In the EEG world, a Brain-Computer Interface is known as a BCI. Webster’s Dictionary defines an interface as a system through which one machine is connected to another machine. Unfortunately, this definition doesn’t explain how the data from the transmitting machine is being processed and utilized by the receiving machine. It also doesn’t elaborate on the interesting history and evolution of BCI. In this post, we’re going to expand on the definition of BCI, as well as dive into what past purposes BCI has served and how has it evolved into the consumer BCI we see today.
First, let’s step back in time to recognize some fascinating accomplishments in BCI history1:
- 1924 : Hans Berger, a German neuroscientist, discovers the electrical activity of the human brain with EEG
- 1970: Defense Advanced Research Projects Agency of USA initiates program to explore brain communications using EEG
- 1976: UCLA’s Brain Computer Interface Laboratory provides evidence that single trial visual evoked potentials could be used as a communication channel effective enough to control a cursor through a two-dimensional maze. Profess or Jacques J. Vidal coins the term BCI.
- 1998: First (invasive, non-EEG) implant in the human brain that produces high quality signals
- 1999: BCI is used to aid a quadriplegic for limited hand movement
- 2002: Monkeys are trained to control a computer cursor
- 2003: First BCI game is demonstrated to the public (BrainGate)
- 2005: Monkey brain controls a robotic arm
- 2008: Voiceless phone calls are demonstrated (The Audeo – TI developers conference)
What are some of the explicit themes visible here? First, BCI has had a long history centered on control applications: cursors, paralyzed body parts, robotic arms, phone dialing, etc. Second, many of the applications are centered on the needs of the disabled community. Third, higher granularities of control require electrode implants (residing on the brain surface—yes, a physician drills through your skull) versus the traditional EEG cap (residing on the head surface).
There are also implicit and subtle themes not fully represented, but deserving mention. First, there is a lot of complex and expensive hardware and software involved in most of the use cases. Second, a scientist or clinician is normally assisting in the use of the technology. Third, the inconveniences to the user are enormous.
Around 2007, NeuroSky launched what is considered the consumer BCI space, as the technology had evolved from medical to consumer-facing. This was a major leap forward for BCI in that now it better met the more exacting consumer requirements for price, ease-of-use, aesthetic design (fewer electrode contacts), comfort, mobility and, of course—no skull drilling. This was new territory for BCI, and clearly certain sacrifices were required. BCI control granularity was attenuated in order to satisfy users who had no patience for BCI products that introduced inconveniences into their lives. What consumer BCI could promise was a scaled down and courser version of control. Imagine one electrode (consumer headset) replacing a full array of 64 or 128 electrodes or, more unbelievably, direct brain implants. A simple rule of thumb is that BCI is full of important trade-off considerations driven by the all-important use case model.
When it comes to control, consumer BCI is not about reliably and accurately navigating a prosthetic limb or a cursor across a screen, but more about delivering an entertaining and empowering value component to the overall user experience. The culmination of consumer BCI for control led to its inevitable use as a mental telepathy vehicle. Telepathic control in entertainment had non-stringent control granularity requirements (instead of moving a cursor two pixels to the left, you needed to simply push a virtual car forward). In this way, BCI and telepathy became joined at the hip and established the traditional, and seemingly only, use case for consumer BCI. Yet, as you will read in upcoming posts, that is not the actual case.
Footnote #1 – http://www.academia.edu/1365518/Brain_Computer_Interface_Past_Present_and_Future