Privacy

Study identity

Your study activity is recorded under a generated anonymous handle (for example @quiet-fox) tied to a random identifier stored in a cookie on your device. We do not ask for your name.

Optional referral challenge

After completing every case, you may optionally create a three-word referral link and provide a separate contact email. Only friends who visit the link before starting and later complete every assigned case count.

Referral records contain a campaign-specific HMAC pseudonym, your code, completed referral credits, and the contact email. They contain no raw study user ID, handle, response, assignment, judgment, or raffle key. An authorised operator holding the separate secret can recompute the pseudonym, so these records are pseudonymised rather than fully anonymous.

The email is deleted after prize verification. Participant-level referral records are deleted after the 30-day dispute window. Referral details are never sent to PostHog or included in research analysis. Read the referral terms.

Optional prize draw

After completing all cases, you may choose to enter a prize draw for one £100 Amazon gift card. Entry is optional and does not affect your participation. To enter, you provide one email address, which we use only to administer the draw and contact the selected entrant. We do not use it for marketing, include it in research analysis, or send it to PostHog.

Your email is stored separately from your survey responses and can be accessed only by authorised researchers administering the draw. We delete the contact details immediately after prize verification.

Read the raffle guidelines.

What we store

For each clip: your verdict, response time, watch time, full-video and evidence replay loops, which exhibits you open, how often and how long you inspect each one, playback speed and scrubbing, the playback and exhibit state when you decide, page focus, viewport size, and technical video loading or error events. We also retain the exact exhibit regions shown and an event-by-event audit trail so the study can be reproduced and checked. For your session, we store a coarse device class, anonymous study setting, completion state, and experience points.

Analytics

We use PostHog to understand where people drop off and how fast the app responds. Autocapture and session recording are switched off; IP addresses are discarded. PostHog receives page navigation and the explicit study events described above, keyed to the same anonymous identifier. It does not receive your raffle email, the clip's ground truth, generator condition, referral code or email, or whether your individual verdict was correct.

Cookies

One httpOnly cookie keeps your anonymous study session so you can resume where you left off and lets the server verify that all cases were completed before accepting a prize-draw entry. The cookie also returns a completed session to the final wrap-up page. Clearing it means the app cannot reconnect you to that session.

A second httpOnly cookie temporarily remembers the first valid referral link used before the study starts. It expires when the 30-minute settlement period ends and is cleared earlier after the completion credit is resolved. It contains the public three-word code and an integrity tag bound to your anonymous browser session.