Paw'd Defiance

Using a Robot to Measure Stress in Teens

May 21, 2019 Season 1 Episode 10
Paw'd Defiance
Using a Robot to Measure Stress in Teens
Chapters
Paw'd Defiance
Using a Robot to Measure Stress in Teens
May 21, 2019 Season 1 Episode 10
UW Tacoma Assistant Professor Emma Rose and UW Tacoma student Christina Nelson.
The process of designing and building a robot that can measure stress in teens.
Show Notes Transcript

UW Tacoma Assistant Professor and UW Tacoma student Christina Nelson discuss Project EMAR, a social robot designed to help address teen stress. Project EMAR (Ecological Momentary Assessment Robot) is a cross campus partnership between UW Tacoma and UW Seattle. Rose and Nelson will talk about why they decided to use a robot to help with teen stress and how they are involving teens in the design process. 




Speaker 1:
0:00
I feel like I bring to the field are, and what I'm passionate about is this idea of making sure that people are included fully in the process and that we're not making decisions on behalf of people, but we are facilitating and opening a space where people can move into that place and see themselves as designers
Speaker 2:
0:22
from UDaB Tacoma. This is pot defined welcome to plot defines where we don't lecture, but we do educate. I'm your host Maria [inaudible] today on the pod teen stress and robots with UTIP Tacoma system professor Emma Rose. And you'd have T senior Christina Nelson Rosen and team of researchers at the university of Washington are developing a robot that can measure stress in teams. We'll talk about the project including house students from area high schools are helping shape the design.
Speaker 1:
0:54
So today we have Dr. Emma Rose and Christina Nelson. Can you briefly introduce yourselves? Yeah, I'm Christina Nelson. I was a writing studies major at UDaB Tacoma and I graduated last quarter and I am dr Emma MROs. I'm an assistant professor at UDaB Tacoma. Um, and my expertise is in human centered design. So what exactly is a project? EMR. So project Emaar is a social robot designed to help better understand and intervene in teen stress. So EMR stands for ecological momentary assessment robot. Ecological momentary assessment is a technique that we use in a variety of disciplines to try and understand how people are feeling right here and right now. So the ecological is where is something happening. So if we wanted to talk about our stress right now or like how stressed are you right now, right here in this booth, right? Versus N momentary stands for the time in which something is happening and assessment that's about capturing the data.
Speaker 1:
1:59
So the name of EMR represents what it is trying to do, which is to better understand, understand teen stress. So let's imagine we're in a public high school. That's where EMR is designed to live. And the reason for that is because teens face more stress than any other age group. And oftentimes that stress is happening in the school environment. So we wanted to design a robot that would be able to better understand and intervene in that stress in school. So it's not like for an educational robot, it's not for our classroom. It would probably live in a public setting like the library. So let's imagine you're a teenager, you're stressed, you walk into your library in your high school and what you might see is a, uh, on top of a table, maybe like about a foot and a half or two feet, a little robot.
Speaker 1:
2:47
Right now it's covered in felt and it's got a cute little face and on its belly it's got a little tablet that you can interact with. And as you walk up and start the interaction as you trigger it by where you're standing or by touching the robot, EMR might wake up out of it state and say, Hey, how are you doing today? Or would you like to tell me about your stress? And at that point, the interactions are a series of you touching the robot and different things on the, on its belly. And the robot will respond with different facial expressions and it'll also speak to you and ask you questions or give your responses. So let's say you indicate that you're pretty stressed by filling out a slider on the belly of the robot. Imara might respond and say, Oh, I'm sorry to hear that you feel that way.
Speaker 1:
3:35
Or might just say, well, thank you for telling me. I appreciate knowing how you feel. So it's some pretty simple interactions in terms of touchscreen and verbal responses and verbal questions. Um, but that's how a team might interact with EMR. So why a robot? Why a robot? Great question. So there've been lots of ways that people have been looking at stress. So we all know that stress is on the rise probably individually, that we feel that stress in our culture in the United States is on the rise. And teens in particular are facing a really big spike in stress levels. And as scientists, we always want to try and understand the world better. And so we're often thinking about measurements and tools. So I have a colleague on this project who was at UDaB Seattle. Her name is dr Ellen beer Lang, and she's been studying teens and stress for years.
