An installation in which the individual and group engagement EEG scores are indiated through a type of LED interface. When the engagement score reaches a certain mark (0.7), the LEDs change colour. This project was started during the 2016 Hack the Brain Hackathon in Amsterdam at Waag Society
- Polle van Duuren (Industrial Designer)
- Carlos Moreira (Biomedical Engineer)
- Lydia Muijen (Allround Artist)
- Ehsan Razaghi (Highschool Student)
We are a young group that is mostly just diving into this matter and curious where BCI can take us!
We were interested in using BCI's to measure a group of people in a social context. Through an exploration of possibilities therein we pinned ourselves down on measuring Engagement using [beta/(aplha+theta)] as described in a number of scientic papers (see later).
We envisioned a situation where people in a social group (colleagues, teammembers, etc) could individually be measured for their level of engagement. Also we could measure the engagement of the group as a whole.
Marketability: This could be relevant in a number of situations. During brainstorm sessions it could be interesting to see when certain ideas are actually interesting for everyone, during pitch presentations or lectures it could be interesting to see the general reactions of the audience, or for teambuilding it can be a challenge to achieve specific group goals.
We have build an installation in which the individual and group engagement scores are indiated through a type of LED interface. When the engagement score reaches a certain mark (0.7), the LEDs change colour. Moreover there is one interface that communicates the engagement of the whole group. Here we identified two different approaches: We can add all individual scores together and divide it by the number of individuals [beta1/(aplha1+theta1) + beta2/(aplha2+theta2 .... etc / # individuals]] We can work with the sum of frequency bands and treat all brains together as 1 brain [betasum/(aplhasum+thetasum)]
Artistic value: Interestingly the second approach gives (slightly) different results than the first one. In this way we are treating the group as a whole as a separate entity, rather than the as a number of individuals. This artistic input is what we consider to be unique about our project: we are showing data of the 'groupbrain', which actually does not exist itself. We can dive a bit into philosophy here and discuss what it would then mean if there is such a thing as a separate group consciousness...
We have worked with Muse and wrote (with a lot of help) a code in Python to extract the values of all individuals, as well as the group. We then used Arduino to activate a number of RGB LEDs that change based on the different engagement scores.
Level of maturity: Ideally some more calibration (and personalization) of the engagement score would be implemented, but we have saved that for another time! At least within the capabilities of our team we have challenged ourselves truly in order to get to where we are, so we are proud of our achievement.
Scientific basis of engagement calculation: Berka et al., EEG Correlates of Task Engagement and Mental Workload in Vigilance, Learning, and Memory Tasks (2007) McMahan et al., Evaluating Electroencephalography Engagement Indices During Video Game Play (2015) Pope et al., Biocybernetic system evaluates indices of operator engagement in automated task (1995)
Novelty and scientific value: Although measuring engagement of individuals in this way is not something new, using this to measure group engagement is something we have not come across. Through our installation we try to challenge scientist to consider how translate our current knowledge about individuals to groups, and what implications this would have on the group as entity.
- Presentation PDF (this is a copy of the slides that were presented during the hackathon)
- Arduino code
- Python code
This repository was made by specific group members and uploaded to a temporary environment. This temporary environment has been manually transferred to a a more permanent Hack the Brain Github account. If any of the former team members wish to administer this repository and manage the master branch, then they can contact the current administrator of the repository to gain the proper acces privileges to manage the master branch
