2. UBHave's
...aim is to investigate the power and
challenges of using mobile phones and
social networking for Digital Behaviour
Change Interventions (DBCIs), and to
contribute to creating a scientifc
foundation for digitally supported
behaviour change.
3. Digital Behaviour Change Interventions
...focus on delivering `information' via
digital means (e.g., a web site) in order
to support intents to change behaviour
5. “...each of these transactions leaves
digital traces that can be compiled into
comprehensive pictures of both
individual and group behaviour...
“Computational Social Science” Lazer et. al
7. “...sampling to capture data from the
sensors of the phone cannot be
performed continuously, as this will
drain the battery rapidly. However,
conservative sampling leads to the loss
of valuable behavioural data...”
K. Rachuri
8. “Study fndings suggested that young, currently
healthy adults have some interest in apps that
attempt to support health-related behaviour
change [...] The ability to record and track
behaviour and goals and the ability to acquire
advice and information “on the go” were valued.
Context-sensing capabilities and social media
features tended to be considered unnecessary
and off-putting.”
“Opportunities and Challenges for Smartphone Applications in Supporting
Health Behavior Change: Qualitative Study” Dennison et. al
10. Towards a framework...
Mobile Web App
Native Mobile App
Reconfgurable
Interfaces
Dynamic Content
Sensing
Notifcations
11. {
“intervention_id”:”my_intervention”,
“questions”: [ … ]
“diary”: [ …]
“sensors”: [ …],
“trigger”:[
{“accelerometer”:”moving”, “survey”:”physical_activity”}
]
}
...that can be 'authored'
Using well-known mobile app design patterns
Native app's benefts, web apps' benefts:
12. ● Questionnaires
● Feedback
● Sensor data collection & management
Part of the path so far...
Mostly measurement. (experience sampling)
Building from a subset of the functionality:
14. ● Battery-friendly sensor data collection
● Triggering notifcations
● Data storage & transmission
“Reinventing the Wheel”
All smartphone-based research needs to
begin by engineering solutions for:
15. ● Pull Sensors
– Accelerometer, Location, Microphone
– Wi-Fi, Bluetooth, Camera
– Active apps, SMS/Call Log Content
● Push Sensors
– Battery, Connection State
– Proximity, Screen
– Phone Calls/SMS Events
Everything as a 'Sensor'
16. Open Source Android Smartphone
Libraries
http://emotionsense.org
https://github.com/nlathia/SensorManager
https://github.com/nlathia/TriggerManager
https://github.com/nlathia/SensorDataManager
17. ● How can we keep users engaged in a
seemingly repetitive task?
– Diversify and sample from the questions as a
“journey” of unlocking feedback
– User needs vs. research needs
● How can we effciently collect sensor data?
– First deployment took a naïve approach
– Current implementation focuses on CPU time
rather than sensor strategy
Design Challenges
18.
19. Sensor & Emotion Data
Valence vs. Sociability
Self-Report:
r = 0.0581
Valence vs. SMS Events:
r = 0.2154
20.
21. “Can I run an ESM study
like Emotion Sense?”
Generalise sensor-
enhanced experience
sampling tool. Currently in
alpha testing.