Speaker 1:
4:28
And she's used a variety of ways to do that, like mobile using a mobile phone or using another way to just try to understand what stress teens are facing. And what she had noticed over time was that, um, after a while some of these data collection tools that we use, like in mobile phones or tablets, they no longer are interesting and therefore teens or any population stops responding. And so it was actually in the back room of Metro, the coffee house on campus where she and I had this conversation that I'll never forget where she was talking about, well, what if we could have like a robot interact with the teens and her expertise is in mental health and stress. And she's like, is that the craziest thing you've ever heard? Is that like a really odd idea? And I was like, no, not at all.
Speaker 1:
5:15
We could see if it works. And as somebody who's trained in design, I'm always like, yeah, let's, let's see if this idea holds true. And so that conversation led to, uh, this project that is now been running for four years. It involves both campuses of Seattle and a U dub Tacoma and it involves, there's been hundreds of students who have worked on the project over the time since we've started. And the, the reason that we wanted to look at a robot for implementation was one, it's never been done. So if you tell a like a scientist or a researcher, like no one's ever looked at teens and robots and stress before, that immediately got us excited and we thought having a novel way to collect stress data could be something that could be engaging more engaging than a phone or a tablet or something or paper.
Speaker 1:
6:08
Um, lots of things that have been done before and, but through the processes we've gone through this. When we started with the idea of like, is this just kind of a weird idea that we'll never kind of go anywhere. We've started to see over time that it really has some legs and that teens are very excited to interact with a social robot, even if it's a modest prototype, which ours are at this moment in time. So we feel like a robot is, it's never been done before. So that's an exciting reason to try it. And also it seems like it is a novel way to try and understand stress from teens. So I want to go back to how this project started. Sure. How did, what was the first step? Well, after that conversation in the back room of Metro, we decided to just try it out and we said, well, you know, we'll devote a little bit of time we think of in the startup world, this would have been called a side project.
Speaker 1:
7:02
We thought we devote a little time to investigating if this could go anywhere. Um, I also have an adjunct appointment in human centered design and engineering at UDaB Seattle. And so I talked to some of my colleagues up there and I was like, do you know anybody who might be interested in working with that with us on this? And there was a graduate student at the time and she really wanted to do more physical computing and physical prototyping. And so alien and I am the graduate student whose name is Christine Cole. She works at Google now. We met and we just talk to her about our idea and the three of us kind of brainstormed and she spent a quarter prototyping and came up with a very modest little robot that would ask a couple of questions that would respond when you asked a question and we prototyped that and started to take it out into the field simultaneously.
Speaker 1:
7:51
Alien and I were doing some informal research with teens to try and understand their world. If this idea had any legs and we would invite them to just hang out and draw pictures, what we call storyboards and design processes and that was the very beginning where it started to snowball. From there. We there, we joke that this project just attracts people over time. People hear about it and they want to get involved. So we just started amassing a team of other students who are interested and eventually we had another professor join our project Maya cheque Mack, who is a professor of robotics and computer science and engineering in the Paul Allen school at UDaB Seattle. And so the three of us started working on it more intently. So from there we have, that was about four years ago, we applied for a couple of different sources of funding.
Speaker 1:
8:41
We got a bunch of rejection along the way, which is also part of, of the job. And um, eventually, uh, two years ago we were surprised and excited to hear that the national science foundation wanted to fund our project. So then we developed a little bit more momentum and started going into different schools and working more intently on both data collection and design. So the, we have involved, like I said, countless, uh, undergraduate and graduate students on both campuses. And then also we have worked directly with Seattle public schools with Tacoma public schools, with chief lash high school in Puyallup, thanks to dr Danica Miller who put us in contact with that school in that community. We also have gone out into informal settings. We, this just this past year, we've been working with the rain incubator here in Tacoma. They have a girls who code club. So we've gone in there to talk to the teens, the girl teens who are in that club about robots and about prototyping.
