The brain is directly involved in all our intentional actions, thoughts, states of mind and behaviors. Unlike most other organs in the body, the way the brain works is far from being fully understood. The fundamental underlying mechanisms involved in functions such as memory, focus or our changing mood are still under active investigation.
Yet, we have the intuitive feeling we "know" how our brain works, at least in a heuristic way: we "live" in our brain. Indeed, brain functions have been studied through controlled observation and disciplined scientific experimentation, a field of science that is referred to as behavioral sciences.
To unify and improve our comprehension of the brain, an active field of research and challenge in neuroscience consists in studying the correlation between observed cognitive states and cerebral activity in order to decipher the brain signal.
One way to study the brain and measure its activity is to evaluate the very small electrical currents that groups of neurons firing together create at the surface of the scalp. This brain imaging technique is called electroencephalography, or simply EEG. This invention accredited to Hans Berger almost a century ago made a big impression when presented to the scientific community at the time, and has been recently described as "one of the most surprising, remarkable, and momentous developments in the history of clinical neurology" (D. Millet, The origins of EEG, 2002). Today, EEG is widely used in clinical and research facilities. Typical applications for EEG include sleep medicine and seizure localization. EEG is a powerful tool yet relatively affordable brain imaging technique that is used in various situations as different as ER or cognitive research protocols.
More recently, EEG has been used for Brain Computer Interface, or BCI which is a way to exchange information between the brain and a machine.
EEG can only measure brain activity along the scalp as opposed to other brain imaging techniques such as say functional Magnetic Resonance Imaging (fMRI) or Positron Electron Tomography (PET) scan. However, neuroscientists realized quickly that EEG could be used in many situations where other imaging techniques were impractical, too expensive or simply impossible to use, and that the high temporal resolution of EEG signal was able to monitor brain activity with a unique high accuracy.
EEG is still for instance the preferred and probably most reliable way of "scoring" sleep, which implies determining precisely the different stages we fall into while asleep , including deep sleep, light sleep, Rapid Eye Movement.
During the day, the brain also fluctuates between different states, influenced in particular by circadian cycles. EEG remains a powerful tool to identify these states and rather logically, increasing interest has been shown lately about the usage of EEG to identify states of the brain while awake.
EEG signal pre-Processing
The signal measured using EEG is a voltage, in the form of electrical waves reflecting the brain activity. Typical EEG signal has an order of magnitude of 10 microvolts, which is a very small voltage. Such a weak signal needs a high level of amplification to be detected and is very sensitive to noise, which includes physiological noise other than cerebral activity (heart, breathing, perspiration, muscle activity) and electronic noise (mainly the 50/60Hz AC current from the mains, other electro-magnetic waves surrounding us, movements). A careful pre-processing of the raw signal is necessary to filter out the noise and keep the useful cerebral information.
The brain music: from Voltage to Cognitive state
Brain waves are very similar to audio waves. The signal detected using EEG is like a very subtle music that our brain would play. Depending on our state and activity, the brain radio plays different music styles: classical music, talk show, pop, techno, jazz,... Depending on the music style that is played, the frequency spectrum varies: for instance, techno music will offer more bass than classical music. The same can be said regarding brain signal: by analysing its frequency spectrum and the relative weights of each frequency band, we can identify which kind of brain activity is produced, and therefore guess which state we are in. Just like music can be rendered more realistically using more audio channels (mono, stereo), brain signal can be analysed more accurately using more EEG channels. Of course, this is not as simple as it may "sound". We can not always find a reliable spectrum signature for any cognitive state: happiness, melancolia, jealousy still have their secret mix! Conversely, some recorded spectrum might correspond to different states. The discrimiation power also depends on the quality of the signal, number of channels... This is an active research topic under constant evolution: one of the reasons we are not as advanced as we could be in this field might be the difficulty of objectively naming and defining the cognitive states (what is happiness, is there only one form of concentration, has "very tired" the same meaning for everybody?). The other reason is the difficulty of gathering quality databases in real life conditions, over a long period of time, on a wide varity of people.
CONSCIOUS LABS scientific mission
CONSCIOUS LABS believes that identifying mental attention states based on cerebral activity is a powerful way to grasp a better understanding of ourselves in a very broad way, and to provide our digital environment the information that was missing to dramatically improve the quality of our interactions.