- B. Z. Allison, D. Valbuena, T. Lueth, A. Teymourian, I. Volosyak, and A. Gr¤ser, “BCI Demographics: How Many (And What Kinds of) People Can Use an SSVEP BCI?” Transactions on Neural Systems and Rehabilitation Engineering 18 (2010).
- P. London, J. T. Hart, and M. P. Leibovitz, “EEG Alpha Rhythms and Susceptibility to Hypnosis,” Nature 219 (1968).
W. Larbig, T. Elbert, W. Lutzenberger, B. Rockstroh, G. Schnerr, and N. Birbaumer, “EEG and Slow Brain Potentials During Anticipation and Control of Painful Stimulation,” Electroencephalogrophy and Clinical Neurophysiology 53 (1982).
B. Z. Allison, A. Vankov, J. Overton, M. Cassarino, and J. A. Pineda, “Selective Attention to Tactile Stimuli During Hypnosis and Waking Conditions, Society for Neuroscience 23 (1997).
- J. R. Wolpaw, N. Birbaumer, D. J. McFarland, G. Pfurtscheller, and T. M. Vaughan, “Brain—Computer Interfaces for Communication and Control.”
N. Birbaumer, and L. G. Cohen, “Brain-Computer Interfaces: Communication and Restoration of Movement in Paralysis,” The Journal of Physiology 579 (2007).
B. Graimann, B. Z. Allison, and G. Pfurtscheller, “A Gentle Introduction to Brain—Computer Interface (BCI) Systems,” Brain-Computer Interfaces: Revolutionizing Human-Computer Interaction (New York: Springer, 2010).
- N. Birbaumer, N. Ghanayim, T. Hinterberger, I. Iversen, B. Kotchoubey, A. K¼bler, J. Perelmouter, E. Taub, and H. Flor, “A Spelling Device for the Paralyzed,” Nature 398 (1999).
- J. R. Mill¡n, F. Renkens, J. Mouri±o, and W. Gerstner, “Noninvasive Brain-Actuated Control of a Mobile Robot by Human EEG,” IEEE Transactions on Biomedical Engineering 51 (2004).
- J. A. Pineda, B. Z. Allison, and A. Vankov, “The Effects of Self-Movement, Observation, and Imagination on Mu Rhythms and Readiness Potentials: Toward a Brain-Computer Interface (BCI),” IEEE Transactions on Neural Systems and Rehabilitation Engineering 8 (2000).
C. Neuper, M. W¶rtz, and G. Pfurtscheller, “ERD/ERS Patterns Reflecting Sensorimotor Activation and Deactivation,” Progress in Brain Research 159 (2006).
G. Pfurtscheller, and C. Neuper, “Dynamics of Sensorimotor Oscillations in a Motor Task,” in Brain-Computer Interfaces: Revolutionizing Human-Computer Interaction (New York: Springer, 2010).
- G. Pfurtscheller and F. H. Lopes da Silva, “Event-Related EEG/MEG Synchronization and Desynchronization: Basic Principles,” Clinical Neurophysiology 110 (1999).
- C. Neuper, M. W¶rtz, and G. Pfurtscheller, “ERD/ERS Patterns Reflecting Sensorimotor Activation and Deactivation,” Progress in Brain Research 159 (2006).
R. Scherer, F. Lee, A. Schlogl, R. Leeb, H. Bischof, and G. Pfurtscheller, “Toward Self-Paced Brain-Computer Communication: Navigation Through Virtual Worlds,” IEEE Transactions on Biomedical Engineering 55 (2008).
B. Z. Allison, D. Valbuena, T. Lueth, A. Teymourian, I. Volosyak, and A. Gr¤ser, “BCI Demographics: How Many (And What Kinds of) People Can Use an SSVEP BCI?”
- D. J. McFarland, W. A. Sarnacki, and J. R. Wolpaw, “Electroencephalographic (EEG) Control of Three-Dimensional Movement: Journal of Neural Engineering (2010.)
- B. Z. Allison, and J. A. Pineda, “ERPs Evoked by Different Matrix Sizes: Implications for a Brain Computer Interface (BCI) System,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 11 (2003).
E. W. Sellers, A. K¼bler, and E. Donchin, “Brain—Computer Interface Research at the University of South Florida Cognitive Psychophysiology Laboratory: The P300 Speller,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 14 (2006).
