First Run with Sample Data

Get the sample materials

Install EyeSort, then obtain the open example data and configurations or the release compatibility bundle. Use the settings supplied with that sample rather than treating GUI placeholder values as universal measurements.

Run the verified menu flow

1. Choose EyeSort → 1. Load EEG Dataset(s). Select one sample .set file and a writable output directory.
2. Choose 2. Setup Interest Areas → Text-Based Sentence Contents and Interest Areas. Load the sample IA file/configuration, confirm the values, and run Step 2.
3. Open Inspect Parsed Regions. Check several condition–item rows and their pixel boundaries.
4. Choose 4. Eye-Tracking Event Labeling. Load the sample label queue or create a broad, nonzero label; select an event-marker format; click Apply All & Finish.

Expected result: EyeSort writes <sample>_processed.set and normally <sample>_processed.fdt to the Step 1 output directory. It may also write eyesort_labeling_summary_NNN.csv. No manual save command is required.

Verify in MATLAB

EEG = pop_loadset('filename', 'sample_processed.set', ...
                  'filepath', '/path/to/output');
assert(EEG.eyesort_processed)
assert(isfield(EEG.event, 'current_region'))
labeled = ~cellfun(@isempty, {EEG.event.eyesort_full_code});
nnz(labeled)
unique({EEG.event(labeled).bdf_label_description})

Expect a positive labeled count. Labeled fixation events should contain original_type, eyesort_full_code, and BDF description fields.

Optional outputs

Generate eyesort_bins.txt only after labels match. Use Modify Event Code Format to inspect readable types or restore numeric codes. Use History Scripts → Save processing history script to create a portable .m workflow and sidecars.

For every click and validation checkpoint, follow the quickstart tutorial.