The importance of phase in EEG-based Brain-Computer Interfaces

In the world that we live, we have to react to sensory inputs all the time: the hustle and bustle of the city (traffic lights, cars, bikes…), notifications popping up in our screen, concerts… Taking the traffic light as an example, has ever happened to you that when you are waiting to cross and the light turns green sometimes you react faster or slower?

Why does this happen? And how we can benefit from this? My research is focused on studying the brain when receives visual sensory inputs, and understand how it works so as to use it to improve the interaction with brain-computer interfaces (BCIs).


Figure 1. Real-world stimuli: hustle and bustle of the cities, notifications popping up in our smartphone, demonstrations, live concerts…

Brain rhythms produce perceptual cycles

Neuroscientific evidence has established that brain rhythms play a crucial role in sensory, cognitive, and motor mechanisms. These brain rhythms can be recorded in various frequency bands and at multiple scales (from single-neuron studies to whole-brain techniques such as EEG), and these brain rhythms produce perceptual cycles that fluctuate over time.

This means that there are more or less appropriate phases (or optimal/non-optimal brain states) for the neural processes under consideration. For example, taking the example of the traffic light back, you might see react faster to the green light because your brain is in an optimal state, and you might react slower when it’s in a non-optimal state.

How can we know our brain state and how can we calculate it in real time? Calculating the alpha rhythm in the visual cortex and using a BCI system.

Figure 2. Brain rhythms produce perceptual cycles.

Methods and expected results

The idea is to register real-time EEG data from a person, analyze the alpha phase, divide the alpha cycle into ten different phases (~10ms apart), send visual stimuli to all the phases, and register the reaction times to the stimuli. Then, select the phases with the slowest and fastest reaction times and make a statistical comparison.

We have adopted a study from the 60s in which they used a straightforward and robust method to relate reaction time (RT) with the EEG alpha phase.

Figure 3. An alpha cycle can be divided into 10 phases (~10ms apart). One of these phases might have associated shorter reaction times (RT) and another phase might be related to larger RT.


Possible applications? If it works, we will be able to register real-time brain signals, pre-process it, calculate the alpha phase and present stimuli at the optimal phase in which the subject is able to integrate it and react faster. This would open a wide range of possibilities to optimize the user interaction with brain-computer interfaces like the improvement of the process of sending cues/stimuli, fully-adapted to the person’s state, time-saving, among others.

Figure 4. Scheme of a Brain-Computer Interface (BCI) system applied to phase dynamics.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s