How Will Brain-Computer Interfaces Change Your life?
Our group of engineers, designers, and neural scientists contemplate our future as cyborgs.
Soon, we will interlace our consciousness with computers. Will you be prepared?
It may sound like science fiction. But the dawn of ubiquitous Brain Computer Interfaces (or BCIs) is fast approaching. Our panelist David Piech, a researcher developing the next generation of brain-machine interfaces, expects risk levels to fall such that clinical uses become standard practice in 10–15 years.
15 years? I’ll think about this later. You could, but increasing interest in the area is drawing efforts to fast-track this technology for commercial uses- like many cases in human history, this could be a case of applied technology preceding a cautious, full-understanding of consequences. These are the kinds of scenarios we like to ponder, so for this Futures Story experience, we sat down with neural engineers and designers, and dug into a future that, whether we like it or not, is on the horizon.
You don’t have time to get the synapse-level review- so the following is a non-invasive procedure to share the key learnings from our discussion. First, we’ll learn about the current state-of-the-art. Then we will explore the requirements for widespread adoption, and perhaps most importantly consider how neurotechnology will reshape our society.
Wrap this Around your Mind
The hero technology of this story is the brain–computer interface (BCI), also known as a neural-control interface (NCI), mind-machine interface (MMI), direct neural interface (DNI), or brain–machine interface (BMI). Some, like Thync, use electrical stimulation on the surface of the scalp to create a mental focusing affect while others, like that used by Braingate, are surgically implanted electrode arrays that help neurologically disabled patients regain motor control.
Will Kasten, a designer/engineer who consulted on the Neuralink project, offered us a useful analogy:
As far as difference between surface or deep insert deep implant. Think of it as a car. You can make a modification, like a spoiler on the car. Things that will affect the aerodynamics on the outside. As opposed to if we modify the engine of a car, with a supercharger you get a much more dramatic impact- but you risk destroying the car.
David Piech, a researcher at UC Berkeley developing the next-generation of neural interfaces, gave us another analogy considering the fidelity of non-invasive sensors:
With non-invasive approaches, the most common approaches — electrodes on scalp picking up very small potential, one analogy is that your watching football. But instead of watching inside- you are outside the stadium, you can actual get a very tenuous glimpse at what’s going on in the game, every once in a while you hear the crowd but you only know the biggest events. People right now can use oscillations that go across very broad regions of the brain and depending on how much power is in those oscillations you can the user can change that power and that can control some signals. What’s much more interesting, to me, is if, instead of standing on the outside of the stadium you actually maybe go to the top of one section of the stands and now you no longer get the most gross information from the game. You now may see how this particular section is cheering. Now, if you go even closer- imagine sitting next to someone, and now you can have an actual conversation with them, and now know exactly what their reaction is to every play.
David explained that as you go more invasive, such that you have an incredibly small electrodes in the brain listening for individual neurons, you get increasingly detailed information. One experiment involves placing electrodes into the motor cortex and having patients repeating particular motions- and if you have the patient repeat different sets of motions you can begin to map out the neurological activity. It becomes a simple data machine learning problem.
However, when invasive surgery is performed and electrodes are installed, there is a risk of affecting neural communication — but David pointed out that the brain is ridiculously redundant. “In practice, you could probably kill of 50–60% of the cells that contribute to the action and see almost imperceptible loss of function.”
But does the type of neural mapping that is occurring on current patient’s brains correlate to a generics model of how the brain works? Recent work has explored the common dynamics of neural data across different individuals. With more research, we learn more about how the brain functions- and the hope for many like David, is to elucidate the fundamental operating principles that will allow a high-bandwidth human/computer link.
The current state-of-the art is not based on a complete understanding of the brain on the synapse level. Do we need to get to that level? Do we really need complete mapping of the brain for this to happen?
Will: I think no. Yes if you want something like in the story, something that is that complete and is permeated throughout your mind. But I think there are devices with much smaller bandwidth with processing signal that could provide some communication. It doesn’t have to be as robust as that, it could be just something that just knows if its guessing things for right you, you’re giving it feedback as to whether it is guessing correctly. And training the machine learning model.
