From raw EEG traces to source-level judgment, one rung at a time. Published lessons and labs.https://eeg.neurokinetikz.com/enhttps://eeg.neurokinetikz.com/curriculum/0-raw-signal/what-eeg-measures/https://eeg.neurokinetikz.com/curriculum/0-raw-signal/what-eeg-measures/Where scalp potentials come from, why synchrony and geometry decide what reaches the electrodes, and where EEG sits among LFP, ECoG, MEG and fMRI.lessonlevel-0https://eeg.neurokinetikz.com/curriculum/0-raw-signal/electrodes-and-montages/https://eeg.neurokinetikz.com/curriculum/0-raw-signal/electrodes-and-montages/10-20/10-10 names and conventions, electrode versus channel versus reference versus ground, impedance, and referential versus bipolar montages.lessonlevel-0https://eeg.neurokinetikz.com/curriculum/0-raw-signal/amplifiers-sampling-recording/https://eeg.neurokinetikz.com/curriculum/0-raw-signal/amplifiers-sampling-recording/Differential amplification, acquisition metadata, events and triggers, clock sync, and the common file formats.lessonlevel-0https://eeg.neurokinetikz.com/curriculum/0-raw-signal/reading-raw-traces/https://eeg.neurokinetikz.com/curriculum/0-raw-signal/reading-raw-traces/Set scale, window and montage for viewing; recognize awake rhythms and state changes; tell eyes-open from eyes-closed by eye.lessonlevel-0https://eeg.neurokinetikz.com/curriculum/0-raw-signal/artifact-atlas/https://eeg.neurokinetikz.com/curriculum/0-raw-signal/artifact-atlas/Identify by eye every common physiological and non-physiological artifact, then prove it on a 20-item drill.lessonlevel-0https://eeg.neurokinetikz.com/curriculum/0-raw-signal/data-hygiene-and-metadata/https://eeg.neurokinetikz.com/curriculum/0-raw-signal/data-hygiene-and-metadata/A first-look checklist for any recording, EEG-BIDS naming, a QC log, and catching flat, DC-offset and mislabelled channels.lessonlevel-0https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/sampling-nyquist-aliasing/https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/sampling-nyquist-aliasing/The sampling theorem, alias frequencies, why decimation needs an anti-alias filter, and choosing a rate for a target analysis.lessonlevel-1https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/time-and-frequency-domains/https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/time-and-frequency-domains/Any signal as a sum of sinusoids; reading amplitude, phase and frequency; the DFT and its bins; resolution as 1/T; amplitude versus power versus dB.lessonlevel-1https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/power-spectral-density/https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/power-spectral-density/Why the periodogram is noisy, Welch's method and its parameters, correct units and scale, and averaging PSDs correctly.lessonlevel-1https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/windowing-leakage-zero-padding/https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/windowing-leakage-zero-padding/Spectral leakage, choosing a window, and what zero-padding does (interpolation) and does not do (resolution).lessonlevel-1https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/filters-fir-iir/https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/filters-fir-iir/Choose a filter type, cutoff and transition band for a stated goal, and predict the distortions it will introduce.lessonlevel-1https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/line-noise/https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/line-noise/Detect 50/60 Hz and harmonics in a PSD, choose among notch, spectral-fit and spatial removal, and verify removal without collateral damage.lessonlevel-1https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/aperiodic-and-periodic/https://eeg.neurokinetikz.com/curriculum/1-signal-fundamentals/aperiodic-and-periodic/The 1/f-like background, separating peaks from slope with spectral parameterization, individual alpha frequency, and why band power conflates the two.lessonlevel-1https://eeg.neurokinetikz.com/curriculum/2-preprocessing/loading-and-bids/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/loading-and-bids/Read every common format with MNE, set channel types and montage, convert to EEG-BIDS with mne-bids, and read events back.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/2-preprocessing/channel-locations-and-bad-channels/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/channel-locations-and-bad-channels/Assign and verify montages, detect bad channels with objective criteria, interpolate, and decide how many bads is too many.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/2-preprocessing/re-referencing/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/re-referencing/Reference dependence, linked mastoids versus average versus Cz versus REST, how topographies and polarity change, and where re-referencing sits in the pipeline.