Speaker 1:
9:39
And so all of our time in the community is a mix of back and forth. Like we are gathering input and design and ideas and trying to understand teens and at the same time we're trying to give something back at the same time. It's a big part of our project. So a lot of the time and a lot of the work that we've done together and Christina has been really instrumental in this, is to go into high schools and into community groups and just teach about design. We use social robotics as the topic, but part of it is just to teach teens about this subject because I went to high school a long time ago, but this was not a focus of academics at all when I was in high school or even when I was doing my undergraduate degree. It was not a field until about 10 or 15 years ago where it really started to take off. So we feel like one of the things that we give back and that we can do while we're in the field is to teach teens about human centered design and about how they themselves can become designers.
Speaker 3:
10:35
Yeah. This actually touches a question that I was gonna ask about what role students play in building Emaar or if they have any, you know, um, input with that. Yeah. So we have quite a large team of student researchers that you'd have to come in YouTube, Seattle. So like at our S our YouTube Tacoma campus, we work with, um, a diverse team of researchers from like computer science, psychology, writing studies, communications, and then we kind of go into the local schools and like we talked about like going into, um, the rain incubator and then working with students. So all of our students kind of engaged, um, high school students in the design process. So we'll go on, we'll teach lessons on like user generated personas, we'll talk about how can we develop robots so that we catered to the needs of diverse groups of teens. And then our students walk them through that process, share their experiences and kind of how they got into the field and that's kind of how we engage in that way.
Speaker 4:
11:30
[inaudible]
Speaker 3:
11:30
we'll get back to this show in just a minute. First I want to mention the global innovation and design lab. Eight up Tacoma. They recently launched initiative will offer workshops as well as credit and non credit courses to a cross section of students, staff, faculty and community members. The global innovation and design lab. We utilize design thinking that paradise is the human experience through a process of ideation, prototyping and testing. These, I think it has been widely embraced by practitioners, researchers and entrepreneurs and activists as a driving force for innovation and transformation for social. Good. For more information about the lab visit the UDub Tacoma website. Okay. Back to the show.
Speaker 4:
12:16
[inaudible]
Speaker 3:
12:17
what are some of the challenges that you face during building Umar?
Speaker 1:
12:22
Boy, quite a few. Um, which is part of the fun and part of the process. Like if it was easy, like no, but someone would have already done it probably. Um, part of the challenge I think is first of all, trying to explain what the robot is and isn't. At the same time when you start to talk to people about robots, some folks are like, Oh, that sounds interesting and exciting. I'm sure there's people listening who think that. And then there's a big part of the population who is like, Oh no, no, robots are gonna like take our jobs or take over the world. Um, so I think just the perception of what robots can and cannot do is something that we immediately sort of step into. And we're in these environments. The robots that we are creating again, are really humble and we are not trying to gather people's information to surveil them.
Speaker 1:
13:14
What it really is at its heart is a way for teens to communicate about their stress in an anonymous way. The data is then aggregated, meaning it's considered altogether. And then the school can have a slice or a picture of what the pulse of the stress is like at the school. And so it helps the school better understand, Oh, is this particular time of the year more stressful than other times? Or if we try to particular intervention, say mindfulness, like does that help with stress levels overall? So one of the, I think challenges is that it's, it's a complex project that we're still trying to figure out how to talk about in a simple way and to let people know that we're engaging in it in an ethical way. And we're really, we're going slow in a lot of ways because we want to make sure that we're making the right decisions and that we're involving our community partners.
Speaker 1:
14:04
So I think explaining it is one of the challenges. I think another challenge is the, the technical side of it. Um, it again, again, it's all of these things that we're developing. Let me step back. A lot of projects and a lot of in the field of social robotics are doing all their work in a lab. So they're developing and prototyping robots and maybe they're bringing people in to interact with these robots, but it's typically done in a lab. And one of the ethics of our project is that we are, and we refer to it as in the wild. So we're not, we're less interested in controlled environments. We're more interested in sort of the messiness of everyday life. And so we want to take our prototypes, even if they're still buggy and not completely working, we want to take them out into the field because those are that conditions that they're going to, to work under.
