Transcript Breakout 2 - drug adherence - Building New Theories of Human

New Computationally-Enabled Theoretical Models to Support Health BC&M
It is the year 2030. You are starting to test a theory that allows
for predictions to support real time intervention on drug use.
WHAT IS NEEDED to predict behavior? Articulate your theory
and design an experiment to test it. THINK BOLD AND HIGH
RISK HIGH REWARD!!
Scribe: Jimi Huh
1.Niilo Saranummi
2.James Lester
3.Brigitte Piniewski
4.Ross Hammond
5.Vicente Traver
6.Loukianos Gatzoulis
Breakout group summary
1
New Computationally-Enabled Theoretical Models to Support Health BC&M
Ingredients of the model
• Current (and past sequence of):
– Physiology
– Affect
– Social context
– Environmental context
• Novelty of context; past experience
• Experience of same-category others
Breakout group summary
2
New Computationally-Enabled Theoretical Models to Support Health BC&M
Ingredients of model (cont’d)
• Stochasticity
• Sensor data from drugs themselves
on actual adherence
• Real-time location of drugs,
individual, and pharmacy/medical
network nodes
• Social network and reference groups
Breakout group summary
3
New Computationally-Enabled Theoretical Models to Support Health BC&M
Outputs of the model
Breakout group summary
4
New Computationally-Enabled Theoretical Models to Support Health BC&M
Experimental Design for Testing
Breakout group summary
5
New Computationally-Enabled Theoretical Models to Support Health BC&M
Conceptual form of the model
• The model as “learning machine”
– Discover the (unobservable) algorithms/heuristics
used by actual individuals to choose behavior
• Model as ally
– Make the unobservable available to the participants;
model can become a tool for “know thyself” and for
individual self-improvement without threatening
autonomy
• Model as a learning companion
– Tamagotchi electronic pet in Japan
Breakout group summary
6
New Computationally-Enabled Theoretical Models to Support Health BC&M
We did not address…
•
•
•
•
Model technical design or feasibility
Funding
Ethical and privacy issues
Sensor arrays and reliability needed for
necessary data streams
• Etc
Breakout group summary
7