We have made an agreement with Museo MARCO to develop and present an event in this venue (estimated quorum: 600 people) where we will implement this system and a couple of others which were requested (also involving Neuro-Lighting and Neuro-Sound) in November 2018. Join us!
We contacted Baby-Robot Studio and Dakota Studio Bar because they are friends of us who are involved in the local artistic scene and we thought they might help. We offered them to prove the NeuroLS System experience and decide if it could benefit their performances. Baby-Robot Studio refused but Dakota Studio Bar accepted provided we did a proper documentation of the experience.
Figure 6. Headset + NeuroLS + LED lights integration. A connection was made between our NeuroLS system with LED lights; programed to vary a sequence depending on the predominant wave signal read by a MUSE band headset.
Figure 5. Construction of 3D-printed pieces. a) Lamp built by a 3D-printer from Innovation gym CoWork with PLA filament. b) Material required for the construction of the LED lamp: LED strip lights; Arduino Mega; Jumpers; Protoboard; monster driver. c) Electronic assembly of the devices.
A connection between MUSE headset and PureData software was stablished, after understanding MUSE Tools. The Pd Patch was programmed to plot and sonify raw FFT neural data. (Fig. I)
Fig. I
A subsequent Pd patch was program to relate the alpha, beta & gamma waves to certain colors of our election. (Fig. II)
Figure 4. Stroboscopic LED lamp Render. Visualization made in solid works to be printed with PLA filament. After NeuroLS Software interface converts the brain signal into audiovisual representation format, the LED lamp receive a DMX signal to reproduce a stroboscopic protocol.
The current objective of the NeuroLS System Version 3 is to change the oscillation between two colors depending on the predominant brain wave of each sensor.
In this case we receive the three Absolute Band Powers (Alpha, Beta, & Gamma) for each sensor (four in total).
Deltha and Theta brain waves were omitted since the nature of the headset doesn't allow the read this wave length properly.
After registering this data to arrays we can then determine the predominant brainwave for each sensor.
Fig. I
Once determined, the predominant brainwave is "lit up" by a red rectangle under the brainwave switch, located on the upper left corner (Fig. II). This switch allows to see and change the two stored colors of each brainwave.
The user can then change the two stored colors through a set of three number boxes (RGB), one for each color. These number boxes, located on the bottom (Fig. II), range from 0 to 255. In addition, there is instant feedback showing the color over the number boxes.
Fig. II
Lastly, the waveform switch and time number box, located on the top middle (Fig. II), change the way these two colors oscillate. The waveform switch has five different waveforms:
UP: upwards ramp
DW: downwards ramp
SN: sinewave oscillation
TR: triangle wave (upwards and downwards ramp)
SQ: square wave (instantly change between both colors).
The time number box allows to change the oscillation time in milliseconds.
All this also has an instant feedback, located in the upper right corner (Fig. II) where the oscillation between colors is visible.
We were able to use some of our research lab’s material for free. We got access to 2 Muse headsets which were not being used. We had to make sure they were working properly. We had never used the actual SDK kit because we normally get raw data and do the whole processing on Matlab. We are getting familiar with the Muse proprietary tools.
Figure 3. Pure Data schematic. A) Listening for incoming messages from UDP network. B) Parse OSC packets into PD messages. C) Write absolute band powers of Alpha and Beta brainwaves into an array for audiovisual representation.
Memo Santos
Fig. I
