This moderated study can evaluate the local M1 Task 7 adult logging implementation and status-labeled prototypes for later concepts. It does not validate production readiness, native-device behavior, or post-M1 workflows. Recruit 8–12 adults aged 18 or older. Do not recruit minors or ask participants to provide medical conditions, real meal photos, account credentials, or private household disputes.
Research questions:
Aim for variation in:
Include participants with access needs where recruitment and prototype support allow. Record demographics only when necessary and consented.
Explain that the prototype is not medical advice, may not work, and uses fake data. Participants may skip any task or stop at any time. Do not ask them to assess their body, disclose diagnoses, or photograph food/home surroundings.
Script:
We are testing a product concept, not you. PlatePilot is intended for adults and provides general wellness information, not medical advice. Please use only the synthetic scenarios. You can pause, skip a question, or stop at any time.
Confirm age 18+, consent, recording preference, and think-aloud expectations.
Do not probe medical history or weight goals.
Prompt: “Set up the concept for the synthetic adult Sam, who prefers simple information.”
Observe disclaimer comprehension, optionality, age boundary, retention understanding, and hesitation.
Prompt: “Sam wants to log the sample lunch photo, then notices the estimate missed an item.”
Trigger a delayed/failure state. Ask the participant to recover with manual entry. Observe estimate interpretation, edit discoverability, status language, and whether the diary remains usable.
Prompt: “Add a sample glass of water, undo an incorrect amount, and explain what Today shows.”
Observe units, reversibility, simple/detailed mode, and any moral interpretation.
Prompt: “Add a meal to Friday’s plan, then show what happens when Friday arrives.”
Check that a plan is not mistaken for a consumed meal or prescription.
Prompt: “Sam considers inviting Alex. Find what Alex would see and how Sam could later stop sharing.”
Ask: “Could Alex export or delete Sam’s private records?” Observe assumptions, coercive cues, and whether acceptance is clearly independent.
Prompt: “Find the trial allowance, restore a fictional purchase, change photo retention, and locate export and account deletion.”
Observe price/allowance comprehension, pressure, provider-pending state, and scope of destructive actions.
Reiterate that the prototype made no changes and collect no real records.
For each task record:
Planning thresholds—not claims of validation:
Do not average away severe privacy or safety failures. A single credible cross-adult disclosure misunderstanding requires redesign and retest.
Cluster findings by flow, then separate usability, accessibility, privacy, content, and implementation assumptions. Report participant count, prototype version, task order, limitations, and contradictory evidence. Store notes under approved retention/access rules and remove recordings on schedule.
Tag recommendations as: