This repository contains the scripts that I used to analyze and plot the ERP data for an experiment on dialectal variation in United States English.
There are two relevant background files for the ERP analyses. 0a_global_params loads packages, pulls participant numbers from MATLAB for exclusion/inclusion, establishes global formatting parameters for plotting, and creates helper functions for number formatting (e.g., p values). These helper functions were mainly used in writing the manuscript. 0b_behavioral is the top-level script that references several underlying data-wrangling scripts for the various surveys and other pieces of background information (not included in this repository). Its primary output is summary_back, which as the name implies is a summary of all the background/behavioral data for each participant.
The group analysis files use the processed data from MATLAB to run a series of within- and between-group ANOVAs, format the values, and create data tables that can be passed onto the plotting scripts, output to CSV files, and joined with the cleaned behavioral data for correlation analyses (not included in this repository). The first script, 1a_group_analysis, is the main analysis script across the whole experiment, while 1b_group_analysis_halves compares the first and second halves of the experiment.
2_plot_format uses the ERP analysis from the main script to pull significant time windows for plotting. It also sets up plotting parameters. 3_erp_plots automatically generates butterfly plots for the two groups with significant time windows labelled.