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/2-preprocessing/filtering-in-practice/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/filtering-in-practice/Goal-dependent high-pass and low-pass settings, the two-pass ICA strategy, boundary events, and when to downsample.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/2-preprocessing/artifact-rejection/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/artifact-rejection/Amplitude and flatness criteria, autoreject, annotation-based rejection, quantifying data loss, and exclusion criteria set before looking at effects.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/2-preprocessing/eog-regression-and-ica/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/eog-regression-and-ica/EOG regression, ICA at intuition level, correct rank and data quantity, component classification with ICLabel as a second opinion, and avoiding over-cleaning.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/2-preprocessing/asr-ssp-and-alternatives/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/asr-ssp-and-alternatives/Artifact subspace reconstruction, when it is preferable and what it risks, plus SSP and wavelet-based cleaning.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/2-preprocessing/pipeline-order-and-reproducibility/https://eeg.neurokinetikz.com/curriculum/2-preprocessing/pipeline-order-and-reproducibility/A canonical, justified order of operations, scripted with a config file, logged, with a per-subject QC report and pinned versions and seeds.lessonlevel-2https://eeg.neurokinetikz.com/curriculum/3-event-related/epoching-and-baseline/https://eeg.neurokinetikz.com/curriculum/3-event-related/epoching-and-baseline/Events to epochs, choosing windows, justifying baseline correction, baseline as a covariate, and overlapping trials.lessonlevel-3https://eeg.neurokinetikz.com/curriculum/3-event-related/the-erp-and-its-components/https://eeg.neurokinetikz.com/curriculum/3-event-related/the-erp-and-its-components/Averaging and SNR, the canonical components, latent component versus observed peak, polarity conventions, and a tour of the ERP CORE paradigms.lessonlevel-3https://eeg.neurokinetikz.com/curriculum/3-event-related/measuring-erps/https://eeg.neurokinetikz.com/curriculum/3-event-related/measuring-erps/Peak versus mean versus area, latency measures, defining windows without circularity, and difference waves.lessonlevel-3https://eeg.neurokinetikz.com/curriculum/3-event-related/topographies/https://eeg.neurokinetikz.com/curriculum/3-event-related/topographies/Plotting topomaps correctly, interpolation and extrapolation beyond the electrode hull, map shape versus amplitude, and map sequences.lessonlevel-3https://eeg.neurokinetikz.com/curriculum/3-event-related/snr-trial-counts-and-design/https://eeg.neurokinetikz.com/curriculum/3-event-related/snr-trial-counts-and-design/Standardized measurement error, trials needed per condition, design confounds, and filter distortion of slow components.lessonlevel-3https://eeg.neurokinetikz.com/curriculum/3-event-related/single-trial-approaches/https://eeg.neurokinetikz.com/curriculum/3-event-related/single-trial-approaches/ERP images, RT sorting, single-trial regression (rERP), and linear deconvolution for overlapping events.lessonlevel-3https://eeg.neurokinetikz.com/curriculum/3-event-related/statistics-for-erps/https://eeg.neurokinetikz.com/curriculum/3-event-related/statistics-for-erps/Within-subject tests on measures, why uncorrected time-point tests are wrong, cluster-based permutation, and exactly what a cluster p-value licenses.lessonlevel-3https://eeg.neurokinetikz.com/curriculum/4-time-frequency/why-time-frequency/https://eeg.neurokinetikz.com/curriculum/4-time-frequency/why-time-frequency/What trial averaging hides: evoked versus induced versus total power, and when a time-frequency analysis is warranted.lessonlevel-4https://eeg.neurokinetikz.com/curriculum/4-time-frequency/stft-and-wavelets/https://eeg.neurokinetikz.com/curriculum/4-time-frequency/stft-and-wavelets/The STFT and its window trade-off, Morlet wavelets defined by frequency and cycles, the time-frequency uncertainty, and edge effects.lessonlevel-4https://eeg.neurokinetikz.com/curriculum/4-time-frequency/multitaper-and-hilbert/https://eeg.neurokinetikz.com/curriculum/4-time-frequency/multitaper-and-hilbert/Multitaper spectral concentration and bandwidth, band-limited amplitude and phase from filter-Hilbert, and when to prefer each.lessonlevel-4https://eeg.neurokinetikz.com/curriculum/4-time-frequency/baseline-normalization-erd-ers/https://eeg.neurokinetikz.com/curriculum/4-time-frequency/baseline-normalization-erd-ers/Why raw TF power needs