Recipes

A cookbook of common analyses against LABO CSVs. Each recipe is self-contained: what it does, which files you need, a pandas snippet, an R equivalent, gotchas, and cross-links.

All recipes assume you've loaded the CSVs from a single run directory and that the files exist (see File layout for gating rules).

Which recipe do I need?

I want to...	Recipe
Measure the delay between a stimulus and a button press	Reaction time
Compute a mean signal (gaze, pose) locked to stimulus onset	Stimulus-locked averaging
Analyse where the participant looked over time	Gaze paths
Align LABO data with EEG / biopac / external trackers	Cross-device sync

Conventions used in these recipes

• Pandas is primary. Every recipe has a copy-pasteable pandas snippet. R snippets are included under a Tabs block where they help; if you use R, they're one-to-one equivalents.
• MonotonicExecutionTime is the reference clock. See Timing.
• FrameNumber is the primary join key within a session. See Join keys.
• Files may be gated out. Not every stream produces a file every run — see each stream's page for gate conditions.

Beyond these recipes

These cover the most common experimental-psychology questions. For anything novel:

1. Figure out whether you need a regular, irregular, or mixed analysis — see Regular vs irregular.
2. Identify the stream(s) — each stream page has a "When to use" paragraph.
3. Pick join keys — FrameNumber within a session, MonotonicExecutionTime across sessions or devices.
4. Write the join; validate with a small spot-check before scaling up.

If you write a reusable recipe, add it here.