Multimodal-Multisensor

Eye, cardiac and questionnaire data from a study of 10 adults across three weekly sessions. The release includes available recordings, saved results and reports.

Keywords: Eye tracking · Cardiac intervals · Questionnaires · Session reliability

Pipeline

Full-size diagram · Edit in Figma

Pipeline source

Run

Run from the repository root with Python 3.11 or 3.12. These commands build the documentation; analysis code and notebooks were removed.

python -m venv .venv
source .venv/bin/activate
python -m pip install -r requirements-dev.txt
python -m pytest tests -q
python scripts/build_docs.py
python -m http.server 8000 --directory build/docs

Open http://localhost:8000.

Results

Study: 10 adults × 3 sessions. Release: 24 raw sensor streams.

Saved ICC(1,1) estimates describe the reliability of within-session standard deviations in the released group tables.

Within-session standard deviation	ICC(1,1)	95% interval
Interval samples	0.22	−0.13 to 0.66
Pupil diameter	0.45	0.07 to 0.79
Response duration	0.61	0.25 to 0.87

Descriptive results; no diagnostic validation. Most participant summaries lack raw-data mappings. Clock alignment and unique-beat timing are unverified. Methods and limits · Report corrections.

Case-study Session 1

Tech Stack

Scope	Tools
Current documentation	Python 3.11/3.12 · Mistune · pytest · HTML/CSS
Archived analysis	NumPy · pandas · SciPy · Matplotlib · scikit-learn

Roadmap

These steps require source evidence that is currently unavailable.

• Document participant-to-recording mappings under the recorded consent.
• Reassess P01 if the original export becomes available.
• Verify acquisition timezones and clock synchronization.
• Validate unique-beat timestamps and record interval corrections.
• Confirm eye-openness units and polarity with labeled calibration frames.

References

1. Technical report
2. Thesis
3. Proposed protocol
4. Dataset citation

License

Code: MIT · Data, image and questionnaire terms