Designing for engagement: Using participatory design to develop a social robot to measure teen stress Emma J. Rose, PhD & Elin Björling Assistant Professor University of Washington Tacoma @emmarosephd
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Designing for engagement:Using participatory design to develop a social robot to measure teen stress
Emma J. Rose, PhD & Elin BjörlingAssistant ProfessorUniversity of Washington Tacoma
@emmarosephd
UX
Long term
Ubiquitous
Engagement
Immediate fixes
At work
Reducing friction
Then Now
EngagementIn Human-Computer Interaction
Relational communication
Relational dynamics
Relational nonverbal behavior
Bickmore & Picard 2005
How might we…
better understand stress in teens
Why stress? Why teens?
Why stress? Why teens?
Why stress? Why teens?
Teens experience more chronicstress, than other age group (Leonard, et al. 2015)
Adolescent brain is extremely vulnerable to the negative impact of stress (Armsten, et al. 2014)
Stress increases the possibility of both mental and physical illnesses (Hamilton, et al. 2014)
Chronic stress is correlated with suicidal ideation (Rosiek 2016)
A Different Way to Measure Stress
Here Now
What is your current stress level right now?0 ---------------------------------------------------------100
Ecological Momentary Assessment (EMA)Collecting data in the moment it happensand within the context in which it occurs (Shiffman, et al 2008)
Accurately measures data over time
Here Now
Ecological Momentary
Introducing EMAR: Ecological Momentary Assessment Robot
Our ideaA social robot for school environments that invites teens to engage and share data in a social and fun way.
Why a social robot?
Getting good data, requires creative approachesEngaging- Data collection is not burdensome or boring (a hypothesis)
Social- Robots often enhance and increase human-human social interaction
Innovative- It’s never been done. Teens are an underexplored population in human-robot interaction.
Methods and approach
How might we…
Build an engaging social robot that resonates with teens?
Human centered approachparticipatory, iterative and data driven
Session 1 Session 2 Session 3
Friend groupIdeationElicitation
n=6
STEM meet upIdeationElicitationFeedback on v1n=15
HS Robot clubIdeationElicitationFeedback on v1n=19
Iterative Prototyping, Research & Design
What we’ve learned … so far
Mental models
Writing scenarios to identify features and functions
Drawing storyboards to place robots within a school context
Informed by popular culture
GLaDOS from the video game Portal
Creative ideas
What does it mean to be a helper? Teens thought EMAR should be a helper
With tasks
With homework
With social interactions
Thought of EMAR as a peer
Appearance and embodiment
(aka) Avoiding the uncanny valley
Reactions to printed images of existing robots
Interactions with low fidelity prototype
Jibo (MIT) “It looks like it cares about you.”
“It’s cute, and simple, that’s why I like it”
“Cute and clean”
“It shows expression, it can move it’s head”
“Very cute and friendly”
“Amazon looking smile.”
“Like a puppy”
“Cute and innocent”
Boxie (MIT)
Pepper (SoftBank, Japan)“Clean”
“Face is creepy”
“White, soulless eyes”“Feels more like a peeper, than a pepper”“I can’t trust it”
“Stalkerish”
Dragonbot (MIT)
“Nightmare Fuel!!!”
“No, god, no!”
“Reeeeallllly creepy”
“Totally would not be seen with this EVER”
Evaluating a
prototypelow fidelity
very
What does it mean to be cute? Teens like and are drawn to cute: small, likeable, approachable
Cute can help establish intimacy, earn trust.
Design cues: Big head, large eyes, smile
What does it mean to feel heard?
EMAR Script A:“I’m sorry you are feeling stressed, try talking to a friend or exercising to help you relax”
EMAR Script B:“Thank you for answering. I like to know how you’re doing.”
How does a robot represent and fit into my world?
Gender and race
“He’s (or she or it) black! Saying yes, even robots can be a different race too! Racial… equality across the board”
- Teen from Session 2
Conclusion & Next StepsTeens are engaged by a social robot, even a (very) low fidelity prototype. EMAR shows promise.
Designing for engagement requires long term, participatory methods.
In the next 3+ years, we plan to further design, iterate, and pilot EMAR in a local high school*.
* Thanks to the National Science Foundation
Acknowledgments
Elin BjörlingMaka CakmakAndrew Davidson
The Directed Research Group students at UW Tacoma & UW Seattle
Thanks to all the teens who have participated in our design sessions.
Emma Rose - [email protected]@emmarosephd
EMAR@RobotEMAR
Related Documents
