Over the past two decades, the international biomedical research community has demonstrated increasingly sophisticated ways to allow a person’s brain to communicate with a device, allowing breakthroughs aimed at improving quality of life, such as access to computers and the internet, and more recently control of a prosthetic limb. DARPA has been at the forefront of this research.
“The tools we use have grown more sophisticated over time … but these still require some form of physical control interface—touch, motion or voice. “What neural interfaces promise is a richer, more powerful and more natural experience in which our brains effectively become the tool.”
The state of the art in brain-system communications has employed invasive techniques that allow precise, high-quality connections to specific neurons or groups of neurons. These techniques have helped patients with brain injury and other illnesses. However, these techniques are not appropriate for able-bodied people. DARPA now seeks to achieve high levels of brain-system communications without surgery, in its new program, Next-Generation Nonsurgical Neurotechnology (N3).
DARPA’s The Next-Generation Non-Surgical Neurotechnology (N3) program seeks non-invasive or minimally invasive neural interfaces to connect warfighters directly through thought to computers or other digital devices to enable fast, effective, and intuitive hands-free interaction with military systems.
In addition, for efficient warfighter multitasking it is imperative that warfighters be able to interact regularly and intuitively with artificially intelligent (AI), semi-autonomous and autonomous systems in a manner currently not possible with conventional interfaces. The N3 program will develop the interface technology required for current and future systems.
“DARPA created N3 to pursue a path to a safe, portable neural interface system capable of reading from and writing to multiple points in the brain at once,” said Dr. Al Emondi, program manager in DARPA’s Biological Technologies Office (BTO). Noninvasive neurotechnologies such as the electroencephalogram and transcranial direct current stimulation already exist, but offer nowhere near the precision, signal resolution, and portability required for advanced applications by people working in real-world settings. “High-resolution, nonsurgical neurotechnology has been elusive, but thanks to recent advances in biomedical engineering, neuroscience, synthetic biology, and nanotechnology, we now believe the goal is attainable.”
“Smart systems will significantly impact how our troops operate in the future, and now is the time to be thinking about what human-machine teaming will actually look like and how it might be accomplished,” Emondi said. “If we put the best scientists on this problem, we will disrupt current neural interface approaches and open the door to practical, high-performance interfaces.”
“As we approach a future in which increasingly autonomous systems will play a greater role in military operations, neural interface technology can help warfighters build a more intuitive interaction with these systems.
DARPA’s Four-year N3 Program
One of the biggest issues researchers face when developing neural interfaces is keeping the tech homed in on the right part of the brain. Our brains are constantly gaining and losing neurons, so the machines often need to be recalibrated as neural connections change.
But through artificial intelligence, researchers could train the interface to automatically pick up on these changes and recalibrate itself accordingly, DARPA wrote in the solicitation. Under the program’s first track, teams would build algorithms that adjust the interface when neurons are lost or added, as well as if there’s any interference between the system and the brain. Teams participating in the second track of the program will explore ways to overcome another limitation of neural interfaces: the human body itself.
The brain receives a constant stream of sensory information from the maze of nerves spread throughout the body, but there’s only so many feelings a given nerve can express. Under the program, teams will also build an AI-powered interface that can stimulate “artificial signals” within the body—creating a sense of burning without heat or touch without physical contact, for example.
Such a system would connect to the upper torso and “maximize information content carried” along major nerves, DARPA said. Teams under both program tracks are eligible for up to $1 million in funding and will have 18 months to build a prototype. Proposals must be submitted to DARPA by March 4.
Potential N3 researchers will face numerous scientific and engineering challenges to bypass those limitations, but by far the biggest obstacle will be overcoming the complex physics of scattering and weakening of signals as they pass through skin, skull, and brain tissue. The interface must be bidirectional and will integrate technology for both neural recording (read out) and neural stimulation (write in). The developed technology must be agnostic to the interfaced DoD-relevant system.
“We’re asking multidisciplinary teams of researchers to construct approaches that enable precise interaction with very small areas of the brain, without sacrificing signal resolution or introducing unacceptable latency into the N3 system,” Emondi said. The only technologies that will be considered in N3 must have a viable path toward eventual use in healthy human subjects.
The Next-Generation Non-Surgical Neurotechnology, or N3, program is a project which will aim to create two new technologies. One will be a minute interface where the user may have to ingest different chemical compounds to help external sensors read their brain activity, ultimately allowing them to control the technology using their mind. The other will be a non-invasive technology that will monitor the brain and the machine.
