Premise: Creating an object that signals to caregivers when a child is experiencing high levels of emotional distress.
Synopsis:
As someone who spent ten years teaching young people to use art for self expression, I care deeply about the emotional wellbeing of young people. The most vulnerable children may be those who are too young to have the language or maturity to express what they are feeling. Without adult intervention, youth are left struggling to mitigate their emotions on their own. This study utilizes the Arduino Grove Beginner kit and Grove EMG Detector to explore the creation of a technology that would signal a color alert to an adult caregiver about the anxiety state of the child.
The Problem:
The Covid-19 pandemic has exacerbated many existing societal problems. The decline of mental health in children is one byproduct of the pandemic. In a study conducted by Mental Health of America (2020) that measured the causes of depression and anxiety in 327,000 children ages 11 to 17, 80% reported that their feelings were linked to loneliness and isolation, with the youngest age groups reporting “frequent recent thoughts of suicide or hurting themselves.” Additionally, the number of 5 to 11 year olds brought to the emergency room for mental health issues has increased 24% during the pandemic (Einhorn & Sesin, 2020). Time Magazine reports that “ the impact of the pandemic on the young mind is not going to stop until the spread of COVID-19 itself does. For parents and other caregivers, that means mitigating the problem” (Kluger, 2020). Children in distress need adult intervention, however many children are not able to express themselves adequately or ask for help. Thus, I undertook how to create a technology that gauges how children are feeling and can alert parents/caregivers to an emotional crisis.
The Moody Design:
First Concept
My original concept was to create a hat that could track the moods of kids so that the adults in their lives would be able to visually see how they’re feeling and help them cope. The hat would utilize an EEG, since neurophysiological research has shown that EEG signals and emotions are connected (Suhaimi et al., 2020, p. 3). The EEG’s electrode sensors inside the hat would record the brainwave readings based on the child’s mood, the readings would signal an LED alert by lighting the brim of the hat. Different colors would be correlated with different emotions. The EEG headband would be sowed into the inner ring of the hat, minimizing the potential discomfort of having an EEG cap with different nodes and cords.
Second Concept
After researching potential prototype materials, I realized that purchasing an EEG would not fit within the budgetary constraints of this project. Thus, I decided to pursue an EMG which would provide readings on muscle tensions rather than on brainwaves. This idea maintained the goal of the project since there is a correlation between muscle tension and anxiety, with muscles being persistently activated by anxiety (Hazlett et al., 1994). Research has shown “that anxiety is often accompanied by muscle over-activity” (Sainsbury & Gibson, 1954, p. 218) My second concept was to run the EMG wires from the back of the hat to the temple and neck and have those readings signal the LED in the hat. This concept also incorporated a design change: moving the LEDs from the brim of the hat to a smiley face logo on the hat, making it more appealing to a younger child and making it more easily visible to an adult. After visualizing this prototype, I considered that a wired hat wouldn’t be a practical option for a child to wear consistently.
Final Concept
I designed my final concept after considering the potential limitations of my earlier ideas. A large part of why the original Moody design was a hat was due to the initial plan of utilizing an EEG, but after removing that aspect from the project I realized that it made less sense to continue with the hat. The tracking of brainwaves would have allowed me to gauge emotionality in a way that is not possible with the EMG. I therefore decided to focus on measuring and conveying one tangible emotion instead of a spectrum of emotions, which I would not be able to differentiate on the EMG. Since anxiety and stress influence muscle tension this seemed like the best variable to focus on. Anxiety and stress also appear as a recurring theme in mental health discussions about children’s ability to cope with the pandemic and accompanying depression, which allowed me to still maintain my initial line of research. This ultimately led to my focus on a Moody bracelet/band that would maintain a singular green color unless the muscle tension increased beyond the normal resting/calm range, at which point the LED light would turn blue when anxiety/stress is felt.
The Experiments
I began this project with no Arduino or EMG experience. In order to learn how to conduct particular experiments I used Youtube and Seeed Studio. Below is a breakdown of the different experiments I did with the Grove RGB LED Stick and EMG Detector.
RGB LED
How to change the colors on RGB LED Stick
After downloading the Adafruit Neopixel Custom Library, I was able to send commands to my RGB LED. To change the LED color, I had to adjust the variables representing Red, Green, and Blue. To produce different colors I input values ranging from 0 – 255. Zero means off and 255 is the darkest version of the color– for example, I could turn off one color and input different values for the other two, to create a new color. This experiment taught me that using different RGB values I could produce any color on the visible spectrum.
