ABOUT
This is my capstone project for the master program in Human Centered Design and Engineering (HCDE) at University of Washington. We formed a team of researcher, designer and engineer to reinvent the candy bowl inside the HCDE main office. We were able to design a product/service at a system level and created a physical and digital interactive prototype.
ROLE
Research analysis, interaction/visual/industrial design, video production
Everyone has come across a candy bowl that is either empty or nearly empty, with only a few remaining pieces of discarded candy at the bottom. This experience is always disappointing, leaving you without the sweet treat you were craving.
Learn from Experts: Secondary Research
In order to better understand the space, we read 10 academic pieces on gamification, motivation, and free food.
A Quick & Dirty Prototype was Born
Nothing speaks louder than a prototype. We decided to test all the important thoughts through a quick and dirty prototype. Using cardboard and laser cutter, we were able to quickly build out a physical prototype to imitate the interactions (2-day work for 3 people). For the digital part, we built a simple website to host our 3 digital challenges to test with users (1-day work for 1 person).
And It's Interactive
Something didn't Work
(One participant was playing with the prototype. I designed the posters on the wall.)
(I built the 3D model and prepared the laser-cutting file in Rhino. We created almost every components (including gears) by ourselves.)
(The video summarizes our process starting from ideation. You can see documentary pictures of us making the prototype.)
Read another case study
THE NEW CANDY BOWL
To change people's behavior of picking candy, and create fun and collaborative atmosphere in the department.
TEAM
Lingshuang(Liz) Chen, Interaction & Industrial Designer
Chen Ye, Design Technologist
Ankur Agrawal, Design Technologist
Allie Deford, UX Researcher
TIME
01/2016 - 03/2016 (10 weeks)
IMPACT
It inspired me to think about how a product could change people's behavior, their relationships with others and the enviornment around it. It was the most complex prototype I ever built.
THE PROBLEM
THE PROCESS
or, Back to Work
The old candy bowl sitting on the HCDE front desk: poor mints!
THE GOAL
After the initial research, we identified three main goals that address the existing problems around the bowl.
Encourage Collective Responsibility for Candy Refilling
Enhance Diverse & Sustainable Candy Consumption
Enrich Fun & Collaborative Social Interactions
THE SOLUTION
The video summarizes our research findings, and the proposed solution in context.
MOTIVATION
When changing behavior, it is important to consider how users are being motivated to use the system being redesigned. Extrinsic motivation, or motivation where users are motivated by something outside themselves, can be used to initially motivate. However, for a system to be successful, it must ultimately rely on intrinsic motivation, or motivation that comes from within the user.
GAMIFICATION
When used properly, gamification can be used to make a system interesting and to keep users coming back over time. However, it must be carefully thought out. Our research pointed us to successful gamification frameworks that encouraged us to think about how we were motivating users and how to keep them interested in the system over an extended period of time.
FREE FOOD
Free food can be used as a motivator to get those who wouldn't normally interact with each other to communicate, which can lead to unlikely but effective collaborations. In a case study looking at why Google gives its employees free food, we learned that one of the main reasons is to give those who work near each other but on separate tasks a chance to get communicate with one another, which can often lead to interesting collaborations across work groups.
Learn from Candy Bowl Managers
We interviewed 3 candy bowl managers about their experience maintaining the candy bowl as well as to get a better understanding of the culture and common problems surrounding candy bowls. This allowed us to quickly understand the culture around the candy bowl without having to perform hours of observation.
With all the qualitative data we had, I led the team through the data analysis and sythesis session. We highlighted the key quotes, mapped them out on the write board, and identified their relationship with each other with affinity diagrams. As we analyzed deeper from people's actions, to their feelings and motivations, three themes began to emerge: Ownership and Procession, Logistics and Social Influence. The three themes were further elaborated as three insights.
The initial analysis happened on white board, the google sheet includeds the results of our findings
3 themes emerged from the analysis, which were elaborated as 3 insights and opportunities
The initial analysis happened on white board, the google sheet includeds the results of our findings
What We Learned: 3 Insights & Opportunities
Free candy in the office provides a place for people to interact with others. However, some less positive social interactions happen when people sitting near the bowl are frequently interrupted by candy takers who feel the need to justify their actions.
“I sit next to the candy bowl, so people talk to me when they‘re taking candy. They comment on how many they’re taking, especially if they’re taking more than one”
- p1
Enrich positive social interactions among HCDE staff, faculty and students.
INSIGHT
QUOTE
OPPORTUNITY
The department thinks the bowl belongs to everyone in the office, with the burden of buying and refilling candy being shared by everyone in the office. But people only take candy from it, and complain to the office manager when the bowl is empty.
