Silent communication: A brain-computer interface that can read minds

In recent times, computers have begun to feel like extensions of our consciousness. Appointments are stored on digital calendars, routes are navigated using GPS, and phones can act as convenient reminders of friends’ birthdays. With computers playing such an important role in modern life, the boundary between our brain and computers is becoming thinner. In the world of neuroscience, brain-computer interfaces (or BCIs) are used to establish communication between the brain and an external electronic device. Unlike mind-reading chips in science fiction narratives, these BCI devices are designed to perform simple tasks such as determining whether a person responded ‘yes’ or ‘no’ to a question without them speaking or moving. Recently, researchers at Western University developed one such BCI and successfully tested it on healthy adults. 

The motivation for developing this BCI comes from research on people in a vegetative state, which is a state of consciousness whereby a person is awake but not aware of their surroundings. While these patients were traditionally though to lack awareness, work by Dr. Adrian Owen (one of the researchers involved in the BCI study) showed that not only do these patients have some awareness, communication between them and the outside world can be established. In his study, Owen asked a patient who was in a vegetative state to imagine playing tennis and exploring her house. Each of these thoughts activates dinstinct regions in the brain. While imagining those scenarios, the patient’s brain activity was measured in an MRI machine, which showed different patterns of activity while she imagined the two scenarios – implying that she was able to understand the instructions and respond to the researchers by thinking about the two scenarios and modulating her brain activity. This marked a shift in how consciousness in vegetative patients is viewed and suggested that communication with these patients is possible. 

Since the vegetative state study, efforts have been focused on optimizing this communication by using technology that’s more accessible than an MRI brain scanner, which is large and expensive to use. This is where the recent study, published in Frontiers in Neuroscience, comes in. Androu Abdalmalak, a PhD student at Western University, set out to use a non-invasive brain imaging technique to make a BCI and test it on healthy adults. Abdalmalak’s study used functional near-infrared spectroscopy (fNIRS), a technique that measures changes in the amount of oxygen in the blood at various positions along an individual’s scalp. Since brain areas that are more active use up more oxygen, this measurement correlates with brain activity. 

An fMRI brain scanner (left) is much more expensive and less portable than an fNIRS system (right). Images by Thomas Angus (left) [CC BY-SA; https://creativecommons.org/licenses/by-sa/4.0], and Elisenicolegray (right) [CC BY-SA; https://creativecommons.org/licenses/by-sa/4.0].

To determine if their BCI works, the research team asked a series of questions to healthy participants while their blood oxygenation levels were recorded by the fNIRS system and participants were asked to imagine playing tennis to answer ‘yes’ and do nothing to answer ‘no’. The rationale goes like this: when participants imagine playing tennis, there should be an increase in activation in motor areas of the brain (areas involved in planning and carrying out movement). Thus, when participants want to answer ‘yes’, activity in motor areas should increase. In contrast, when they want to answer ‘no’, the level of activity in motor areas should not change since this task does not involve the motor areas. Indeed, this is what the team observed. The participants’ responses were accurately classified by computer programs. That is, computer programs were able to determine which patterns of brain activity indicated a “yes” and which patterns indicated a “no”, confirming the viability of this approach. As a result, this study suggests that the combination of fNIRS with motor imagery tasks can be used as a BCI. 

You may be wondering ‘why was this research done on healthy participants?’ The reason for that is that the new methodology has to be tested with healthy individuals before it can be used clinically. After these promising results from the BCI on healthy participants, the next step is for it to be used with patients that have disorders of consciousness (disorders in which consciousness is affected due to brain damage). Some of the patients with this group of disorders are able to carry out mental functions such as imagining an action but are unable to communicate with the external world using things like movement and language. 

While brain imaging with MRI has been used in research to communicate with these patients for over a decade, fNIRS offers some added possibilities. Unlike brain imaging, which uses bulky multi-million-dollar machines, fNIRS offers a cheaper portable alternative. This avenue of research offers a lot of hope for these patients and their loved ones, as this could be a device they can keep in their home to use on a daily basis. And while sci-fi brain chips are still beyond the horizon, realistic BCIs pave the way forward by restoring communication with damaged minds. 

Original article: Abdalmalak, A., Milej, D., Yip, L. C. M., Khan, A. R., Diop, M., Owen, A. St. Lawrence, K. (2020) Assesing time-resolved fNIRS for brain-computer interface applications of mental communication. Frontiers in Neuroscience, 14: 105.