Scalp to Source
Progress

About

Scope

This site is a structured, interactive curriculum in scalp EEG analysis: from reading a raw trace to source estimation, inference and applied electives. It is methodologically mainstream. MNE-Python is the vehicle; EEGLAB and FieldTrip equivalents are named in sidebars. Every lesson is drafted for expert review and carries a status (draft, review, published) until the author promotes it; nothing is presented as settled that the author has not checked.

Author

This site is Scalp to Source, published at https://eeg.neurokinetikz.com/ (§13 item 1, answered 2026-09-19). Author name and affiliation: TODO(confirm). The curriculum structure, lesson outlines, dataset directory and reading-list seeds were written by the author; draft prose, widgets, notebooks and extraction scripts were built from those outlines with AI assistance and are reviewed by the author before publication.

Product neutrality

This site does not promote any research program, theory, device or app — including the author’s own. Where consumer and mobile EEG devices are discussed (Level 7), the treatment is comparative and neutral: devices are named once as a fact and otherwise described by class (gel or dry electrodes, channel count, hardware bandwidth, reference). No device is described with vendor marketing language; no product is recommended. The author’s own recordings are not used anywhere on the site.

Datasets

This site uses only publicly licensed datasets. It never vendors a dataset. The repository contains only short extracted snippets and precomputed products with provenance, notebooks that download subsets from official sources, and a public directory of datasets with their citations, access steps and licenses. Snippets are derivatives (cropped, resampled, re-referenced, filtered) and are labelled as such; where a dataset’s license statements conflict, both are shown verbatim and the data descriptor is always attributed. Dataset pages are generated from the author’s catalog by script, never typed by hand.

License

Proposed licensing (to be confirmed by the author, §13 item 4):

  • Text and figures: CC BY 4.0.
  • Code (site, widgets, DSP library, notebooks, scripts): BSD-3-Clause.
  • Data assets are excluded from the text license. Every snippet and precomputed product shipped on this site carries its own source license — CC0, CC BY 4.0, ODC-By 1.0 or PDDL by default — recorded together with its dataset, subject, run, generating script and checksum in data/manifest.json. Assets under a share-alike or non-commercial license ship only if the author accepts those terms (§13 item 15); none do by default.

Attribution for a derived asset must name the source dataset and its license as recorded in the manifest, not only this site.

Cite this site

See Cite this site for how to cite the site and every dataset it uses. The site’s own citation still needs an author name and a version or commit to be complete: TODO(confirm).

Contribute

The site is built from the repository neurokinetikz/eeg, which is private, so it takes no issues or pull requests from outside at the moment; corrections are welcome by whatever route you reached the author. (§13 item 3, answered 2026-09-19. The repository being private is also why every notebook’s Open in Colab button is disabled: Colab cannot read a private repository.) Scientific content is reviewed by the author before it is published; please cite a source for any number or claim you propose. Labels on drill items and annotations are algorithmic until the author reviews them, and pull requests that change labels should say why.

Privacy

There are no accounts, and your work stays on your machine: progress — lessons marked complete, exercise results, capstone self-assessments — is stored only in your browser’s localStorage, and can be exported and imported as JSON. Nothing you do on a page is sent anywhere.

Analytics: the site uses Plausible, which counts page views without cookies and without any cross-site identifier, so there is nothing to consent to and no profile of you to build. It is the one third-party request the site makes: loading a page fetches a script from plausible.io and reports that the page was viewed. If you would rather it did not, any content blocker stops it and the site works exactly the same without it.