A reader may be the type of person who is perfectly content with existing computer interfaces and their manageable frustrations. Maybe society is regressing to single finger typing on smartphones, and struggling to position current devices in the perfect compromise between the needs of the hands and those of the eyes, and a little disappointed every time an application-update needlessly reshuffles all the features, but what can be done? This is the status quo. One will carry on coercing a computer with the tools available and there will be no complaints so long as the changes are not too drastic to keep up with. Someone may be this type, it is a fine way to be, or someone might be one of those disruptive and disabled-friendly people who are determined to turn the world into one giant brain-computer interface.
Commonly referred to by their initials, Brain Computer Interfaces (BCI) encompass a wide variety of technologies which aspire to connect brain signals directly to the control of computers. The technology is not entirely new, but for most of the thoughtful history of BCI they have been “invasive,” which is to say, surgically implanted. This type of BCI is the most well-developed and inspiring in that it promises to return sight to the blind, hearing to the deaf and touch to the amputated, applications also known as neural-prostheses, but these are distinctly non-consumer grade. If someone is not suffering from a severe ailment then it is unlikely they are going to agree to have chips and wires installed inside of their mind. There is another option however. The EEG, or electroencephalogram, measures the electrical signals generated by the brain from the dry side of the skull. Long used in clinical and research settings as a measurement device, the technology is now being deployed in several consumer grade products as a user interface.
Neurosky, InteraXon and Emotive all offer EEG headsets that can be calibrated to any person and then used to interact with a computer. They look like pretty swanky gadgets, so there is definite technophile appeal, and they come with software packages which allow a casual user to enjoy them without a great deal of understanding. Among the hobby engineers they seem to be pretty popular as a scientific pet-project, and among the gamer crowd there is some immersion appeal. Although, if someone is merely interested in showing off their hip mental wizardry on the day to day circuit, there may be some difficulties. EEG is a very low resolution technology, meaning that far from being the reader of the the mind, these devices are simply calibrated to perform some function when they recognize a certain electrical signal on the scalp.
The signals they read are the result of thousands if not millions of outer cortex level neurons all firing together. With the bio-feedback software these devices include, someone can train themselves to reproduce certain electrical signals, but the number of actions these devices can be used to elicit on the computer is limited to the number that have been programmed by the user. There are also some accuracy and latency issues at play here, so the responsiveness of these products is not at all comparable to that of the mundane hand-operated interfaces and devices we use already. Sadly, as in this video of the game MindFlex and in other product demos one can find on YouTube, people using these BCIs appear to have very little control over the ability to produce the desired electrical signal.
The different headsets measure different signals, some more complicated such as visualization of movement, but more often they measure something described as “concentration” which correlates to a reading of the brainwaves. The user’s ability to control these readings is hard to take seriously within the context of usefulness. While it would be a huge advantage for a paralyzed person, cycling through the whole alphabet every time someone wants to pick a letter is not very helpful to the consumer crowd. The limitations are not surprising, these devices are in the early stages of consumer use and may get better, but it does raise some doubts as to the actual potential of non-invasive EEG for brain computer interfacing.
The doubts raised are not meant to discourage interest, nor are they indisputable, the technology is very cool, and there may very well be a means of improving non invasive BCIs such that they are valuable tools. Certainly there are uses for them already, and advantages to consumer grade products over the complicated, expensive and messy clinical EEG electrodes, but when it comes to the science fiction fantasy of strong BCI, it is certainly still part of the future and it may well not include non invasive options.
Editorial by Eric Anderson