SSL vs handcrafted features under age-matched speaker-disjoint splits

EVALUATION SSL DOES NOT BEAT HANDCRAFTED SCOPE-LIMITED svd see artifact

Scope caveat. This run is a pilot slice -- 613 of the 13,535 speakers measured elsewhere on svd. It may satisfy the repeat contract and still be a narrower measurement than a later run on the same corpus; both facts are shown because either alone would mislead.

0.6496

mean ROC-AUC, eGeMAPS (age-matched)

no CI in the artifact

0.6227

mean ROC-AUC, WavLM (same folds)

n = 10

repeats × 5 folds

n=10 repeats

613

speakers

recording count not in artifact

Power contract (R6)

Paired Wilcoxon over 10 repeats gives a minimum attainable p of 0.001953 = 2/2¹⁰, against a Holm-tightest threshold of 0.005556 = 0.05/9 for the pre-registered family. Feasible. This is arithmetic on the design, computed before the data: it says what the run could have shown, not what it did.

Per-seed detail — every repeat, not a summary

seed	speakers	recordings	age healthy	age patho	age gap	AUC age only	AUC WavLM	AUC eGeMAPS
0	616	10,243	34.32	35.21	0.893	0.5655	0.6151	0.6392
1	610	10,402	34.43	34.96	0.531	0.5585	0.6253	0.6335
2	616	10,200	34.20	35.04	0.839	0.5474	0.6240	0.6539
3	608	10,682	34.51	35.14	0.632	0.5787	0.6307	0.6718
4	608	10,160	34.25	35.13	0.882	0.5684	0.6217	0.6519
5	610	9,579	34.08	34.84	0.760	0.5292	0.6139	0.6444
6	610	10,427	34.41	35.08	0.670	0.5490	0.6183	0.6466
7	608	9,968	34.41	34.83	0.420	0.5391	0.6307	0.6602
8	620	10,117	34.33	35.51	1.181	0.5359	0.6277	0.6562
9	626	10,652	34.16	35.08	0.924	0.5623	0.6196	0.6386

eGeMAPS beats WavLM on 10 of 10 seeds. Paired difference -0.02693 (95% CI [-0.03172, -0.02151]) — the interval excludes zero, and it excludes it on the losing side for the SSL model.

Encoders registered: HeAR, WavLM-base+, Whisper-small-enc; encoders actually run: WavLM-base+, eGeMAPS(baseline). Scope stated in the artifact: HeAR and Whisper not extracted on this host; COUGHVID excluded per F2. A registered-but-unrun encoder is a hole in the family, not a free pass — it is why the falsifier is not formally closed.

The matching is checked, not assumed

Age matching at a ±3.0-year tolerance collapses the age-only AUC to 0.5534 and leaves a mean age gap of 0.7732 years. The artifact records matching_worked = True. Without that collapse the comparison beneath it would be meaningless, so it is measured every run rather than assumed once.

Reproduction provenance

artifact	`autoresearch_results/V1_ssl_vs_handcrafted.json` · view raw · last written 2026-07-28 16:05 UTC
objective	`ROC-AUC vs clinical labels (JUDGE-FREE)`
wall clock	499.7 s (0.14 h)
written up as	F7 in FINDINGS.md

What this page does not have

No 7-step reasoning entry exists for this run. The sibling convention renders Diagnosis · Citations · Hypothesis · Prediction · Verdict · Learning on every per-experiment page, sourced from a reasoning_annotations.json. This program never wrote one, and neither does it keep an append-only experiment_log.jsonl; the run history is the set of artifact files listed on the run ledger, ordered only by their write time. The pre-registered prediction for this run does exist — it is on the V1 hypothesis page, recorded before the run. Everything else on that convention’s checklist is absent here and is shown as absent rather than quietly dropped.