I led the UX research process including: contexual inquiry and user testing. As project manager, I assisted in the interaction design and graphic design, wrote the final research paper, and presented at the semi-finals and finals during the CHI Student Design Competition where our team won first place.
Joyce Sakata, Mengdi Zhang, Shi Pu, Kritika Versha
Actively maintaining emotional balance and well-being is important to avoid future mental health problems. Motivating people to do positive intentional activities and taking an active role in maintaining their mental health can increase long-term happiness and avoid more serious mental health problems.
Beam is a mobile application that uses random acts of kindness to analyze and improve the happiness and mental health of its users. We used a combination of theory, design, and user research to influence users to become more optimistic – linking the use of the application with mental health and mood improvements.
We began by gathering survey data from 78 college educated students and professionals. We targeted individuals who we believed incurred more stress on a daily basis and would benefit most from our application. Our primary goal was to uncover the types of activities individuals currently do to reduce stress and improve their mental health.
Individuals often talk to others in times of stress or act more passively by either listening to music or doing nothing at all.
30% of our respondents reported that they do not do anything when they are feeling depressed.
Very few interacted with social media or wrote about their feelings in either a blog or journal format to reduce stress
We interviewed six individuals between the ages of 23 and 30. Our primary goal was to discover users’ current outlook on life and what factors promote this outlook. Our team built an affinity wall diagram to analyze the data from our interviews
Individuals experience negative stress everyday.
Engaging in activities reduces stress.
Many are passive and do not do anything to improve their mood.
Individuals do not want to share their personal feelings on social media.
Individuals judge others who share their personal feelings on social media.
we developed personas of our typical users to help guide the design of our low-fidelity prototype. We used our personas to create storyboards to illustrate how our application would be used and help our persona improve his/her individual mental health.
Several comparator sites offer user-based navigation, and have fewer, more clearly labeled global links many users access only a limited portion of himss.org
Comparator websites that used search engines powered by google or bing offered users more advanced search and filtering functions and better search results.
Comparator websites that introduced related content into their sidebars during search and article display provided a better browsing experience for users.
We asked ten users with various skill levels tested the paper prototype by performing a set of tasks. The tasks consisted of selecting a challenge, selecting a mood before and after the challenge, journaling, and reviewing sample data visualizations.
Users enjoyed the cat avatar and seeing its different expressions.
Users preferred simple interfaces with fewer options.
Users liked the ability to do a short self-reflection responses as well as a longer manual entries if desired.
Users liked the data visualizations and the ability to see their progress over time and emphasized the need for diagram clarity.
We conducted a second round of user testing to assess the usability of the application as well as its effectiveness in improving levels of happiness. To do this, we asked five users to perform three challenges throughout the week and another five users to perform five challenges over the course of a week. Users were texted a challenge and asked to write a short self-reflection after the challenge. They were also asked to self assess their mood levels prior to the study as a baseline and to reassess after each challenge.
Users felt good about themselves after completing the challenges given to them. They felt that continuing with the challenges of random acts of kindness every week would bring a positive long-term impact.
Challenges that directly affected an individual were more effective than a more general random act of kindness that did not involve people.
Challenges that were similar to normal activities of our users were less effective than challenges that were outside the norm of our users.
Although users found the self-reflection process useful, they expected them to be tied into the analytics in some way.
During the CHI conference, we presented our poster and design through three rounds of judges. We were selected as one of the four finalists where we presented a more in-depth presentation. After each round, we were asked questions about our process and design.
The application allows the user to select and complete a random act of kindness challenge. Several challenges are randomly selected from a database for the user to choose from. Examples of the types of challenges include: buying a cup of coffee for a stranger, complimenting a co-worker, writing an appreciative note to a friend or loved one, or making lunch for a roommate. The challenges are scoped such that they are simple enough to complete within the day, yet have a direct affect on another person. In most of the described challenges, the user receives immediate feedback from the recipient.
By having a user reflect after completing a random act of kindness, we can foster positive self-reflection. The application tracks short-term and long-term mood improvement changes through self-reporting or a micro-expression capturing process. The user is then presented with data visualization to display his/her progress over time.
Research findings from Fox and Bailenson have shown that users who are presented with virtual representations of their physical-selves are significantly more active than those who did not receive any vicarious reinforcement. By incorporating a cat avatar, we hope to be able to promote the idea of virtual self-modeling in the user and improve the effectiveness of the application.
In the future, we would like to look at context based and pervasive interaction design, where challenges are suggested based on the location of the user. (ie. if the user is at a cafe, it may suggest buying a stranger a cup of coffee; if the user is at school, it may suggest that the user help a student with his/her homework.) Although existing studies have shown short term and long-term benefits related to completing random acts of kindness, our week-long studies were not long enough to see if benefits could be maintained over a longer period of time. Additional user testing would be needed to learn if users who continue to do random acts of kindness will continue to see benefits.