Jody Medich, the Director of Design at Singularity University Labs, believes that useful mapping should go beyond the brain and intelligently understand the user’s context. A human-machine partnership where we expect the interface to be natural- vastly superior to the clunky half-way step UX of smart-phones and computers today, with the machine intuiting our intention in a future where we are empowered rather than encumbered.
Jody: [We need something] that can create mapping of not just what’s coming from your brain, but uses smart computing to understand the context you are in and situation you are in- something could be as simple as say, understanding your focus depth and know whether or not it’s okay to send you a bunch of notifications or something silly- versus something that keeps everything else at bay. Something with a simple interface would greatly improve our mental abilities.
Sensory overload!
Bandwidth and latency. David reminded us that these are the two most present problems that neural interfaces aim to solve. Write that down.
- Bandwidth
- Latency
Humans are bandwidth limited in an environment of increasing data- how can we empower ourselves to deal with this deluge? The hope with BCIs is that a infinitesimal latency of communication between humans and computers will help. David proposes: If it currently takes 5 seconds to complete a Google search- what if we could get that down 10 or even 100 fold?
Jody- Multiple factors to latency. One is, the method we are fed data. Real world- senses, movement, these types of things. Fed data from computers through visual screen interface and linguistic command structure. Latency not just in response time to the query but in the method of accessing it. Brain machine interface, we have to take into account cognitive ergonomics just how we do physical ergonomics that amplify physical response of machine. BMI must match our sensory processing that we use to exist in the real world. Might amplify our brains functionality. Could accelerate things we aren’t aware we are triggering.
Neural scientists and researchers hope to understand enough about the human mind to create this unfathomably fast connection- but Jody provokes us to consider the affects of which we are unaware.
Matt Bell, Director of Design at DXLab , Futures author and the voice in your mind right now, hopes that neural interface developments help us unpack and understand the inner workings of our brain so that we can gain a better footing before forging a connection that begins to meddle with those secret processes. Will uncovering the abyss of our subconscious help us unlock our deepest motivators and prepare for the day that we, in a very physical and literal sense, “link” to our on-line accounts?
What comes to mind when you think about the “Killer App?”
Speaking of our on-line accounts- we all know that a flourishing technology hinges on more than neat-demos or one-hit-wonder apps. What will really drive us to make such a drastic leap into the realm of cyborgs and technokenisis? Kevin Fishner, philosopher and VP of Product at Hashicorp, thinks there are three main categories of utility for BCIs.
- Speed of Communication
2. Ability to Process More Data
3. Augmenting Biological Function
While augmenting biological function is at the forefront of current application with devices tailored specifically to a smaller subset of differently abled people, Kevin pointed out that speed and bandwidth will become interesting to all of us as the technology advances.
I think the part that is interesting to me is , do we really want those first two things? I think, as someone who lives in a society who is often overwhelmed by the amount of communication we have to deal with, and the amount of information — the idea of having more of that- even if I had a machine augmenting those capabilities- I wonder is that the direction that I would want to bring my life into? So in terms of the killer app, I’d say it’s communication- because, as we’ve seen through history: Humanity has been brought closer together as communication gets quicker. So we go from letters, to telegrams, to email, to text messages, to direct brain to brain interaction. Do we want that? And does it even matter if we want that, because it’s probably going to happen anyways?
Jody implored us to consider the entire sphere of our communication. We don’t just communicate with other humans. We live in a world which gives us superpowers- our devices offer us increased capabilities and vast amounts of data- but what if we were able to use neural interfaces to help manage and data such that we could focus on deepening our human interactions?
Jody: I talk about making super humans instead of supercomputers. Today we train humans how to become the controllers of the device, we don’t expect the device to adapt to how we process information- unfortunately that’s not how our brains function
Humans have a limited working memory, the short-term task-based memory that helps us deal with data flowing in and out, and at the moment the kind of flood of data that a BCI ushers in is beyond our current capabilities. What if we could augment this capability and thus tap into the super-powers promised by BCI technology? Jody went on to discuss another very important matter concerning working memory, something called the “Doorway effect.”