normalization, dB versus percent versus z baselines, choosing an edge-safe baseline, and computing ERD/ERS.lessonlevel-4https://eeg.neurokinetikz.com/curriculum/4-time-frequency/phase-itc-and-coupling/https://eeg.neurokinetikz.com/curriculum/4-time-frequency/phase-itc-and-coupling/Intertrial phase coherence, phase resetting versus additive evoked models as competing accounts, and phase-amplitude coupling with its confounds.lessonlevel-4https://eeg.neurokinetikz.com/curriculum/4-time-frequency/is-it-really-an-oscillation/https://eeg.neurokinetikz.com/curriculum/4-time-frequency/is-it-really-an-oscillation/Test for a spectral peak before reporting band power, detect bursts, run cycle-by-cycle analysis, and recognize non-sinusoidal waveforms.lessonlevel-4https://eeg.neurokinetikz.com/curriculum/4-time-frequency/statistics-for-time-frequency/https://eeg.neurokinetikz.com/curriculum/4-time-frequency/statistics-for-time-frequency/Cluster-permutation tests over channel × time × frequency, TFCE, limiting the search space with a priori ROIs and bands, and reporting.lessonlevel-4https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/volume-conduction/https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/volume-conduction/One source produces high zero-lag coherence across all sensors; why that invalidates naive sensor connectivity, and what survives.lessonlevel-5https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/sensor-space-connectivity/https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/sensor-space-connectivity/Coherence, PLV, PLI/wPLI, imaginary coherence and orthogonalized envelope correlation; what each is robust to; directed measures; surrogate testing.lessonlevel-5https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/laplacian-and-csd/https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/laplacian-and-csd/The surface Laplacian via spherical splines: reference independence, spatial sharpening, what it attenuates, and its use for ERPs and connectivity.lessonlevel-5https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/forward-models/https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/forward-models/Head models and conductivity assumptions, fsaverage as a template, coregistration, the leadfield, and expected localization error.lessonlevel-5https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/inverse-problem/https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/inverse-problem/Ill-posedness, minimum-norm family estimates, LCMV beamformers, point-spread and resolution, and which claims EEG source estimates can defend.lessonlevel-5https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/source-space-connectivity/https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/source-space-connectivity/Leakage in source space, symmetric orthogonalization, parcel-level connectivity, and conservative interpretation.lessonlevel-5https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/spatial-filters-for-decoding/https://eeg.neurokinetikz.com/curriculum/5-connectivity-and-source/spatial-filters-for-decoding/Common spatial patterns, filters versus patterns and why only patterns are interpretable, xDAWN, and the bridge to decoding.lessonlevel-5https://eeg.neurokinetikz.com/curriculum/6-inference/multiple-comparisons/https://eeg.neurokinetikz.com/curriculum/6-inference/multiple-comparisons/Enumerate the search space; compare FWER and FDR control, Bonferroni, max-statistic, cluster permutation and TFCE; pre-specify ROIs and windows.lessonlevel-6https://eeg.neurokinetikz.com/curriculum/6-inference/mixed-models-trial-level/https://eeg.neurokinetikz.com/curriculum/6-inference/mixed-models-trial-level/What averaging-then-testing discards, linear mixed models with subject and item random effects on single trials, convergence, and reporting.lessonlevel-6https://eeg.neurokinetikz.com/curriculum/6-inference/effect-sizes-and-power/https://eeg.neurokinetikz.com/curriculum/6-inference/effect-sizes-and-power/Effect sizes for within-subject EEG designs, SME as precision, and simulation-based power analysis using pilot data.lessonlevel-6https://eeg.neurokinetikz.com/curriculum/6-inference/circularity-and-flexibility/https://eeg.neurokinetikz.com/curriculum/6-inference/circularity-and-flexibility/Double-dipping, orthogonal contrasts and collapsed localizers, preregistration, and multiverse analysis.lessonlevel-6https://eeg.neurokinetikz.com/curriculum/6-inference/decoding-as-inference/https://eeg.neurokinetikz.com/curriculum/6-inference/decoding-as-inference/Time-resolved decoding with proper cross-validation, chance level and permutation significance, temporal generalization, and what above-chance decoding means.lessonlevel-6https://eeg.neurokinetikz.com/curriculum/6-inference/reporting-standards/https://eeg.neurokinetikz.com/curriculum/6-inference/reporting-standards/Reporting