Inspect Parsed Regions and QC Results

Inspect the reconstructed stimuli

After Step 2, choose EyeSort → 2. Setup Interest Areas → Inspect Parsed Regions. Select a processed dataset if prompted. The read-only table shows one row per condition–item combination and two columns per region: literal text and [start, end] pixel boundaries.

Check examples from every condition and stimulus family. Verify that:

• region text is in the intended column and order;
• spaces appear where expected;
• boundaries advance by the correct character widths;
• the first boundary starts at the measured offset;
• right-to-left boundaries proceed in the expected screen direction.

Export the table to CSV or tab-delimited text and retain it with analysis records.

Read Step 2 diagnostics

Inspect EEG.eyesort_step2_diagnostics for matched trial/condition/item triggers, complete condition–item keys, assigned boundaries, fixation/saccade counts, missing text keys, missing or invalid X values, and fixations assigned to regions.

d = EEG.eyesort_step2_diagnostics;
[d.assigned_boundary_events d.fixations_assigned_to_region]
d.missing_text_file_keys
d.unused_text_file_keys

Stop conditions

Do not continue when zero boundaries were assigned, fixation-region assignments are unexpectedly sparse, or condition–item keys disagree. Correct trigger formatting, IA keys, X-field names, offset, character width, region order, or reading direction and rerun Step 2.

Next: optionally import IA columns, then build a label queue.