EEG ObservatoryASO
ValidationPASS
EvidentiaryVERIFIED
SMOKETEST DATASET — SubjectJ LRHand motor-task EEG. Condition labels are remapped for pipeline testing only and carry NO bereavement meaning. These results must never be reported as grief/bereavement evidence.
Views · projects · runs
Acquire
Validate
Register Events
Epoch
Preprocess
Extract Features
Analyze
Test Hypotheses
Verify
Report
ROC-AUC
0.722
Permutation p
0.0196
50 perms
Null mean ± SD
0.502
± 0.026
Cohen's d
1.086
Majority baseline
0.684
ROC curve (out-of-fold)
GroupKFold · AUC 0.722
00.250.50.751false positive rate00.250.50.751
Permutation null distribution
Label-shuffled negative control
0.4530.4650.4770.4890.5010.5130.5250.5370.5490.561null AUC02468observed 0.722

Confusion matrix & leakage controls

pred control
pred deceased
true control
591
25
true deceased
180
104
Cross-validation PASS GroupKFold(5)
Overlap-leakage grouping PASS
Participant-ID audit NOT APPLICABLE
Held-out participant INSUFFICIENT DATA
Held-out stimulus INSUFFICIENT DATA

Alternative model scoreboard

ModelStatusVerdict
A · Constraint propagationUNRESOLVEDDeceased-related state is discriminable; consistent with A but not yet distinguished from D/E without matched confound controls.
B · Global representational extinctionUNRESOLVEDRequires an independent person-identity preservation test to weaken/support; not available in this dataset.
C · General emotional salienceUNRESOLVEDNot excluded: no valence/arousal measurements to rule out a nonspecific salience account.
D · Familiarity / autobiographical memoryUNRESOLVEDNot excluded: no familiarity/memory-richness controls in current design.
E · Attachment intensityUNRESOLVEDLiving-attachment control present but attachment-strength covariates absent; cannot separate attachment intensity.

Robustness multiverse (feeds evidentiary state & claim gate)

PASS
Variants
10
Significant fraction
100%
AUC range
0.659–0.744
Median AUC
0.719
Analytic variantAUCvs null upper
features=all_features0.722PASS
features=spectral_only0.670PASS
features=no_connectivity0.715PASS
features=time_domain_only0.710PASS
model=lda0.744PASS
model=linear_svm0.659PASS
cv_folds=30.709PASS
cv_folds=100.729PASS
seed=430.722PASS
seed=440.722PASS
Effect is stable across the admissible analytic multiverse — supports the current claim level.

Evidentiary state · falsification · challenge matrix

Prespecified failure conditionStatusNote
Deceased-related states cannot be distinguished from matched controlsPASSDeceased vs controls AUC=0.722, p=0.0196.
Permutation testing removes the apparent effectPASSEffect survives permutation null.
Effects disappear under participant-level validationINSUFFICIENT DATASingle-participant dataset cannot test participant-level generalization.
Classification explained by emotional arousalUNRESOLVEDNo arousal measurement available to exclude an arousal account.
Person representation globally decays rather than remaining distinguishableINSUFFICIENT DATARequires an independent person-identity preservation test.
No dissociation between explicit knowledge and neural responseINSUFFICIENT DATARequires explicit-recall contrast with concurrent conscious report.
Results depend materially on preprocessing choicesPASSEffect stable across the admissible analytic multiverse.
Results depend on a few contaminated trialsNOT TESTEDTrial-influence analysis not yet executed.
Longitudinal behavior contradicts the proposed architectureINSUFFICIENT DATASingle session; longitudinal trajectory not testable.