"The responsibility, I have decided, I am disowning that responsibility. Now it’s like whoever decides or remembers to pick up some candy, they can do it. I will go get some, I just don’t do it all the time, I am trying to empower other people to bring in the candy." –p3
Encourage all staff in the office to refill the candy bowl, and foster the collective sense of responsibility.
The bowl is usually filled with a variety of candy at first. But some types of candy run out more quickly than others because people tend to pick the same types of candy. This leaves the less desirable ones at the bottom of the bowl for a long time before they are finally selected.
"It’s all babe ruth and 3 musketeers [left in the bowl], which happen to have the same color packaging." "They pick something that they can see, but something they like" -p2
Make the candy bowl consumption sustainable, in both speed of consumption and diversity of choice.
Generate Solutions via Idea Jam
Given the 3 key insights, I organized and led the idea jam section in order to come up with as many ideas as possible to solve the problems. During the idea jam, each of us was asked to draw quick sketches of an idea around one aspect of a problem. Each round takes 2 minutes, and then we moved on to the next aspect. After all aspects of a problem were finished, we presented our ideas to the group and talked about what they *liked* about the idea. We conducted the activities for all 3 problems.
Idea jam works better than traditional brainstorming because it urges us to put our independent thoughts into visuals without interference, and clarifies the problems by deconstructing them. In 2 hours, 3 of us generated more than 20 distinct ideas to solve the problems from different aspects. We spent another 4 hours documenting and fleshing out the ideas with user scenarios and interaction flows.
Finally, by leveraging how well the solution addresses the 3 problems and the complexity of engineering, we narrowed down to one idea with a single physical form, and 3 possible digital interactions.
The concept is a candy bowl with integrated modules catering to the needs of 3 different user groups.
2 HOURS
3 PEOPLE
20+ IDEAS
IDLER
RUNNER
GIVER
Those who would like to take a break and play (providing fun & engaging experience)
Those who are busy and just want to grab some candy and go (providing quick ways to get candy)
Those who would like to share good candies to others, either obtained after playing or brought from elsewhere
(enabling & encouraging sharing behavior)
3 Types of Users & Solutions
We created 3 possible digital interactions for the final system, which we called Challenges.
The objective of this Memory Testing game is to flip over pairs of cards and match all the turned-down cards in as few moves as possible. We choose this game for its simplicity, fun, and good fit for screens with small size. The choices of the pictures can add another layer of fun, as we can use pictures related to the people in the department.
The idea behind Co-drawing is to encourage social interaction inside the department by providing a way for people to have fun drawing together. The users will be provided with incomplete drawings started by previous users, and continue drawing for 30 seconds based upon the partially finished drawing. This challenge was chosen for it’s attention to collaboration and group interaction.
Sliding puzzle is a game that challenges the players to move pieces which are stuck in a fixed frame around in the right sequence to recreate the original picture. We are aware that this challenge is more complex than the previous ones. We choose it to help us determine how much effort users will willing to put forth to earn candy.
The objective of this Memory Testing game is to flip over pairs of cards and match all the turned-down cards in as few moves as possible. We choose this game for its simplicity, fun, and good fit for screens with small size. The choices of the pictures can add another layer of fun, as we can use pictures related to the people in the department.
How We Tested the Design
There were 2 main goals for our testing: one was to test if people could correctly understand the system, and interact with it in ways it was designed; another was to test users' emotional reactions to different challenges.
Putting users and the product/interface in their place
Because people's understanding and behavior is affected by contexts and situations, we decided to test the product in its real scenario. The prototype was placed in the hall way where students, staff, and faculty would pass by.
Lack of framework to test emotion
We ended up testing the users with traditional methods used in usability studies, in which interactions are formalized in terms of tasks, goals and efficiency. I reviewed papers on the topic, and found some tools for emotion report and evalution, such as PrEmo. However, it was not immediately available(free) for us. The only way we could possibly evaluate is through observation of facial expression and self-reported words describing their feelings.
Our initial plan was to leave the prototype unattended with automated pop-up questionnaire at last. But the result of the pilot test revealed the inappropriate time setting of the questionnaire, and the lack of qualitative data. We then changed it to a researcher standing with the prototype, with invited participants who walked by. They were directed to explore the prototype, and after they were done, the researcher asked them about their experience using the prototype and for any feedback they might have. Ultimately, we had 7 participants complete our study in 1 hour. Their feedback uncovered several key usability issues.
The purpose of the charity box and the sugar rush box is unclear.
Change: Simplifies the system while remains the feature by combining the two boxes into one. The purpose of the box becomes ambiguous, but accomodating.
It is unclear the the drawer contains different candy.
Change: Making the drawer with better quality candies transparent and prominent.
The sliding game is too time consuming.
Change: Providing only challenges that take less than 30 seconds to complete, and don't require considerable intellectual effort.
The buttons are hard to select due to their size.
Change: Change the height of the button to 68px, and the size of the font 30+.
