Trial-independent EEG familiarity classification remains statistically significantly above chance for the 4-class, equations-only, and faces-only tasks under Bonferroni-corrected permutation tests¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Permutation tests (1000 permutations, HistGradient Boosting proxy) showed all three familiarity tasks significantly above chance: 4-class (adjusted p≤0.0030), faces-only (adjusted p=0.0060), equations-only (adjusted p=0.0360). → Isuru Nanayakkara and Thilina Halloluwa 2026
Evidence¶
Isuru Nanayakkara and Thilina Halloluwa 2026¶
Isuru Nanayakkara and Thilina Halloluwa. (2026). Automating Learner Assessment: Benchmarking Machine Learning and Deep Learning Models for EEG-Based Familiarity Prediction. https://arxiv.org/abs/2608.16541 doi:10.48550/arxiv.2608.16541
q2 · i? · design · r2
Empirical permutation testing (1000 permutations, trial-group-level label shuffling) using HistGradient Boosting as a computationally efficient proxy; the article reports that "all three tasks achieve statistical significance" under Bonferroni correction. No effect size is printed.
"Under this correction, all three tasks achieve statistical significance: the 4-class task (p≤0.0010, adjusted p≤0.0030), the faces-only task (p=0.0020, adjusted p=0.0060), and the equations-only task (p=0.0120, adjusted p=0.0360)."
Discussion¶
Related Claims¶
- Trial-level spectral features robustly differentiate stimulus domains (faces vs. equations) but no familiarity-driven spectral shift survives FDR correction — related
- Domain-separated EEG classifiers decode familiarity within each stimulus category, with peak LOGO Mean Recall of 0.8992 for faces-only (HistGradient Boosting) and 0.8670 for equations-only (CNN) — related
- Temporal leakage from shuffled neighboring epochs inflates EEG familiarity classification, with peak weighted F1 dropping from 0.9853 (CNN, stratified) to 0.6038 (CNN, Group K-Fold) — related
- Familiarity recognition is represented by a distributed multivariate spatial-spectral pattern rather than localized single-feature shifts, necessitating multivariate machine learning — related
- Offloading is decodable from metacognitive monitoring and control brain processes — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
trial-independent-eeg-familiarity-above-chance |
title |
Trial-independent EEG familiarity classification remains statistically significantly above chance for the 4-class, equations-only, and faces-only tasks under Bonferroni-corrected permutation tests |
description |
Trial-independent EEG familiarity classification remains statistically significantly above chance for the 4-class, equations-only, and faces-only tasks under Bonferroni-corrected permutation tests |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — isuru-nanayakkara-and-thilina-halloluwa-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.