Speaker 1:
14:52
And so that's a challenge too with like we want to have it more realistic, but that realism control when you have less control, there's just more things that can go wrong. So that's definitely one of the challenges. But one of the themes that we're writing about in a paper right now is how imperfection makes connection. So when we bring like a buggy robot that doesn't say the right thing, like two teens are like laughing about it and then they're laughing at each other and they're having this sort of like funny interaction. And we're like, well, the robots should definitely not be perfect. Because when you see the, the, the sort of, the glitches and the fact that it isn't this infallible thing than it actually brings people together, especially for students as well. You mentioned that you go to high school, students want to be laughing.
Speaker 1:
15:36
They don't want to be tested. Exactly. Exactly. And I think knowing that what we already know about how stressful things are, I mean the fact that we're just there, I often think about like if this project never actually goes anywhere and you know, whether it does or not remains to be seen, but even if we're just out there in the world talking to teens about their stress and, and acknowledging that it's a real thing. And then also getting them to talk to each other about their stress and maybe laughing because a robot didn't behave the way it should have. Like that feels like success in some small ways. Like it's, it's a small thing, but, um, it seems like along the way to creating more connection, which also seems to be counterintuitive about the project when people were like, Oh, again, I hear this mostly from adults, but like, Oh, teens already spend so much time with their faces in their phones. Like why would we bring technology into this situation to try and intervene? But it's, it's a social technology, like they are social robots to have people together and um, convening together and communicating together while they're interacting with this technology. Right. Um, I want to go back to Christina [inaudible].
Speaker 3:
16:48
How did you get involved with this? Yeah, so I had had Dr. Rose during my, uh, I think sophomore year of college and I'd taken her classes on like, um, technical writing and technical communication. And then I kind of merged into the user centered design. And then I got an email about a year later and it was from Dr. Rose and it was about the project and talking about how she wanted to get people involved to teach what we later did as the social robot design challenge. And I was reading the description when I had first entered technical communication. I didn't really ever imagine myself doing any kind of research. Just imagine that I'd be writing a lot of papers, so then when I saw that there was a research opportunity, I was like, I need to take advantage of that and what better way than to do it with a professor that I was really fond of and who I knew was doing really great work in the community and doing community engaged research was really important to me. A, I want to go back to the question where you go to high schools and you let students interact with the robot. What are students thinking about? What are their interactions? What are some of the common CC?
Speaker 1:
17:46
Yeah, that's a good question too. Well, there's a of ways that we engage teens and sometimes we don't take any robots out. We just go out and ask them what they would want in a robot. The particular methodology that we use in human centered design is called participatory design. So it's not the idea of like we are the designers and we're bringing something to you for feedback, but rather we are co-creating this together. So we really come with a very um, humble stance as designers to say we are not the audience we're designing for and they are experts in their lives. And so we really need to create a space for them to feel like that they can have a say in how the technology is designed. So again, we came with like a really blank slate slate and said, how would you design it?
Speaker 1:
18:36
And Christina mentioned the social robot design challenge, which is what we did just last year where we engaged seven different schools in Tacoma and Seattle and had an app. We taught teens about design and we had them prototype their own robots. Now a lot of high school students know about robotics challenges, but they're typically physical robots that you take to a competition and maybe they smash things or they carry things around and they do particular mechanical things. So we were riffing off of that idea of a robotics challenge, but doing it with social robots instead. And so we're less interested in the technology and more interested in the functionality and what teens would want. So they, we went in and taught them about human centered design and taught them about prototyping, about sketching, about getting feedback and doing research. And along the way they built their own robot prototypes and they built them out of humble materials like cardboard boxes or pipe cleaners or for and things that they found are around.
Speaker 1:
19:38
And then all the teams or all the schools sent at least one team to UDaB Seattle. And we had a, uh, a showcase where the teens could show what they designed and why they designed them. That the way that they did and how they talked about their designs became data for our project. It became the resources that we draw on to think about the kinds of designs that we would need to build. So to answer your question about what did teens, what do they do or say in that instance they were like, what does this have to do with me? And how would a robot, how would a robot intervene or address stress? Um, and, but by the time of the end of the competition, they came up with like really, really interesting, innovative ways to address the problem. And they were all different. They were all really, really different.