Some labels are confusing.
Change: Modify the language from "sugar rush" to "grab and go".
In a nutshell, our previous design is too complex. It demands 3 kinds of interactions between users and the system, and attempts to establish one correct understanding of the system for all kinds of users. However, this couldn't work because of the diversity of people's understanding and the situations. Instead, it should accommodate the varying perspectives of users, and make it really simple what's the *one* thing that users should interact with.
Iteration, Iteration, Iteration: Second Prototype
We then incorporated those changes into a new design and a final prototype. We primarily used laser cutting on acrylic sheets and birch boards for all of the components.
The final prototype is a complete system with three individual parts: the physical design, the electronics system design, and the digital interaction design.
We designed and implemented a control system that would run the mechanics of the prototype. The core controller is Arduino Uno, which talks to different sensors and modules. The components work together as the following chart shows.
Electronics System that Made it Smart & Responsive
Digital Challenges that Made it Engaging & Fun
Digital Interaction of Sweet Treat primarily includes the challenges that users must complete in order to earn higher quality candies.
Collaboration not Competing
Without careful thought, it's easy to be trapped into the stereotype of gamification, which is adding leader boards, points and badges to the system. We deliberately avoid doing so because it conflicts with our goal of encouraging sharing and collaboration.
Sense of Community
Given the limitation of time, we were only able to implement one game in the final prototype. The previous Memory Testing game was selected. I iterated on the interface (color, font, and content) to convey the sense of community and fun.
Long Wow Experience
To simplify the usage flow, we decided to only provide the user one challenge at a time. But the challenge will rotate every week, which helps keep the experience fresh for an extended amount of time.
Main Features
Interactive Challenge
In order to make candy consumption sustainable, we needed to limit the user's ability to get to the good candy. To do this, we designed a digital challenge which users must complete in order to earn the candy. To better show off what they would earn by completing the challenge, we made the drawer containing the better candy transparent; this will encourage them to play the challenge in order to earn the better reward.
This system enhances sustainable candy consumption by limiting the number of types of candy that users can choose from. Social elements added in the challenge, such as words encouraging candy sharing and refilling, department branded pictures, and co-creation process, will also promote a collaborative atmosphere in the office centered around the candy machine.
Grab and Go
In order to meet the needs of those who just need a quick fix, we designed a tray at the top of the machine where users can grab a piece of candy without completing a challenge. This section will initially be filled with less desirable and cheaper candy than the candy in the drawer, but can be added to by users who want to share extra candy they earned by playing games, as encouraged by prompts on the screen.
Moreover, we believe sharing candy in this manner will increase a positive environment and promote a cooperative and helping nature in the office.
Auto Refilling
We designed the refilling drum, which is divided into 4 sections, to be the main candy storage. It can refill the drawer 4 times by rotating itself when the drawer is almost empty. As soon as the IR sensors detects the drawer is empty or almost empty, it sends a signal to controller to rotate the drum, which refills the drawer.
We designed this because we heard that refilling the candy bowl is a hassle to candy bowl managers. So, we designed Sweet Treat to have a large candy storage area and the ability to refill itself, which reduces the frequency of refilling.
Encourage Collective Refilling
While the drum is able to refill the drawer several times, occasionally the drum needs to be refilled. When the drum is empty, we show prompts on the screen to encourage people to refill it from the refill spout. We put two LEDs near the refilling entrance to indicate the drum is full or not to help users know when the drum needs refilling.
We learned from our research in HCDE office that the burden of filling the bowl should belong to everyone who uses it. This feature encourages collective responsibility by making Sweet Treat easy and fun to fill.
My Reflection & Learning
I learned a lot from this experience. I'm also suprised and proud of what we've accomplished at last. However, there are still some things that I would do or suggest the team do differently.
Practical Plan of Time and Resources
I would leave more time for building and testing. Completing the final prototype was very stressful, and some of that stress could have been reduced and the electronics could have been firmed up if we had more time. We also should started purchasing the materials needed for prototyping earlier with practical estimate of the volume.
Real-world Practice is Too Complex for a Prototype
While we designed the system under an ideal scenario, the prototype itself raised up more issues when putting in a real-world. Because it's made of birch board, it is actually easily breakable. The mechanism we designed and made was also not durable enough. Some people put their fingers in it trying to stop a moving drawer when it's closing, which broke the wooden gears easily. The prototype has to remain as a prototype, rather than a functioning product.
Try New Methods for Evaluation
Our testing for the first prototype was primarily focused on usability issues inside the screen. With more time for preparation, we could have tested the emotion, and the embodied interaction happens outside the screen. Originally we had the creative plan to test the effectiveness of the candy bowl in controlling assumption, but cancelled because of obstables in execution.
You may try out the interactive prototype below for the digital component (note: the digital interface has nothing to do with an iOS platform).