Forgetting of Passwords: Ecological Theory and Data
summary
Paper adapts ecological theory of long‑term memory and derives ACT‑R–style models that predict login duration and recall odds, validated in a month‑long field experiment controlling frequencies from 1-to-8-day frequencies. Models fit observed behavior with RMSE 1.8 s for entry time and recall odds RMSE 0.09, and show recall probability rises steeply and plateaus after about five successful logins. CMU password meter and a neural net estimate strength and reveal that higher‑strength passwords take longer to enter and are less memorable.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:47Ecological Theory
- 5:00Password Usage
- 5:27Field Study
- 7:37Results
- 9:05Fitted Model
- 10:36Logging Frequency
- 11:52Reciprocity
- 12:25Mean Way
- 13:10Creation Strategies
- 13:28Other Related Work
- 14:48Conclusion
related talks
12mThe more accounts I use, the less I have to think': A Longitudinal Study on the...
19mDesign and Evaluation of a Data-Driven Password Meter
27mFast, Lean, and Accurate: Modeling Password Guessability Using Neural Networks
12mProspects for Improving Password Selection
01h 05mPasswords are not going away
35mHow UX Research and Behavioral Science Can Help to Improve People's Cyber Security Behaviors
31mBreaking Bad (Passwords)
32mzxcvbn: Low-Budget Password Strength Estimation