Skip to content

PROJECT_CONTEXT.md

Latest commit

 

History

History
121 lines (83 loc) · 4.54 KB

PROJECT_CONTEXT.md

File metadata and controls

121 lines (83 loc) · 4.54 KB
Raw

BAP Dynamometer Analysis — Project Context

Handoff document for Cursor agents working on this repo locally on Mohd's MacBook.

Goal

Build analysis tooling for BAP in-scanner dynamometer (grip force) data, starting with understanding and inventorying the raw files, then linking them to existing behavioral outputs, and eventually running more advanced analyses.

Current phase: exploratory — get a solid sense of the gripforce files (structure, coverage, force ranges, alignment with behavioral data) before advanced modeling.

Repository

  • Code: /Users/mohdasti/Documents/GitHub/dynamometer-analysis
  • Remote: dynamometer-analysis on GitHub (analysis code only; raw data stay outside git)

Data locations (local Mac — not in git)

Resource Path
Raw gripforce CSVs /Users/mohdasti/Documents/LC-BAP/BAP/BAP data
Behavioral / prior analysis (Nov 2025) /Users/mohdasti/Documents/LC-BAP/BAP/Nov2025

Confirmed (Mac): BAP data/ holds 85 sub-BAP### folders plus BAP tablet data/. Nov2025/ is a flat folder of behavioral CSVs and LC master spreadsheet exports; primary trial-level file is bap_beh_trialdata_v3.csv.

Raw dynamometer data layout

BIDS-like hierarchy:

BAP data/
├── sub-BAP001/
│   ├── ses-1/
│   │   └── InsideScanner/
│   │       └── subjectBAP001_{Task}_session1_run{N}_{M}_{D}_{H}_{Min}_gripforce.csv
│   └── ses-2/ ...
├── sub-BAP103/
└── ... (~90 subject folders)
  • 202 gripforce CSV files total (all under *InsideScanner* folders)
  • Files match pattern: *_gripforce.csv

Task types (from filenames)

Analysis uses oddball tasks only; MVCnPRAC calibration runs are excluded from pipelines (MVC comes from behavioral mvc column).

Task File count Status
Aoddball 84 Included
Voddball 72 Included
MVCnPRAC 46 Excluded from analysis

Filename pattern

subject{BAP###}_{TASK}_session{N}_run{R}_{M}_{D}_{H}_{Min}_gripforce.csv

Example:

subjectBAP103_MVCnPRAC_session3_run1_7_8_12_44_gripforce.csv
  • {M}_{D}_{H}_{Min} ≈ acquisition date/time (e.g. July 8, 12:44)

Gripforce CSV format

  • No header row
  • 4 comma-separated numeric columns
  • ~100 Hz sampling (column 2 steps ≈ 0.01 s)
col1,  col2,       col3,          col4
257,   0.010000,   15358.043546,  11.000000   ← MVCnPRAC example
0,     0.010030,   11286.366915,  1.000000    ← Voddball example
Col Likely meaning Notes
1 Marker / event / run code e.g. 257 in MVCnPRAC, 0 in Voddball — not force
2 Elapsed time within run (seconds) Use for within-run alignment
3 Absolute timestamp (seconds, large offset) Use for cross-modal sync (scanner, behavior)
4 Grip force Values ~1–13 in samples seen; units TBD — check acquisition notes

Behavioral data (Nov2025)

Folder: /Users/mohdasti/Documents/LC-BAP/BAP/Nov2025

Prior behavioral processing already exists here. The agent should inventory this folder and determine how tables map to gripforce files (subject, session, run, task, event timestamps).

Expected linking keys: sub-BAP###, session, run, task name (Aoddball, Voddball, MVCnPRAC).

Suggested next steps for the agent

  1. Confirm both data paths exist and list Nov2025 contents.
  2. Run python scripts/run_inventory.py to build a data inventory (subjects, sessions, tasks, file counts, run durations, peak force per file).
  3. Inspect Nov2025 behavioral outputs and document join keys to gripforce files (see docs/DATA_LINKAGE.md and generated reports/dataset_inventory.md).
  4. Paths are configured in config/paths.local.yaml (copy from config/paths.example.yaml).
  5. Produce a short summary report before any advanced analyses.

Conventions

  • Keep raw data outside git; only scripts, config templates, and derived summaries in repo.
  • Prefer simple, focused scripts over heavy abstraction at this stage.
  • Match existing conventions in Nov2025 if present before inventing new schemas.

Useful shell commands (run on Mac)

export BAP_DATA="/Users/mohdasti/Documents/LC-BAP/BAP/BAP data"
export BAP_BEHAV="/Users/mohdasti/Documents/LC-BAP/BAP/Nov2025"

find "$BAP_DATA" -path "*InsideScanner*" -name "*_gripforce.csv" | wc -l

find "$BAP_DATA" -path "*InsideScanner*" -name "*_gripforce.csv" \
  | sed 's/.*subjectBAP[0-9]*_//; s/_session.*//' | sort | uniq -c

ls -la "$BAP_BEHAV"
find "$BAP_BEHAV" -maxdepth 3 -type f | head -40