E. W. Sellers, T. M. Vaughan, and J. R. Wolpaw, “A Brain-Computer Interface for Long-Term Independent Home Use,” Amyotrophic Lateral Sclerosis 11 (2010).
J. Jin, B. Z. Allison, C. Brunner, B. Wang, X. Wang, J. Zhang, C. Neuper, and G. Pfurtscheller, “P300 Chinese Input System Based on PSO-LDA,” Biomedical Engineering 55 (2010).
- G. Bin, X. Gao, y. Wang, B. Hong, and S. Gao, “VEP-Based Brain-Computer Interfaces: Time Frequency, and Code Modulations,” IEEE Computational Intelligence Magazine 4 (2009).
G. Bin, X. Gao, Y. Wang, Y Li, B. Hong and S. Gao, “A High-Speed BCI Based on Code Modulation VEP,” Journal of Neural Engineering 2 (2011.)
C. Brunner, B. Z. Allison, C. Altst¤tter, and C. Neuper, “A Comparison of Three BCIs Based on ERD, SSVEP, or a Hybrid Approach Using Both Signals,” Journal of Neural Engineering 24 (2011).
B. Z. Allison, J. Faller, and C. Neuper, “BCIs that Use Steady-State Visual Evoked Potentials or Slow Cortical Potentials,” in Brain-Computer Interfaces: Principles and Practice (Oxford: Oxford University Press, 2012).
- B. Z. Allison, C. Brunner, V. Kaiser, G. M¼ller-Putz, C. Neuper, and G. Pfurtscheller, “A Hybrid Brain-Computer Interface Based on Imagined Movement and Visual Attention,” Journal of Neural Engineering 7 (2010).
G. Pfurtscheller, and C. Neuper, “Dynamics of Sensorimotor Oscillations in a Motor Task.”
C. Brunner, B. Z. Allison, C. Altst¤tter, and C. Neuper, “A Comparison of Three BCIs Based on ERD, SSVEP, or a Hybrid Approach Using Both Signals.”
- B. Z. Allison, “Toward Ubiquitous BCIs,” in Brain-Computer Interfaces: Revolutionizing Human-Computer Interaction (New York: Springer, 2010).
- Y. Miyawaki, H. Uchida, O. Yamashita, M. Sato, Y. Morito, H. C. Tanabe, N. Sadato, and Y. Kamitani, “Visual Image Reconstruction from Human Brain Acitivity using a Combination of Multiscale Local Image Decoders,” Neuron 60 (2008)
- Schaefer R. S., J. Farquhar, Y. Blokland, M. Sadakata, and P. Desain, “Name That Tune: Decoding Music from the Listening Brain,” Neuroimage 56 (2011).
- Jon Hamilton, “Mind Reading: Technology Turns Thoughts into Action,” www.npr.org.
- S. K. Talwar, S. Xu, E. S. Hawley, S. A. Weiss, K. A. Moxon, and J. K. Chapin, “Rat Navigation Guided by Remote Control,” Nature 417 (2002).
- M. S. George, “Transcranial Magnetic Stimulation for the Treatment of Depression,” Expert Review of Neurotherapeutics 10 (2010).
W. Wang, J. L. Collinger, M. A. Perez, E. C. Tyler-Kabara, L. G. Cohen, N. Birbaumer, S. W. Brose, A. B. Schwartz, M. L. Boninger, and D. J. Weber, “Neural Interface Technology for Rehabilitation: Exploring and Promoting Neuroplasticity,” Physical Medicine and Rehabilization Clinics of North America 21 (2010).
- J. A. Pineda, and B. Z. Allison, “Method and System for a Real Time Adaptive System for Effecting Changes in Cognitive-Emotive Profiles,” U.S. Patent Serial No. US 7,460,903 (2008).
- A. Y. Chow, A. K. Bittner, and M. T. Pardue, “The Artificial Silicon Retina in Retinitis Pigmentosa Patients,” Transactions of the American Ophthalmological Society 108 (2010).
- J. E. Huggins, “BCIs Based on Signals from Between the Brain and Skull,” in Brain-Computer Interfaces: Revolutionizing Human-Computer Interaction (New York: Springer, 2010).
F. Nijboer and U. Broermann, “Brain—Computer Interfaces for Communication and Control in Locked-in Patients,” in Brain-Computer Interfaces: Revolutionizing Human-Computer Interaction (New York: Springer, 2010).