Jody: When you are switching context. You are creating something called an interruption in your working memory. Which means your brain is walking into this new context and it says I don’t need this information anymore, and it dumps it. It’s called the “doorway effect” for a reason. It’s that phenomena you feel when you walk into the other room to get a glass of water and then forget- you can’t remember why you came into that room. That’s it. So, right now the way we interact with our devices is a ton of doorways, you’re constantly switching between whatever task at hand or whatever UI, between the screen and the real world- or any other switching, and it’s really killing our brain’s ability to process information. Now, I want to use the super powers I can from my devices- but I want to be able to be human, I want to amplify what makes me a human being — and I don’t want to become a computer.
Cognitive vs. emotional and moral enhancements
Our panel agreed that the current thinking on BCIs is that augmenting bandwidth and latency to facilitate data processing was the crux upon which our computer-machine consciousnessness would be built. But what about the use of neural interfaces to augment our emotional and moral aptitudes?
Lance Cassidy, CEO of DXLabs and Fireside moderator, brought up an incredible study:
There was an interesting study that asked students about various mental traits and whether they’d be willing to use an enhancement technology to improve them. The students were very willing to use an enhancement to improve cognitive traits like attention, alertness, and rote memory. But they were loath to enhance other traits like empathy and kindness. Only 9 percent of people were willing to be enhanced in kindness. With traits like memory and language ability, the students said they’re part of me, but rather remote from my sense of self.
He then asked our panelists where they thought the line will be drawn for most people for what they would want to enhance and where they would draw the line personally.
Will: I won’t say personally, but I think that that line will creep further and further each year.
Our choice of neural enhancements has a lot do do with what we value as a society. Kevin explained that most of the drug use that occurred at his university was specifically for academic performance. While the students could easily see the benefit of taking Adderall to increase their concentration and test scores…
Kevin: …if you were to ask them, does kindness make you better at being a student? They would probably say no. And if we got asked, does kindness or empathy make you better at your job in a capitalist society, most people would unfortunately say no. It might not help as much as logic, and reason, in where we live today. So, I think- we could get into this, um- reinforcing feedback loop- where we are improving skills that fit in our society and we are over-optimizing.
But in my life-life, I think being able to see- having greater empathy and see through other people’s lives is one of the most powerful things, and that is what I’d hope we could gain with this. So, if I had to choose, I would like to think I would choose empathy.
Jody hit the question out of the park with her exceedingly refreshing honesty:
I don’t want more empathy.
And that hits on a huge issue. When sociopolitical turmoil, economic engines, and marketing hinges on a population whose heart-strings, and crocodilian cerebellum keys are played in an orchestra of distraction what is the ideal amount of empathy- And what does it mean to the authenticity of our identities if we augment our empathy? And what if we want to revert back?
Matt: I think a big question, is: is it reversible? If I could enhance certain aspects of my behavior, basically becoming a human chameleon based on the context- whenever I need to have increased empathy, bandwidth, memory- I’d probably do that. If I could reverse it. I can envision spies using this like chameleons- changing their abilities. Above this, personally, I really like the idea of being able to understand your own motivations and subconscious.
Not In My Brain Yet
We considered the adoptability of BCIs and the factors which would lead to a society seamlessly connected to super-powering technology. The current state-of-the-art for neural augmentation involves invasive surgery — electrodes are implanted deep into brain tissue in order to get precise signals into and out of the brain. But as the fidelity of sensors increases and the risks involved in less-invasive procedures decreases, we envisioned a world delving deeper into neural augmentation.
We do not need something that is completely mapped to our brains and permeated through our minds. What if we combined lower-fidelity technology with intelligent computing to learn and adapt to us?
Will: …there are devices with much smaller bandwidth with processing signal that could provide some communication. It doesn’t have to be as robust as that, it could be just something that is- that just knows if its guessing things for you, you’re just giving it feedback as to whether it is guessing correctly. And training the machine learning model, okay this user- this is the sort of thing you’re looking for.
Social Impact and Disparity
It is critical to consider how modifying our right-brained and left-brained aspects might exacerbate the divide in human behaviors- but what about the divide between the rich and poor? Not only are we concerned with economic disparity- but what are the multifarious ways in which BCIs could positively or negatively impact our future? How can we ensure a positive outcome?