To reach high temporal and spatial resolution, N3 will focus on two approaches: noninvasive (Technical Area 1 –TA1) and “minutely” invasive (Technical Area 2 – TA2) neural interfaces. Noninvasive interfaces will include the development of sensors and stimulators that do not breach the skin and will achieve neural ensemble resolution (<1mm3).
Minutely invasive approaches will permit nonsurgical delivery of a nano transducer: this could include a self-assembly approach, viral vectors, molecular, chemical and/or biomolecular technology delivered to neurons of interest to reach single neuron resolution (<50μm3). In this application, the developed technology will serve as an interface between targeted neurons and the sensor/stimulator. They should be sufficiently small to not cause tissue damage or impede the natural neuronal circuit. The sensors and stimulators developed under the minutely invasive approach will be external to the skull and will interact with the nanotransducers to enable high resolution neural recording and stimulation.
Both noninvasive and minutely invasive approaches will be required to overcome issues with signal scattering, attenuation, and signal-to-noise ratio typically seen with state of the art noninvasive neural interfaces. Systems that are larger or requiring a highly controlled environment – such as magnetoencephalography (MEG), or magnetic resonance imaging (MRI) – and proposals describing incremental improvements upon current technologies, such as electroencephalography (EEG), may not be considered responsive to this BAA and may not be evaluated.
If early program deliverables overcome the physics challenges, along with the barriers of crosstalk and low signal-to-noise ratio, subsequent program goals would include developing algorithms for decoding and encoding neural signals, integrating sensing and stimulation subcomponents into a single device, evaluating the safety and efficacy of the system in animal models, and ultimately testing the technology with human volunteers.
Final N3 deliverables will include a complete integrated bidirectional brain-machine interface system. Non-invasive approaches will include sensor (read) and stimulator (write) subcomponents integrated into a device (or devices) external to the body. Minutely invasive approaches will develop the nanotransducers for use inside the brain to facilitate read out and write in.
Minutely invasive approaches will also develop the external subcomponents and integrated devices that interact with the internal nanotransducers. N3 developed technologies may move beyond the traditional voltage recordings associated with action potentials, and include different types of signals, such as light, magnetic/electric fields, radiofrequency, and neurotransmitter/ion concentrations. These atypical signals may require the development of new algorithms to enable accurate decoding and encoding of neural activity. To that end, the N3 program will include a computational and processing unit that must provide task relevant decoded neural signals for control in a DoD-relevant application. It must also provide the capability to encode signals from a DoD-relevant application and deliver sensory feedback to the brain. The processing unit must decode/encode in real time with minimal system latency.
DARPA intends the four-year N3 effort to conclude with a demonstration of a bidirectional system being used in a defense-relevant task that could include human-machine interactions with unmanned aerial vehicles, active cyber defense systems, or other properly instrumented Department of Defense systems. If successful, N3 technology could ultimately find application in these and other areas that would benefit from improved human-machine interaction, such as partnering humans with computer systems to keep pace with the anticipated speed and complexity of future military missions.
Proposers must use approaches that ensure confidentiality, integrity, and availability (also known as the CIA triad) to prevent spoofing, tampering, or denial of service. It will be necessary to adequately secure the connection between the integrated device, the processing unit, and the system user’s brain. Proposers must incorporate inherently safe techniques into any wireless and electronic portions of their system, and proposals must describe the specific protocols and techniques to be used.
Ethical, Legal, and Societal Implications (ELSI)
DARPA has invited federal regulators to participate from the beginning of the N3 program, serving as aids for researchers to help them better understand regulatory perspectives as they begin to develop technologies. Later in the program, these regulators will again serve as a resource to guide strategies for submitting applications, as needed, for Investigational Device Exemptions and Investigational New Drugs.
DARPA is being similarly proactive in considering the ethical, legal, and social dimensions of more ubiquitous neurotechnology and how it might affect not only military operations, but also society at large. Independent legal and ethical experts advised the agency as the N3 program was being formed and will continue to help DARPA think through new scenarios that arise as N3 technologies take shape. These individuals will also help to foster broader dialogue about how to maximize societal benefit from those new technologies. Separately, proposers to N3 must also describe mechanisms for identifying and addressing potential ethical and legal implications of their work. As the research advances, published N3 results will further facilitate broad consideration of emerging technologies.