Update study dataset (2 files)

Exported by scripts/figures.mjs on the dashboard's default task selection.

README.md

@@ -8,7 +8,7 @@ tags: [human-study, web-agents, grounding, hci]
 Every answered task from the PageGuide web study that the analysis is currently counting, one
 row per task, straight from `study_task_results_v2`.
 
-Exported 2026-08-14T04:56:50.743Z · 180 rows · 42 participants.
+Exported 2026-08-14T04:57:14.121Z · 76 rows · 40 participants.
 
 ## Conditions
 
@@ -22,9 +22,9 @@ Each participant sees both arms, interleaved task by task:
 | Facet | Rows | Tasks | Selected by |
 |---|---|---|---|
 | Find × Text | 28 | EDU-v1, HARRY-v1 | the dashboard |
-| Find × Visual | 45 | TREE-V1, SVSF-V1, TESLA-V1 | the dashboard |
-| Guide × Text | 49 | gv2-ms9j3200-u0i9nm, gv2-ms9hwloh-15fuvy, gv2-ed05972e-i5fi3b, gv2-ed35d549-ct71ub | the dashboard |
-| Guide × Visual | 58 | gv2-msf1kfhv-o9o8n8, gv2-msf0vpxs-qucehj, gv2-msf0i7ll-hffde9, gv2-msf28nd3-gig8aj | the dashboard |
+| Find × Visual | 13 | TESLA-V1 | the dashboard |
+| Guide × Text | 18 | gv2-ms9hwloh-15fuvy | the dashboard |
+| Guide × Visual | 17 | gv2-msf0i7ll-hffde9 | the dashboard |
 
 Some tasks are held out of a card — a disputed answer key, a duplicate re-recording, a run whose
 ground truth cannot be scored in full. The table above is exactly what produced this export.