At F8 2017, Facebook announced a brain-computer interface (‘BCI‘) program with a goal of creating a silent speech system capable of typing 100 words per minute straight from your brain. Last month the social networking giant has provided an update through a blog post.
As per the update, Facebook is supporting a team of scientists at the University of California, San Francisco (‘UCSF‘) through Facebook Reality Labs (‘FRL‘) a division dedicated to augmented reality and virtual reality R&D. The ultimate goal of Facebook is to build a non-invasive, wearable device that lets people type by simply imagining themselves talking.
The Facebook funded team at UCSF has published the results of the research in a recent publication available here. The research at UCSF involved human participants who were already undergoing brain surgery to treat epilepsy listening to questions and responding aloud with answers while the researchers used high-density electrocorticography (ECoG) recordings to detect when human participants heard or said an utterance and to then decode the utterance’s identity in real time. The ECoG arrays were surgically implanted on the cortical surface of one hemisphere for each human participant.
The human participants were put to questions like: which musical instrument do you like listening to? (with options as Piano, Violin, Electric guitar Drums Synthesizer, None of these). The researchers were able to decode produced and perceived utterances with accuracy rates as high as 61% and 76%, respectively. However, the present AI algorithm as of date can only recognize small set of words and phrases and far from Facebook aim of building non-invasive, wearable device.
Facebook proposes to indirectly measure brain activity by detecting shifts in oxygen levels within the brain. A method similar to how Pulse Oximeter is able to measure the oxygen saturation level of our blood through fingers in a safe, non-invasive way. FRL also partnered with Mallinckrodt Institute of Radiology at Washington University School of Medicine and APL at Johns Hopkins.
Dr. Philip Kennedy and his Neural Signals, Inc
In 2007, computational neuroscientist Frank Guenther of Boston University in collaboration with Dr. Philip Kennedy and Neural Singals Inc. had developed the first speech BCI way back in 2007. The system used electrodes implanted in the brain of a man with locked-in syndrome to eavesdrop on the motor cortex’s plans to speak. The study went as far as detecting vowels. The project ended after FDA had withdrawn permission to use the devices in any more patients. FDA reportedly asked Dr. Phil Kennedy for more safety data, including on the neurotrophic factors which the team couldn’t provide. Years later in 2014, Kennedy underwent a brain surgery to insert intracranial electrodes to experiment on himself. The plan didn’t go well and left Kennedy speechless for few months. Neural Singals Inc. is currently developing a multichannel, implantable recording system in collaboration with Andrew Wilder, CEO of Ripple Inc.
Others in the race to control Human Brain
- Defence Advanced Research Projects Agency, United States Department of Defense
- Kernel, co-founded by Bryan Johnson who made a fortune of US$ 800 million by selling his payment services company Braintree to Paypal. Kernel aims to develop “neuroprosthesis,” a device that will act as a supplement to existing brain.
- Paradromics, Austin, Texas-based startup that is working neuro-prosthetic device to create streaming capabilities between the brain and computers.
- Neuralink, Elon Musk backed entity aims to develop flexible “threads” that can be implanted into a brain and be used to control your smartphone or computer with just your thoughts.