Emotiv Systems, an electronic-game company from San Francisco, wants people to play with the power of the mind. Starting tomorrow, video-game makers will be able to buy Emotiv’s electro-encephalograph (EEG) caps and software developer’s tool kits so that they can build games that use the electrical signals from a player’s brain to control the on-screen action.
Emotiv’s system has three different applications. One is designed to sense facial expressions such as winks, grimaces, and smiles and transfer them, in real time, to an avatar. This could be useful in virtual-world games, such as Second Life, in which it takes a fair amount of training to learn how to express emotions and actions through a keyboard. Another application detects two emotional states, such as excitement and calm. Emotiv’s chief product officer, Randy Breen, says that these unconscious cues could be used to modify a game’s soundtrack or to affect the way that virtual characters interact with a player. The third set of software can detect a handful of conscious intentions that can be used to push, pull, rotate, and lift objects in a virtual world.
The notion of using brain activity to interact with computers isn’t new. A number of schools–such as the University of Minnesota; University of California, San Diego; and Purdue–have research labs devoted to decoding thoughts from the brain and manipulating cursors on a screen, which is especially useful for disabled people. In addition, companies have cropped up in the past couple of years claiming to offer an effective brain-computer interface for video games or for biofeedback purposes. For instance, S.M.A.R.T. BrainGames, a company based in San Marcos, CA, sells games and EEG caps designed to treat people with attention deficit/hyperactivity disorder.
To use Emotiv’s system, a person puts on the EEG cap and adjusts it to her head, making sure that most of the sensors touch the scalp. The system automatically picks up blinks and emotional states. However, in order to move virtual objects, such as a box on a computer screen, a person must go through a series of training sessions in which she concentrates for about 10 seconds on mentally moving the box. Tan Le, one of Emotiv’s cofounders, says that there is a large amount of machine learning built into the software, so the more a person concentrates on a specific task, the more precisely the system follows the mental instructions.
Since Emotiv’s technology is currently patent pending, the company will not disclose the details of its system. However, Le claims that the company’s heads of research–optics expert Allan Snyder and former Bell Labs chip engineer Neil Weste–have made a number of scientific discoveries that are worthy of academic research papers. But so far, none have been published, and no game manufacturer has publicly committed to using the technology.
Bin He, professor of biomedical engineering at the University of Minnesota, in Minneapolis, is skeptical of the system. He says there are some universal challenges with such systems and specifically with the EEG caps, even in a lab environment. Typically, neuroscientists use a conductive gel to help the cap’s sensors collect as much information as possible. And even with the gel, signals that represent thoughts are very weak, he says.
There are other electrical signals that are stronger, however, and Emotiv’s system could be detecting such signals in order to move virtual objects, says Alan Gevins, executive director of the San Francisco Brain Research Institute and president of SAM Technology. For instance, the electrical potential produced at the scalp from eye movements and facial and scalp muscles are at least 15 times stronger than those produced by activity in the brain. It is possible that when a person concentrates on pushing, lifting, or turning a virtual object, his or her eyes, jaw, head, or tongue move in a characteristic way, Gevins says.
But the fact that there is still much work to be done in EEG research in general leads Gevins to be skeptical of new consumer products. “You need not only advances in sensor technology, but also in application-specific signal analysis and, most importantly, [in] understanding which brain signals relate to thinking,” he says. “And once you have that in the lab, it needs to be refined and refined and then refined some more.”
While there still may be work to be done, Emotiv is betting that its technology is good enough for prime time, and it plans to make the system available to the general public in 2008. The question of whether or not Emotiv’s technology will be a successful commercial product remains to be answered. Using the power of concentration to move virtual objects may have a novelty factor, but it is a slow and an unnatural way to manipulate objects in many gaming scenarios. Also, in order for Emotiv’s system to work, the sensors on the cap must maintain contact with the scalp, which means that head and body movements should be kept to a minimum. Taking it out in the real world, Gevins says, could be difficult. “It doesn’t do anyone any good to bring something out of the lab prematurely,” he says.
Toronto wants to kill the smart city forever
The city wants to get right what Sidewalk Labs got so wrong.
Saudi Arabia plans to spend $1 billion a year discovering treatments to slow aging
The oil kingdom fears that its population is aging at an accelerated rate and hopes to test drugs to reverse the problem. First up might be the diabetes drug metformin.
Yann LeCun has a bold new vision for the future of AI
One of the godfathers of deep learning pulls together old ideas to sketch out a fresh path for AI, but raises as many questions as he answers.
The dark secret behind those cute AI-generated animal images
Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.