Speaker 1:
20:27
But there were some common themes across all of their prototypes that again, that we are now thinking of those themes and thinking about how to incorporate it into a functional prototype. The other thing is that when we go back and bring the prototypes that we have now, we get a range of reactions. We have some teens who are really excited and once a hug the team or hug the robot and interact with the robot. And we have others who might not be or seem like that they're really engaged. But then later in the session they might say something about stress or say something about how they're connecting with the robot in kind of a different way. And then we have some that are probably not interested at all and we want to account for and acknowledge that there's going to be a big diversity of people who are interested in interacting with the robot.
Speaker 1:
21:15
Um, Christina, what have you learned from your work? Yeah, I think the biggest thing for me has been the importance of community engaged research. So having a really participatory, transparent and ethical relationship with our community partners has been really important. I think so often in research we see this kind of taking from communities and in our project we've been able to really maintain the fact that this was an ethical project where, and we can take data and collect data from students, but we can also return the fact that we're sharing human centered design with them and we're engaging them in that process so that it feels really meaningful for them as well as us. What are some of the challenges that you faced so far? Yeah. I think in the beginning I really struggled with seeing myself as a researcher. So really feeling like I was going to be part of the design team and being a researcher because I'd never imagined myself in that role.
Speaker 1:
22:01
But then when I was going into schools and I was seeing students and I was watching them design right along with me, they would ask kind of like, how did you get involved in this? And like how can I be part of this community as well? And when I was watching them kind of transform into designers, it was something that I saw myself kind of evolve into as well. So watching them and then reflecting on my own journey really kind of contributed to my own self-growth. Um, dr amaryllis, you have a doctorate in human centered design, right? Can you explain more about that? Sure, yeah. It is. Human centered design is both a philosophy and an approach. And the philosophical side of it is that when you center people in the design process, when you focus on the people that you're designing the technology for, you're going to build a better product or a better service.
Speaker 1:
22:50
Um, it's also, uh, an approach. It has steps along the way. So it gives you a roadmap for following that process to try and understand people. My take on it in particular or what I, I feel like I bring to the field or in what I'm passionate about is this idea of making sure that people are included fully in the process and that we're not making decisions on behalf of people, but we are facilitating and opening a space where people can move into that place and see themselves as designers. What are some of the questions that he more asks? There are some really simple ones like, uh, how are you feeling today or on a scale of one to a hundred, how stressed are you right now? And the script of what EMR says and does is constantly evolving. It's one of the things that we're getting feedback from people on, including asking people, well, what would you want the robot to, to you?
Speaker 1:
23:46
And it's fascinating to see what teens say. Even sometimes an utter and it's like we had a teen in one session, someone said I'm really stressed and the teen had put in the script. So EMR and a robot voice says, well that sucks. And everybody loved it because it sounded like something that, you know, a peer or another team might say. So we are still working on the exact kinds of language and it's going to be derived and designed from what um, what resonates with teenagers. So, so the project includes a diverse group of people. Yes. Why? Well, I think this goes back to the philosophy of human centered design as well. There's lots of technologies out there in the world that are designed by people who, let me step back and say this, prior to becoming a professor, I worked in the technology industry.
Speaker 1:
24:39
Like this was my gig. This is what I did for a living. It's something called user experience where you go into companies and communities and you're building products and services that work for them. And at that time the tech industry and still is, is pretty white and pretty male. And so oftentimes they're creating technologies that are reproducing the way that they see the world. What human centered design gives us is this ability to open up spaces but and to have different conversations with a more diverse group of people, but at the same time, the people who are doing the designs themselves also need to represent that diversity or we're going to design our own biases right into our products. So I think both in terms of people who are doing designing but also who gets included in design is really important. And I think when you, we always talk about human centered like center humans and you'll make good decisions, but like which humans are being centered in that process because who they are is really going to result in very different outcomes.
