This repository contains MATLAB scripts to build pooled datasets and run the analysis pipeline.

• scripts/run_all.m: runs all main scripts except dataset creation.
• scripts/main_000_create_dataset.m: builds data_final/*_pool.mat from raw *_spk.mat files.
• scripts/main_000_behav.m: behavioral preference/saccade analysis, report and figures in report_000.
• scripts/main_001_anova_lda.m: ANOVA + LDA outputs in processed and report in report_001.
• scripts/main_002_states.m: state decoding outputs and report in report_002.
• scripts/main_003_crossdecoding.m: crossdecoding analyses and report in report_003.
• scripts/main_004_unitstability.m: unit stability figures/statistics and report in report_004.

Most scripts use this pattern:

• f = mfilename('fullpath')
• if empty, fallback to pwd
• otherwise compute project root from script location
• build project-relative paths from that root

This means scripts do not require a specific MATLAB working directory.

At project root:

• data_final/ (generated pooled *_pool.mat files)
• processed/ (analysis outputs)
• scripts/ (pipeline scripts)
• report_000/ ... report_004/ (generated reports/figures)

1. Open MATLAB and set current folder to project root (DynamicGeom).
2. Run full analysis pipeline:
   • run('scripts/run_all.m')

• scripts/run_all.m removes existing report folders before running scripts.
• Report files are overwritten ('w' mode) on each run.
• If you run individual scripts, each script creates its own report folder if needed.