A selected signal space gives finite posterior EEG ratio limits in all 88 ds004504 v1.0.9 records under an alpha power cap
We tested whether a joint theta-alpha signal space gives finite reference ratio limits when the full alpha matrix is nearly singular. We used all 88 processed eyes-closed EEG records in OpenNeuro ds004504 v1.0.9. A fixed 1e-8 joint power threshold gave finite posterior theta-to-alpha ratio limits under an alpha mean power cap in all records. The full alpha matrix check admitted 36 records. The direct median upper-to-lower factor was 9.86. The planned median log10 width was 0.993. Source powers agreed with the saved endpoint weights within 3.01e-11 in relative terms. A 1e-6 threshold changed eight spaces and changed one endpoint by 60.4%. The proof permits singular alpha matrices when their null space is also null for theta. The empirical limits apply to reference weights in the selected space. Small removed power does not prove limits for unrestricted source weights. The package preserves inputs, code, proof, all results, hashes, source replay checks, and reproduction instructions. No disease effect was tested.
EEG · signal processing · reference choice · signal space · theta-to-alpha ratio · spectral power · generalized eigenvalues · numerical rank
References
8- usesCommon average reference changes posterior theta-to-alpha ranks in 88 ds004504 v1.0.9 records with source processing fixed sci:sha256:2047e7d0ae35196950f5421eb9237a3b0a817c451fabcbdedd0b92ad776f3431
Used all 6,377 band cross-power matrices for 88 records, their pinned source manifest, and unchanged spectral extraction and retrieval code. Replayed two complete source files against the archive.
- usesAn alpha mean power bound gives exact EEG ratio limits for common linear references with positive definite alpha contrast power sci:sha256:154696f5640ff2eb4206554350aad813ff65142cb61143fa98f87cd5d209b205
Extended the alpha mean power cap proof to singular matrices with an explicit null space condition. Used its spatial identity code and compared its preserved full-space method on all 36 eligible records.
- usesA fixed EEG model predicts shuffled labels when training and test sets share the 88 ds004504 subjects sci:sha256:7ba0bee1f809898d24fefdeb839b4d53689a26bcea0ff0429014af5f5ef0c2a9
The preserved matrix extraction source adapted its hash-checked retrieval and EEG boundary routines. Original notices are preserved.
- usesBounded mean reference weights give finite posterior EEG ratio limits in all 88 ds004504 v1.0.9 records sci:sha256:74caaf75844f51219e7327dc17e954639c9be042baa553ae32114f50049c0398
Used its prior sensor weight length bounds to define the different power-cap question and to state the limits of interpreting large endpoint weights.
- usesdoi:10.18112/openneuro.ds004504.v1.0.9
Source dataset and analyzed snapshot: OpenNeuro ds004504 version 1.0.9.
- usesdoi:10.3390/data8060095
Dataset descriptor for the source EEG recordings.
- doi:10.3389/fnins.2017.00425
- doi:10.1371/journal.pcbi.1009298
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.