diff --git a/plans/audio-review-fixes-handoff.md b/plans/audio-review-fixes-handoff.md
index 375accf047..2f20da8072 100644
--- a/plans/audio-review-fixes-handoff.md
+++ b/plans/audio-review-fixes-handoff.md
@@ -166,10 +166,16 @@ Three judgement calls worth knowing:
   | Chain output peak | Samples changed | Error vs. correct | Verdict |
   | --- | --- | --- | --- |
   | 0.48x | none | -70.0 dB | nothing clamped; 16-bit rounding only |
-  | 1.35x | 2.1% | -25.1 dB | **the author listened and could barely tell** |
+  | 1.35x | 2.1% | -25.1 dB | **barely detectable by ear** |
   | 2.69x | 19.3% | -9.1 dB | plainly audible grit |
   | 4.00x | 48.8% | -3.3 dB | not the same recording |
 
+  **Verified by ear, 2026-08-09.** The author could barely tell the two apart
+  at 1.35x and heard the difference "very clearly" once the chain was pushed.
+  So the mechanism is confirmed real and the fix does what it claims — and the
+  severity genuinely is conditional on overshoot, rather than the pair having
+  been a bad test. This is the listening step §4a-1 was waiting on; it is done.
+
   So: **a chain that stays under full scale is unaffected** — the fix changes
   nothing for it. The bug is real and worth fixing (it destroys data that the
   lane was about to scale down anyway, and there is no reason to accept that),