Will: Provided that the world stays the way that is right now, it’s hard to not see how those with the means to acquire this will do so and those without the means won’t. Um, that is disparity gap will be larger than anything we’ve ever seen in human history if that goes that way. My concern is less about social context and more if the military starts doing that. And then you have a bunch of super soldiers.
Jody to Will: That study David was talking about where they are trying to find commonalities between brains, that’s happening at DARPA.
Our panelists touched on an age-old question concerning the responsibility of research and researchers themselves to understand the implications of their work and for whom they work for. No future is inevitable. We have to be mindful that our history shows innovations that promise great power are often keenly watched by those who wish to entrench themselves- and the pure, scientific curiosity that drove the development of atomic physics didn’t preclude the proliferation of the tools of atomic destruction.
Kevin Fishner asked us to ponder how the gap between the rich and poor may be exacerbated by this technology as distribution is limited. Moore’s law has created cheap computing, but the gap has grown. How can we ensure a future where an equal opportunity exists for access and use of BCIs?
Kevin: What I’m curious about is, what is the business model going to be? Because there is the notion that, writing- is not really a business model. I guess it would be the book publishing, and there’s a very limited access of people who can publish- rather than people that can write. I’m wondering about that. I think that we’re seeing something similar with smart phones, it consolidated so much wealth, — android. Access to wireless internet. I’m curious, what is the business model around these interfaces going to be? And will it consolidate wealth, versus more democratized access to these devices. Based on where we are currently, pessimistically I would say it would probably consolidate and the business model would be a central supplier that would create different levels of access to generate more money, unfortunately. But I don’t know, I’m curious.
Conclusions
We get it. In the next couple decades BCIs will start to make headway into our lives and forever change what it means to be human. *Exhales* So what do we need to do now to prepare for a future where BMI is ubiquitous?
Will: I mean, I think talking about it is a start. If this conversation that we are having now were taking place all around the country, around the world. Then we’d be in a great place. I think we’re not there yet, and we are a ways from there socially. the only way it’s gonna happen is if all the people who are sitting here and care about it make other people care about it.
The onus is on each of us to choose what is most important in our lives- as humans we are inherently search for meaning in what we do, and on the impact we can achieve- and in our talk it was clear that the coming age of human-machine integration has the potential to be the largest shift both in our culture, and in our manifest power to change the our inner and outer worlds completely. This is a topic worth discussion, deep contemplation, and as soon as possible: action.
Jody: Learn to be adaptable and creative. Things are gonna change. You’re gonna have to figure out how to help shape it. A lot of times I talk to people, and I talk to a lot of people about this, their first reaction is “Oh my god, how do I stop this from happening?” And I have to help them realize, it is happening right now.
Matt: Historically our technological advancements tend to precede our full scientific understanding of underlying phenomena. We invented the pulley but didn’t really know the dynamics and mechanical properties that made it work, all the physics behind it. We invented a medicine with a herb that tends to help with some kind of ailment and THEN we figured out why, biochemically it works. My fear with this, is that if we don’t fully understand the brain first, we’re messing with it in maybe a permanent way, maybe not- you might be do something that you regret.
If BCIs allow us to override our subconscious motivators with behaviors that align with our surface-level ideal selves, we might need to start pondering how we, each of us, want to participate in this future. Think of our 12 year old selves being allowed to fundamentally restructure our brains — would you trust them to decide on the fundamental restructuring of your mind? It is clear that in this coming future, we must be mindful of how important it is that we understand ourselves deeply before potentially, irreversibly changing who we are.
What is Futures Story?
Futures is a publication of illustrated sci-fi short stories by DXLabs that explores the potential impact of technologies on our society. Futures Story is a live reading experience and fireside chat with leading experts. This Futures Story reading is based on Replenish by Matt Bell and Tia Kansara.
References
https://www.digitaltrends.com/cool-tech/neural-dust/
https://bcs.mit.edu/news-events/news/machine-learning-system-processes-sounds-humans-do
https://mit.applysci.com/2017/07/direct-brain-path-for-sight-sound-via-implanted-microscope/