How to alter LED timing
After figuring out how to change the colors, I wanted to understand how to control the timing and duration of the lights. I was able to do this by using “for loops” to light up each pixel in the LED strip one by one, using the delay command to control the timing.
How to make the colors remain static
Since I initially started experimenting with a sample code from the Neopixel Library, I was having trouble maintaining color consistency and was dealing with blinking colors. I realized that I had to modify the code to keep only one color sequence and remove all the delays from the “for loop.”
How to determine which LED light comes on from 1 – 10
Once I realized that each LED in the strip has a particular address (similar to the string index in Python) I was able to understand how to control which ones came on. For example, 0 is the address of the first LED and 9 is the address of the last one. Indexing the individual LEDs gave me more control over lighting and positioning specific LEDs.
EMG
I had to familiarize myself with the EMG as well as the serial monitor and serial plotter in Arduino, as both aid in displaying what kind of readings are being received. This code takes the value of the EMG and plots numbers and graphs. During my EMG experiments, I tried to position the electrodes for optimal readings. However, regardless of positioning the variance remained difficult to detect, even as I picked up objects of different weights to stress the muscle.
Since the variation on the serial plotter was minimal I divided all the readings into two points by setting a threshold of 450 — if the values rose above 450 I defined that as the muscle being tensed. If the values were less than 450 I defined that as the muscle being relaxed. I chose 450 because my EMG was fluctuating around this point when I moved my arm. These experiments were particularly helpful in allowing me to determine values I could later use in connection with the LED.
Conclusions & Future Work
By using knowledge I learned from my earlier LED and EMG tests, I was able to input a code that responded to my movements; remaining green when the muscle was relaxed and turning blue when the muscle was tense. I considered this experiment successful since it aligned with my final concept of mapping muscle tension to create a color signal. As the problem I chose to address is both challenging and important with real world implications, I was glad that I was able to complete tangible research with technologies I was unfamiliar with.
The code below is for the last iteration of the project.
I believe that with further study, this product could be a viable tool for addressing mental health problems in young people by providing adults with a visual cue when a child is experiencing heightened levels of anxiety. Future experiments would require a more advanced EMG detector that would be more sensitive to changes in muscle tension. Furthermore, to carry out the bracelet idea there would need to be technology developed in the future that allows for the EMG to be incorporated into a wristband or bracelet.
References:
Adafruit NeoPixel Überguide. (2013, August 30). Adafruit Learning System. https://learn.adafruit.com/adafruit-neopixel-uberguide/arduino-library-installation
DIY Electromyography (EMG) Part 2 | Detailed Set-up & Testing. (2020, November 7). [Video]. YouTube. https://www.youtube.com/watch?v=PQe9PesahRM
Einhorn, E., & Sesin, C. (2020, December 16). Covid is having a devastating impact on children — and the vaccine won’t fix everything. NBC News. https://www.nbcnews.com/news/education/covid-having-devastating-impact-children-vaccine-won-t-fix-everything-n1251172
HAZLETT, R. L., McLEOD, D. R., & HOEHN-SARIC, R. (1994). Muscle tension in generalized anxiety disorder: Elevated muscle tonus or agitated movement? Psychophysiology, 31(2), 189–195. https://doi.org/10.1111/j.1469-8986.1994.tb01039.x
Kluger, J. (2020, July 23). The Coronavirus Seems to Spare Most Kids From Illness, but Its Effect on Their Mental Health Is Deepening. Time. https://time.com/5870478/children-mental-health-coronavirus/
Sainsbury, P., & Gibson, J. G. (1954). SYMPTOMS OF ANXIETY AND TENSION AND THE ACCOMPANYING PHYSIOLOGICAL CHANGES IN THE MUSCULAR SYSTEM. Journal of Neurology, Neurosurgery & Psychiatry, 17(3), 216–224. https://doi.org/10.1136/jnnp.17.3.216
Suhaimi, N. S., Mountstephens, J., & Teo, J. (2020). EEG-Based Emotion Recognition: A State-of-the-Art Review of Current Trends and Opportunities. Computational Intelligence and Neuroscience, 2020, 1–19. https://doi.org/10.1155/2020/8875426
Zuo, B. (n.d.-a). Grove – EMG Detector – Seeed Wiki. Seeed Studio The IoT Hardware Enabler. https://wiki.seeedstudio.com/Grove-EMG_Detector/
Zuo, B. (n.d.). Grove – RGB LED Stick (10 – WS2813 Mini) – Seeed Wiki. Seeed Studio The IoT Hardware Enabler. https://wiki.seeedstudio.com/Grove-RGB_LED_Stick-10-WS2813_Mini/