Working R21 feasibility prototype
Daily Decisions
A web platform for studying how older adults use conversational AI, algorithmic recommendations, and independent information search when making repeated health and financial decisions.
Demonstration artifact. The participant records and dashboard values shown here are simulated and illustrate platform capabilities; they are not pilot findings.
Explore the working prototype
Each participant demonstration is fixed to Day 1 and opens a representative decision in the selected assistance format.
Chat assistant
Ask a conversational assistant questions about the decision before choosing.
Open demonstration →Algorithmic recommendation
Review a one-shot recommendation, confidence estimate, and expandable rationale.
Open demonstration →Independent search
Search a controlled set of sources and decide without an AI recommendation.
Open demonstration →Researcher dashboard
Follow aggregate condition patterns into participant histories, trial-level interaction replay, and CSV exports for R or Python.
Implemented at submission
- Participant flow across all three assistance formats
- Controlled error manipulation and next-day outcome reveal
- Task-bounded interaction telemetry and chat transcripts
- Researcher monitoring, trial replay, and analysis-ready CSV exports
Supported by the R21 award
- Expansion and expert validation of the full scenario bank
- Reminder, retention, and missed-session workflows
- Security, data-governance, and production deployment work
- Accessibility and usability testing with older adults