Finger ERP Analysis

This mini-project computes event-related potentials (ERPs) from ECOG data aligned to finger movement events. It extracts 1201-sample windows (200 ms pre-start, 1 ms at start, 1000 ms post-start) for each trial, averages per finger, and plots the mean ERP per finger.

Files

• fingers_erp.py: Implements calc_mean_erp that loads trial metadata and ECOG signal, performs QA, averages ERPs per finger, and plots the results.
• main.py: Script entry point that calls calc_mean_erp with the provided data files and prints the resulting matrix shape.

Data layout

Expected inputs (CSV):

• events_file_ordered.csv: columns starting_point, peak_point, finger (all integers); each row is one trial.
• brain_data_channel_one.csv: one column of ECOG samples (floats), row count aligned with the indices in events_file_ordered.csv.

Placement rules: the script will look for each CSV in this folder, its parent folder (repo root), or any subfolder whose name contains data (case-insensitive). Keep the original filenames.

Environment

Python 3.10+ with the following packages: pandas, numpy, matplotlib. Install with:

pip install -r requirements.txt

Or directly:

pip install pandas numpy matplotlib

Running

From the repository root:

python main.py

This will compute the per-finger mean ERPs, display the plot, and print the ERP matrix shape.