The word robot means different things to different people. In popular culture, robots have been presented as hyper-intelligent killers, loveable goofs, and everything in between. There is no standard definition and that can make it hard to design a robot, especially one that will interact with people on an intimate level.
Dr. Elin Björling and Dr. Emma Rose are working in collaboration with faculty and students on both the Tacoma and Seattle campuses of the University of Washington to develop a robot that will measure moods and stress in teens. Björling studies stress in teens and uses a technique called ecological momentary assessment [EMA] to gather data.
The idea behind EMA is to capture feedback about an experience in real-time. “Stress fluctuates throughout the day, throughout the week, there are highs and lows, there are extremes and that’s the kind of data that makes sense to capture as it’s happening,” said Björling.
EMA data is traditionally collected by answering questions on paper, cell phone or tablet. The process is straightforward if not necessarily engaging. “If we’re going to ask them [teens] to keep doing this again and again then it’s got to be interesting.”
Dr. Rose received her Ph.D. in human-centered design and engineering from UW in Seattle. A major component of this discipline is finding out what people need from a particular design. “It’s very easy to project your own wants and needs onto someone else,” said Rose. “We want to make the process participatory by including people in the design of the technologies they will be using.”
Björling and Rose enlisted the help of UW master’s student Kristine Kohlhepp to build a prototype robot. Kohlhepp graduates in June with a degree in human-centered design and engineering.
The robot Kohlhepp created—EMAR—is more WALL-E than Ultron. EMAR is short, boxy, with big round eyes and a hand drawn smile. He or she (EMAR has no assigned gender) has a touch screen and a child-like voice.
EMAR seems friendly and that’s the point. The eventual goal is to have EMAR live in a high school, interact with students, and collect data. “We want students to form a relationship with EMAR, so we want them to feel like EMAR is listening and understanding what they’re saying,” said Kohlepp.
Björling, Rose, and Kohlhepp have held multiple user-designed sessions with teens and tweens to find out what they would like to see in future versions of EMAR. “We get the basics out of them,” said Björling. “We want to know what kind of robots they like and don’t like.”
One common piece of feedback is that students want EMAR to help in some way. What that looks like is still in the future. The robot’s current design includes a standard greeting and question. EMAR asks students to rate their current stress levels by touching a meter on the robot’s screen. EMAR responds with a statement that acknowledges how the student feels.
The hope with EMAR is that students will feel comfortable enough to talk about what’s going on in their lives. “There are a lot of adolescents that don’t share how they are doing with anyone—and that’s never good,” said Björling.
Using a robot to collect information raises privacy concerns. One way to address this issue is by aggregating the data which would keep it anonymous. School administrators could use these records to get a pulse on their student body as a whole. “I look at it as a measurement device,” said Björling. “It’s an assessment tool that can be used to find out if a particular intervention is working.”
A real-world example where EMAR could be of value is with a student suicide. “My mother is a teacher and, tragically, one of the students in her school committed suicide this year,” said Kohlhepp. “After this tragedy, her school made many efforts to monitor and support students as they coped. It’s situations like these where I believe EMAR’s ability to assess the mental health of the student population can add real value.”