Speaker 1:
25:46
If you are centering people who are already very privileged and have a lot of resources, you're going to come up with different design choices. You're going to focus on like how do we make life easier for people who already maybe have a lot versus if you sent her different audiences, you're going to come up with different concerns and considerations. There's lots of work right now being done on facial recognition technology and if you are part of any marginalized or oppressed group, you know how dangerous and tricky of a situation that might be. And so to say like, well the technology will just recognize people's faces and not think about how this might be used to bring about some really terrible outcomes for particular groups of people is something that needs to be thought about. So if we don't have diverse groups of people involved in design, if we don't have diverse people in organizations that are doing the designing, we are just going to risk just reinforcing existing oppressions and marginalizing people who are already marginalized based on the project.
Speaker 1:
26:50
What are some things you've learned so far from the outcomes? Well, I, one thing that I've learned is how amazing and cool teens are to work with. I feel like I've worked in this field for a long time and it's so interesting when you're working with adults or professionals and when you go and you show them a design and they're like, yes, that's fine, but I have some concerns and you show a design to a teen and they're like, that's stupid. Like they're just so honest. Yeah. They're also so creative. So I feel like one of the big surprises and gifts of this project is to just spend time with teens who I think are underappreciated and stereotyped in our society and to really see kind of the creativity they have and, and be able to just be in that space to acknowledge and recognize that is, is a huge gift.
Speaker 1:
27:39
So I think that's one thing is like the teens part is amazing. Um, the other thing is being able to work with students like Christina and the rest of my research team at UDaB Tacoma, not just because we're all engaged in research and we're working really hard, which we are, but also to see the community that has been created by the students who have taken up this opportunity and to see how they have all like connected and bonded and learn these new skills and seeing themselves in different lights. Um, Christina is presenting a paper at a conference in the fall that we're writing together and she's really taken the lead on and it's just amazing to see that process. There's members of our research team that we work with and see them go through this process of like applying to grad school and getting in and being excited or you know, looking for that first job out of college and getting it.
Speaker 1:
28:27
And I feel like just being able to just be there in a classroom setting, you get to see that as a professor. But I feel like this research group experience is different. Like we just were, we're more like peers and we're just rooting for each other. And so that's, that's a huge gift personally of, of getting to do this work is just to get to spend time with our amazing EMT students and some pretty incredible, um, like Emma said earlier, like the sense of community has been really incredible. That was one of the first things that I noticed when I got on this design team was the strong presence of women on the team and then how we were always just encouraging each other to be better, to push forward, to kind of pursue research for the sake of doing this for teens. And that like component of like working towards creating a robot that's going to really better hopefully the teen experience when they're in high school.
Speaker 1:
29:16
Um, I think that's been really motivating for me as a researcher and a student. Right. Um, so how does being involved in research project like this help students? Well, I think hearing firsthand from student experiences, that's really helpful. But I do think it's that transition of seeing themselves as, um, more than just studying. Like, I think when you're in a class, you're expected to do the work, you got all assignments, it's kind of laid out. Like if you want to be successful, if you do the assignments like the professor is asking you to, then you're most likely going to be successful. And I think what's really different in a research group environment is that like there's no one answer on how to do this and it's messy and it's emergent. And sometimes it might feel a little chaotic because we're just trying to figure out as we go.
Speaker 1:
30:07
So there's no script. So I think for students to sort of step into those situations too, and be able to be just self starters and do things for themselves and just be able to have a little bit more ownership and also get to see the process of research from the inside. I think that that is, um, is key. It can be really helpful because no matter where you go after college, that's going to be a skill that you're going to need to refine is to figure out, okay, there's no script for this. What do I do next?
Speaker 2:
30:38
Thank you to our guests and a big thing. Get to our senior lecturer, Nicole Blair for letting us play your music on the show. Thank you to Mooney, our recording studio, and thank you for joining us today. Be sure to subscribe and go to iTunes, Spotify, Google podcasts, Stitcher, and pocket casts. [inaudible].
×

Listen to this podcast on