filters, reference, rejection, ICA, epoching, measurement and statistics to COBIDAS-MEEG standard; figures with uncertainty; sharing code and data.lessonlevel-6https://eeg.neurokinetikz.com/curriculum/7-applied/decoding-and-bci/https://eeg.neurokinetikz.com/curriculum/7-applied/decoding-and-bci/Benchmark CSP + LDA, Riemannian classifiers and EEGNet with moabb, within- and cross-subject, without leakage.lessonlevel-7https://eeg.neurokinetikz.com/curriculum/7-applied/sleep-eeg/https://eeg.neurokinetikz.com/curriculum/7-applied/sleep-eeg/Sleep stages and graphoelements, a hypnogram with yasa, and spectra across the night.lessonlevel-7https://eeg.neurokinetikz.com/curriculum/7-applied/real-time-and-neurofeedback/https://eeg.neurokinetikz.com/curriculum/7-applied/real-time-and-neurofeedback/Streaming with LSL, ring buffers and causal filters, latency budgets, online artifact handling, feedback design, and what neurofeedback evidence supports.lessonlevel-7https://eeg.neurokinetikz.com/curriculum/7-applied/clinical-eeg-primer/https://eeg.neurokinetikz.com/curriculum/7-applied/clinical-eeg-primer/What clinical review looks like, the appearance of epileptiform discharges in published examples, why spike detection is hard, and what a research analyst must not claim.lessonlevel-7https://eeg.neurokinetikz.com/curriculum/7-applied/mobile-and-consumer-eeg/https://eeg.neurokinetikz.com/curriculum/7-applied/mobile-and-consumer-eeg/Dry and low-channel-count systems, motion artifacts, which analyses survive with 4–8 channels, and validation against a research system — comparative and product-neutral.lessonlevel-7https://eeg.neurokinetikz.com/curriculum/7-applied/resting-state-biomarkers/https://eeg.neurokinetikz.com/curriculum/7-applied/resting-state-biomarkers/Resting-state features, test–retest reliability, and the biomarker replication problem, with the confounds of real clinical archives.lessonlevel-7https://eeg.neurokinetikz.com/curriculum/7-applied/multimodal/https://eeg.neurokinetikz.com/curriculum/7-applied/multimodal/EEG–fMRI, EEG–TMS and EEG–eye-tracking setups, their artifacts, and the corrections that exist.lessonlevel-7https://eeg.neurokinetikz.com/curriculum/7-applied/pipelines-at-scale/https://eeg.neurokinetikz.com/curriculum/7-applied/pipelines-at-scale/mne-bids-pipeline on a BIDS dataset, containers, batch execution, provenance tracking, and a reproducibility statement.lessonlevel-7https://eeg.neurokinetikz.com/labs/w-aperiodic-explorer/https://eeg.neurokinetikz.com/labs/w-aperiodic-explorer/Adjust offset, exponent, optional knee and peaks on real resting PSDs; the model and residual update live, a quick fit and a precomputed specparam reference are shown, and two subjects or conditions can be compared side by side.labhttps://eeg.neurokinetikz.com/labs/w-bad-channel-detective/https://eeg.neurokinetikz.com/labs/w-bad-channel-detective/Mark the bad channels in five 30-second segments, then open the metric panels and compare your call with an objective detector.labhttps://eeg.neurokinetikz.com/labs/w-build-a-signal/https://eeg.neurokinetikz.com/labs/w-build-a-signal/Add and remove sinusoids (frequency, amplitude, phase), see the sum and its spectrum, and rebuild a real 2-s eyes-closed epoch from its top-k Fourier components with the residual.labhttps://eeg.neurokinetikz.com/labs/w-burst-detector/https://eeg.neurokinetikz.com/labs/w-burst-detector/Threshold an alpha amplitude envelope on 60 seconds of real eyes-closed EEG and watch "sustained alpha" resolve into a handful of events — with the burst count moving elevenfold on the threshold alone.labhttps://eeg.neurokinetikz.com/labs/w-cluster-permutation-viz/https://eeg.neurokinetikz.com/labs/w-cluster-permutation-viz/Step through permutations on a real two-condition contrast, watch the null distribution of maximum cluster statistics fill, and read what the resulting p-value does and does not license.labhttps://eeg.neurokinetikz.com/labs/w-csp-explorer/https://eeg.neurokinetikz.com/labs/w-csp-explorer/A CSP filter and the matching pattern side by side, with the angle between them measured — and what a cross-validated accuracy actually belongs to.labhttps://eeg.neurokinetikz.com/labs/w-dipole-to-scalp/https://eeg.neurokinetikz.com/labs/w-dipole-to-scalp/Place and rotate current dipoles in a spherical head model and watch the scalp potential update live; then put the same dipole through two head models — one shell against four — and measure what the skull does.labhttps://eeg.neurokinetikz.com/labs/w-epoch-builder/https://eeg.neurokinetikz.com/labs/w-epoch-builder/Move tmin, tmax and the baseline window on real trials and watch the single-trial image, the condition averages and the difference wave update