HapTurk: Crowdsourcing Affective Ratings of Vibrotactile Icons
summary
Project translates 10 high‑fidelity C2 tactor vibrations via VibViz into two proxy modalities—waveform and designed visualizations—and into low‑fidelity Android phone vibrations using pulse‑width modulation. Local lab ratings (slider 0–100, equivalence testing with ~15–20 SE delta) show duration and pleasantness often preserve, energy often fails, and designed visualization outperforms waveform while phone vibrations better capture speed/urgency. Deployment on Mechanical Turk yields 88% agreement with local proxies, demonstrating viable but imperfect crowdsourcing of vibrotactile affective ratings.
AI-written summary · about
chapters (12)
- 0:00Introduction to HapTurk
- 1:10Challenges in haptic design
- 3:37Proxy modalities for haptics
- 5:02Designing perceptual vibrations
- 8:38Proxy modalities implementation
- 9:22Local study setup and goals
- 10:25Equivalence testing explained
- 11:30Local study results
- 13:18Remote Mechanical Turk study
- 14:43Key findings and takeaways
- 15:47Future work and improvements
- 16:30Q&A session
related talks
10mSubjective video quality assessment for mobile devices
47mTesting the Agents That Look You in the Eye
15mEvals Are Not Unit Tests
38mVR Testing - The Testing Trinity
34mUsing robots to test mobile apps
46mUsability Testing is easier than you think
01h 15mAgile Mobile Testing Workshop
05mPsychology study background and prospects on iOS devices