I worked on this project for my graduation thesis during my MFA program in Pennstate. In the midst of the pandemic, the rates of mental health disorders were strikingly higher. The intention behind this research is to understand and communicate why the human mind feels lonely from a scientific and social perspective.The project uses conceptual illustrations to narrate the reasons behind increasings levels of loneliness we experience today, and analyses tweets to study how people talk about topics like mental health and depression.
The world is more connected than ever before, but the number of people experiencing loneliness and struggling with mental health disorders is increasing. With the onset of industrialisation, people have moved out of their communities and severed close ties with their core network groups. As the effects of a global pandemic reached every corner of the world, loneliness has become one of the leading causes of early mortality rates. The time seemed apt to try and understand what was happening to our social ties and how it is affecting us.
Since the topic in question is a deep and complicated one, the main idea was to break down different types of information into different components.
In order to simplify the narrative of this immensely complex topic, I decided to divide the project into three main components:
Component 1: Visual Essay - Data Analytics from Literature sources
Component 2: Visual Essay - Text with moving illustration explaining the science behind loneliness.
Component 3: Analysis of 1000 Tweets with hashtags #depression and #depressed presented in the form of a Data Visualization.
All of these components exist within one UI framework. The user experience was carefully considered to ease the audience into absorbing this information with interactive scrolling and data visualisation techniques.
This component contains the synthesis of information gathered from various literature sources about how the human brain processes lonely emotions. I created a simple storyline to understand how loneliness stems from a biological perspective. The illustrations show a different narrative with the intention of communicating what the text explains using a simpler storyline.
Intention : To view emotions as a natural consequence of the human evolutionary process.
When dealing with heavy topics, authentic conversations can only be created with simple and straight forward stories. Through this project, I learned the value of reading, writing and editing the script multiple times till there was no excess to distract the viewer. After multiple drafts, I solidified the text and decided to introduce the viewer into the topic with data that was current - emphasising on the relevancy and importance of this topic.
The story-boarding phase for this project was slightly different since it did not have the usual components of a story. I started thinking about an alternative narrative that explains the metaphorical meaning behind the text on an individual level.
This was the easiest part. I knew I wanted the information presented to have an evocative and lyrical quality reminiscent of children’s books. This theme is followed throughout the project.
Based on these parameters, the following shows the first sequence of the user experience.
In the second sequence, the user is then presented with some data points, introducing them further into the topic and its relevancy. In this section, each of the data point is presented as a gif.
In this component, the history behind loneliness and and its scientific explanation is explored using a combination and text and illustrations. I experimented by tying interactive illustrations with text as the user scrolls through the website.
Both the text and the illustration show different storylines but explain the same thing — something that a lot of editorial and conceptual illustrations aim to do. I worked with Lottie files to use my illustrations to speak to the text and vice-versa.
The third component of this project was a collaboration between PHD student Pranav Venkit and me.
In order to understand how people use social media to have conversations about mental health, we analyzed tweets with hashtags #depression and #mental health. Using these tweets, a variety of interesting trends were observed: from how people talk about their mental health to emerging topics related to mental health and depression.
All of the data was collected using Natural Language Processing: Tweets were extracted with the hashtags #depression and #mental health for specific periods and then run through different models to understand the sentiment and recurring trends in these topics.
The data was visualized using Flourish. The sub-topics that were explored using this data analysis are as the following:
The timings of the tweets were very revealing. Most of the tweets related to mental health were shared during the late nights and early morning - when people felt the most alone.
Within the mental health framework, what we talked about was very important to understanding the current conversations we were having. Based on the analysis, we divided the data into nine subtopics.
The most number of tweets showed our need to share individual experiences with our communities.
We designed the interface such that each individual tweet can be read upon hover. (Note: we only analysed public tweets, so as to preserve privacy)
The data was thus plotted with respect to emotional categories. This gave a lot of insight into understanding the nuances of the emotional rollercoaster one feels when dealing with mental health issues. The data framework presented was made interactive - the user can click on different classes to understand the distribution of tweets better.
Using the data points categorized into emotions, we further sub categorized them with respect to social dimensions and belonging. Hovering over the colored dots will reveal the exact tweet.
Understand 'how' people talked was also important to identify the strategies people used to talk about while feeling vulnerable. For example, it was very interesting to note the use of satire and sarcasm while talking about one's own emotions.
The interactive framework allows you to see how the tweets group and regroup depending on tone of language.
In the final analysis, tweets were studied with respect to sentiment values using AI and NLP libraries trained on social media data.
75% of the tweets were given a negative sentiment score by the AI model.
The final component in the project was the interactive dashboard where users can input values and tags to categorize the data. This provides an exploratory approach to finding information allowing freedom to navigate the data set by themselves.
This was my final project presented as the graduation thesis during my MFA education in Graphic Design at Penn State University. This project explores the intricacies of mental health and the importance of having more conversations that approach the problem in a scientific and pragmatic way. This project allowed me to experiment with several different techniques to use data to craft and communicate a compelling story.
Along with being able to play with interactive story-telling and narrative techniques using new media, this project helped me learn how to combine multiple components in a project that work together as a system.
