MindDrawPlay
from Leipzig & Rostov with Love :)
MindDrawPlay – is a research project of interactive educational audio-visual art, representing translation of brain waves and heart beat to visual and sound spaces, flows and controls.
It has been grown on a base of research work in the direction of
Brain-Computer Interfaces and heart activity analysis. The project integrates several applications and biosignals recording devices.
The story
MindDrawPlay has started in 2017 as a hobby project during years of my PhD in computational neuroscience in Max Planck Institute for Human Cognitive and Brain Sciences in Leipzig.
I couldn't imagine how far and deep it will go in my life, when I got MindWave mobile EEG device just to play, and was curious - how can I use it in an artistic way, connecting brain waves patterns to sound samples and drawing.
Motivation
The main idea is - to represent an inner state of a person
in dynamic interactive ways with audio-visual biogenerative models,
to analyze brain and heart responses and to adjust the interaction between.
The key aspects:
technical and artistic
1) to process the brain and the heart activity with mobile devices
2) to extract valuable quantitative metrics of mental and emotional states
3) to create audio-visual real-time models incorporating obtained metrics
4) to adjust parameters of the models according to changes of the state
5) to analyze recorded biosignals data to study biofeedback experience
social and media
6) to exhibit the project and collect feedback from different people
7) to make the project available as open-source / low-cost product
8) to bring mobile biotechnology from labs and industry to daily experience
9) to expand people communication by augmentation with biosignals
ontological
10) to learn more about the brain and the heart, to explore the mindspace
Overview and examples
brief overview of the project
with BrainBit (EEG) + Callibri (ECG)
Biogenerative Sphere with adaptive Attention levels, Statistics
Integration with neuroart:
neuroadaptive animation flow,
Revelation of Dreams collection
Biogenerative Cubes with adaptive adjustment of Attention estimation
Biogenerative Neuron (particlesGPU)
Brain & heart responses on music
with adaptive Attention modulation
on audio spectra visualization
3 examples of biogenerative visuals,
markers and Statistics module
Biogenerative Brain (as points cloud)
Integration with neuroart:
neurointeractive images change,
MindMembrane collection
NeuroMembrane as base universal module for biosignals integration:
TouchDesigner network with audio spectrum visualization
early MindDrawPlay application with brain waves streaming
to TouchDesigner vis OSC protocol (collaged with 3D brain from Resolume)
MindDrawPlay (2025, v.3.0)
Bands Utility (processing of EEG/PPG signals)
Implementation
The project includes several Qt / C++ applications dedicated to obtaining and processing of brain waves (EEG) and heart beat (ECG/PPG) signals from different mobile neurointerfaces (BrainBit / NeiryHeadband / Muse2) and mobile cardiodevice Callibri.
Core part of mathematics related to the signal processing as well as an audio translations of biosignals are also performed in these applications. The acquired biosignals and extracted metrics are then streamed to TouchDesigner (TD) via OSC protocol.
QRS Utility (processing of ECG signals)
Definitions: audio-visual translation of biosignals
Audio-visual translation of biosignals means video, where sound and visuals have direct connection to biological signals recorded from a person in forms of electrical activity. Such video is an interactive to person inner state - the presented objects are changing according to changes in the brain and heart activity. Therefore, a biofeedback relation is established and presented in audio-visual form.
Definitions: biosignals
Biosignals are brain waves and heart beat, which are recorded from a person via special devices:
- Brain waves in form of Electroencephalogram (EEG) - Heart beat in form of Electrocardiogram (ECG)
Biosignals Metrics and its utilization
There are metrics derived from the brain and the heart activity based on EEG and ECG/PPG signals.
These metrics are used in two ways for creation of audio-visual translations:
1) biofeedback parameters for the modulation of generative visual models and neuroart animation
2) biofeedback parameters for the modulation of an audio translation module
For the brain activity, besides, the relative expression of different bands (Theta, Alpha, Beta, Gamma, %),
there are also mental levels (Attention, Meditation, Cognitive Control, Cognitive Load, %).
All these metrics reflect mental state of a person - of being more or less focused, cognitively loaded or calm. There is no one established way for estimation of Attention and Meditation levels, it can vary for different fields and application. Overall, Beta waves is clearly associated with Attention, whereas Alpha and Theta waves – with Meditation, but other waves also have a significant impact on the estimation of these mental levels and states. Cognitive Control is a metric based on relation of Beta/Theta waves, whereas Cognitive Load is based on Theta/Alpha relation.
For the heart activity, there are currently 2 metrics - Heart Rate and Pressure Index.
Heart Rate estimation is based on detection of QRS complexes from ECG signals,
which is implemented with a modification of the Pan–Tompkins algorithm.
Pressure Index is a specific metric (also known as Baevsky's stress index),
which is based on the analysis of heart rate variability.
Audio translation module
There are 6 sound samples in Bands Utility playing in loop for all related brain activity metrics.
For Attention related sound - its pitch is changing accordingly to changes of Attention level.
When you are more focused - the tone becomes higher, and when more relaxed - the tone is lower.
For other sounds - their volume is changing in direct relation to levels. Additionally, there are inner parameters to amplify volume of Gamma (which usually has very low level) and to decrease
volume of Alpha and Beta.
For the heart beat - on each detected beat from ECG signal in QRS Utility - a sound of heart beat plays.