together.labhttps://eeg.neurokinetikz.com/labs/w-erp-averager/https://eeg.neurokinetikz.com/labs/w-erp-averager/Add trials one at a time and watch the component emerge from the noise, then turn the same trials into the measurement error curve that answers "how many trials do I need?".labhttps://eeg.neurokinetikz.com/labs/w-evoked-vs-induced/https://eeg.neurokinetikz.com/labs/w-evoked-vs-induced/Every trial contains the same burst; jitter its phase and watch it leave the ERP while the time-frequency map barely moves — with the law that governs it drawn through the measured curve.labhttps://eeg.neurokinetikz.com/labs/w-filter-sandbox/https://eeg.neurokinetikz.com/labs/w-filter-sandbox/Flagship: live filter design on real 10-s traces, with the trace, spectrum, impulse and step response updating as you drag cutoffs and order.labhttps://eeg.neurokinetikz.com/labs/w-garden-of-forking-paths/https://eeg.neurokinetikz.com/labs/w-garden-of-forking-paths/Analytic flexibility under the null and on real data — 486 defensible pipelines over one oddball experiment, counted two ways.labhttps://eeg.neurokinetikz.com/labs/w-ica-component-gallery/https://eeg.neurokinetikz.com/labs/w-ica-component-gallery/A drill on at least 40 real ICA components — classify each from its topography, time course, spectrum and ERP image, then read the evidence.labhttps://eeg.neurokinetikz.com/labs/w-latency-budget/https://eeg.neurokinetikz.com/labs/w-latency-budget/Where the delay in a real-time loop comes from, decomposed into filter, buffer and processing — and what a zero-phase filter costs when you insist on running one live.labhttps://eeg.neurokinetikz.com/labs/w-measurement-explorer/https://eeg.neurokinetikz.com/labs/w-measurement-explorer/Switch measurement method and window on the same ERP and watch both the value and its across-subject variance move.labhttps://eeg.neurokinetikz.com/labs/w-montage-explorer/https://eeg.neurokinetikz.com/labs/w-montage-explorer/A head with selectable 10-20 / 10-10 / 10-5 density; click or tab to an electrode for its name and coordinates; highlight which positions a given montage contains.labhttps://eeg.neurokinetikz.com/labs/w-phase-clock/https://eeg.neurokinetikz.com/labs/w-phase-clock/Trial phases as unit vectors on a circle, ITC as the length of their average — with the small-N floor drawn beside it, and an added evoked component raising ITC without any phase reset.labhttps://eeg.neurokinetikz.com/labs/w-raw-scroller/https://eeg.neurokinetikz.com/labs/w-raw-scroller/Scroll real multichannel EEG on a 32-channel 10-20 subset, with window length, scale and montage controls and toggleable annotations.labhttps://eeg.neurokinetikz.com/labs/w-reference-explorer/https://eeg.neurokinetikz.com/labs/w-reference-explorer/Switch the reference on a real ERP and watch the waveform, the amplitude at a chosen electrode and the topography change together.labhttps://eeg.neurokinetikz.com/labs/w-sampling-explorer/https://eeg.neurokinetikz.com/labs/w-sampling-explorer/Sample a continuous signal at an adjustable rate with or without an anti-alias filter, compute alias frequencies, and decimate a real 500 Hz recording with and without filtering.labhttps://eeg.neurokinetikz.com/labs/w-spot-the-artifact/https://eeg.neurokinetikz.com/labs/w-spot-the-artifact/Drill: label the artifact type in real 5-s multichannel segments, get immediate feedback with the tell-tale features highlighted, and keep a running score.labhttps://eeg.neurokinetikz.com/labs/w-tf-baseline-explorer/https://eeg.neurokinetikz.com/labs/w-tf-baseline-explorer/Raw, un-normalized time-frequency power for C3 and C4, with the baseline window and the normalization applied live — so a badly placed baseline visibly rewrites the map.labhttps://eeg.neurokinetikz.com/labs/w-threshold-tuner/https://eeg.neurokinetikz.com/labs/w-threshold-tuner/Move a peak-to-peak rejection threshold and watch the per-condition rejection counts and the resulting ERP change together.labhttps://eeg.neurokinetikz.com/labs/w-volume-conduction-sim/https://eeg.neurokinetikz.com/labs/w-volume-conduction-sim/Put one or two sources inside a conducting sphere, watch them arrive at every electrode at once, and read the coherence matrix that produces.labhttps://eeg.neurokinetikz.com/labs/w-wavelet-explorer/https://eeg.neurokinetikz.com/labs/w-wavelet-explorer/The wavelet, its spectrum and the time-frequency map of a real epoch, all driven by the same two numbers — and then the same epoch through multitaper and filter-Hilbert beside it.labhttps://eeg.neurokinetikz.com/labs/w-welch-explorer/https://eeg.neurokinetikz.com/labs/w-welch-explorer/Welch parameters on real resting EEG — segment length, overlap, window, periodogram toggle and zero-padding — and a window mode that shows leakage on a pure tone plus real EEG.lab