Every audio translation reflects brain and heart state and its responses on visuals during recording.
BrainBit mobile neurointerface
Callibri multi-modal biodevice
Artifacts processing
The substantial update in MDP v.3.0 is in improvement of the signal processing part due to implementation of artifacts correction with rASR (Riemannian Modification of Artifact Subspace Reconstruction) approach. This technique allows to eliminate efficiently short artifacted intervals by interpolation based on signal properties and, therefore, provides with more robust and better quality signal.
Below is a demonstration how it works with BrainBit neurointerface.
Statistics module
Statistics module provides with several ways for analysis of recorded biosignals,
the key aspect is an option to see both the brain and the heart metrics changes for choosen time period.
Besides, there is an option for time markers and streaming of them directly from TouchDesigner.
Integration with TouchDesigner (TD)
The main visual and interface parts are implemented inside TD projects.
TD serves as a base environment for variety of biogenerative translations
and integration of the biosignals with neuroart (Stable Diffusion images and animation).
The core visual representation module in TD is NeuroMembrane.
NeuroMembrane - visual representation of biosignals (brain and heart activity) metrics in form of animated circular plots, which show dynamics of metrics changes for specific time interval (usually 1 minute). NeuroMembrane emphasizes the most expressed brain wave at the moment by visual trace effects. Membrane part is used in the title, because the circular plots response like membrane to the changes in brain waves expression levels - stretch and compress. Examples: Gamma (yellow) on the right image,
Theta (cyan) on the video on center (high Theta flow) with additional particles animation effect.
NeuroMembrane is the base universal module for biosignals integration.
Besides visualization part it also includes an adaptive Attention estimation mechanism.
Overview of TouchDesinger networks
The current main TD network and other added so far modules are presented below.
the main network
NeuroMembrane network
NeuroMembrane network:
Theta, Alpha, Beta, Gamma, Heart Rate, Pressure Index visualization
MindScapes network
MindCubes network
MindNeuron network
MindSphere network
v.2.0 (2023)
This video is an overview of 2nd version of MindDrawPlay.
At that time, the project started to include several
Qt / C++ applications dedicated to processing of both brain waves (EEG) and heart beat (ECG) signals from BrainBit and Callibri devices. Most of mathematics as well as audio translations of biosignals were done in these applications. The key change was made by incorporating TouchDesigner as the main visual interface combined with biogenerative animations implemented there.
Audio translation
There are 2 sound samples playing in loops for Attention and Meditation, Attention level modulates pitch of the sample, Meditation level - volume of the sample, for Heart Beat there is a single play of corresponded sound sample on each detected beat.
v.1.0 (2017-2020)
This video is an overview of 1st version of MindDrawPlay made as a single Qt / C++ application obtaining EEG signals from MindWave / MyndPlay / BrainLink Lite devices.
This application is not much in usage (only some modules).
Combining technology advances – such as mobile EEG devices and musical knowledge – such as pentatonic scales (tones from hang and tank drums), it allows to see, to hear the brain activity represented by set of sounds and to use brain waves as a brush for drawing, as parameters for image filtering, attention modulated pictures flows, in “puzzle gathering”, “find the same” and “go through” games.
Examples
MindDrawPlay examples can be divided on 3 groups:
1) biogenerative animations controlled by Attention / Meditation / waves expression
2) animations based on neuroart images, NFT collections:
MindMembrane, Revelations of Dreams, DreamFlowMind
3) live performances as a VJ or with with dancing artists
Examples below are presented mostly in chronological way -
from the first simple to more new and comprehensive.
[ update in progress ]
Attention modulates size and surface distortion of the globe.
Physarum generative particles model: Attention level modulates speed and length of particles movement, Meditation - angle of the movement.
Attention and Meditation levels modulate brush size, brush style (blurring degree) and speed of the canvas rotation.
These examples based on Nvidia Flex particles model.
MindFlex 1: Attention level modulates cohesion parameter of the particles movement.
MindFlex 2 & 3: Attention level modulates border area for the particles movement, several other parameters are linked with midi controller.
based on ParticlesGPU TouchDesigner node: Attention level modulates through sound spectrum scaling parameter size of the particles, Meditation level - Drag parameter ~ particles spread.
based on ParticlesGPU TD node: Attention level
modulates speed of the particles movement.
Brain activity and sound control both structure and coloring of the form: audio level modulates amplitude of waves, Attention modulates size of particles and waves period, bloom intensity level, audio reactivity delay, Meditation - glow level, music: Suduaya
Heart Brain Dance
Audio-visual translation of brain waves and heart beat in a dancing performance. The music of heart and brain is changing depending on dance movements and Attention level, projections of heart and brain images and waves are animated on heart beats.
Collaboration with DomTanzTheater:
choreography: Tamara Ryazantseva,
brain: Alisa wearing neurointerface,
heart: Pavel wearing cardiodevice.
Attention and Meditation levels modulate landscapes structure, sound level - saturation of landscape color.
Attention level modulates structure,
Meditation and sound levels - colors.
Attention and sound levels modulate
amplitude of generative waves.
Modification: Audio and Attention reactive particles: sound and Attention levels modulate size of particles and camera changes, Attention modulates area of particles, its traces length and spread of particles, music: Klaada
Based on Particles node in TD: Attention modulates Axial force and Twist strength, Meditation - Vortex force