Skip to content
Buy one Spikeling unit units

Extracellular recording

Curriculum status

These entries are architecture specifications, not final lesson plans. They define teaching intent, logistics, outputs and scientific boundaries; release-ready settings, worksheets, notebooks and answer keys still require empirical validation.

Return to the module catalogue

Family source cluster: [P3; P4; O2; O4; G7]

EXT-01 — Intracellular versus extracellular signals

Concept / theme Extracellular waveforms are geometry-dependent field measurements, not scaled copies of membrane voltage.
Audience / context Advanced undergraduate. Formal practical, workshop or modular course. Prior modules: FND-02, DAT-02.
Logistics 45 min; 1–3; boards: 0–1; software: Spikeling extracellular GUI; Jupyter/SpikeInterface; prepared dataset: GUI-generated or prepared timestamped traces. Equipment: Standard USB cable; worksheet/protocol card.
Mode / stages GUI simulation. Stages: 6–11.
Spikeling relationship 6 — GUI simulation workflow; 7 — Conceptual analogy; 8 — External computational comparison

Learning outcomes.

  • Identify and predict the principal behaviour described in extracellular waveforms are geometry-dependent field measurements, not scaled copies of membrane voltage.

  • Configure or document compare ground-truth Vm with template and dv/dt extracellular modes and record Vm, source spikes and four contact traces.

  • Measure, calculate or compare compare ground-truth Vm with template and dv/dt extracellular modes using an explicit operational rule.

  • Interpret the result and state why forward model is reduced and µv-like; no tissue/morphology solution is implied

Roadmap.

1. Initial prediction or classification.

2. Configure the board, GUI, simulation or dataset and record metadata.

3. Compare ground-truth Vm with template and dV/dt extracellular modes.

4. Acquire or inspect Vm, source spikes and four contact traces.

5. Produce aligned intra/extra figure and representation critique.

6. Compare conditions or models and justify the chosen measurement.

7. Answer a limitation question: Forward model is reduced and µV-like; no tissue/morphology solution is implied.

Inputs → outputs. Students receive a configuration/protocol prompt, variable definitions and any required starter data. They manipulate compare ground-truth Vm with template and dv/dt extracellular modes. The reusable output is aligned intra/extra figure and representation critique.

Sources / connections / priority. Sources: [P3; P4; O2; O4; G7]. Natural follow-ons: EXT-02, EXT-04. Development priority: Core module. Boundary: Forward model is reduced and µV-like; no tissue/morphology solution is implied.

EXT-02 — Electrode geometry and tetrode projection

Concept / theme Distance and orientation change the multichannel amplitude pattern that supports unit separation.
Audience / context Advanced undergraduate. Formal practical, workshop or modular course. Prior modules: EXT-01.
Logistics 60 min; 1–3; boards: 0–1; software: Spikeling extracellular GUI; Jupyter/SpikeInterface; prepared dataset: GUI-generated or prepared timestamped traces. Equipment: Standard USB cable; worksheet/protocol card.
Mode / stages GUI simulation. Stages: 6–11.
Spikeling relationship 6 — GUI simulation workflow; 8 — External computational comparison

Learning outcomes.

  • Identify and predict the principal behaviour described in distance and orientation change the multichannel amplitude pattern that supports unit separation.

  • Configure or document move source/electrode geometry or load saved tetrode geometry and record source/contact distance, channel amplitudes.

  • Measure, calculate or compare move source/electrode geometry or load saved tetrode geometry using an explicit operational rule.

  • Interpret the result and state why projection law is pedagogical and clipped; not an exact volume-conductor model

Roadmap.

1. Initial prediction or classification.

2. Configure the board, GUI, simulation or dataset and record metadata.

3. Move source/electrode geometry or load saved tetrode geometry.

4. Acquire or inspect source/contact distance, channel amplitudes.

5. Produce geometry-to-feature map.

6. Compare conditions or models and justify the chosen measurement.

7. Answer a limitation question: Projection law is pedagogical and clipped; not an exact volume-conductor model.

Inputs → outputs. Students receive a configuration/protocol prompt, variable definitions and any required starter data. They manipulate move source/electrode geometry or load saved tetrode geometry. The reusable output is geometry-to-feature map.

Sources / connections / priority. Sources: [P3; P4; O2; O4; G7]. Natural follow-ons: EXT-05. Development priority: High-priority extension. Boundary: Projection law is pedagogical and clipped; not an exact volume-conductor model.

EXT-03 — Recording chain: noise, hum, reference and filtering

Concept / theme Recorded extracellular data combine signal, independent/common noise, line hum, referencing and filters.
Audience / context Master’s. Formal practical, workshop or modular course. Prior modules: EXT-01, EPH-05, EPH-04.
Logistics 60 min; 1–3; boards: 0–1; software: Spikeling extracellular GUI; Jupyter/SpikeInterface; prepared dataset: GUI-generated or prepared timestamped traces. Equipment: Standard USB cable; worksheet/protocol card.
Mode / stages GUI simulation + analysis. Stages: 6–11.
Spikeling relationship 6 — GUI simulation workflow; 5 — Recorded-dataset analysis; 7 — Conceptual analogy

Learning outcomes.

  • Identify and predict the principal behaviour described in recorded extracellular data combine signal, independent/common noise, line hum, referencing and filters.

  • Configure or document manipulate noise/hum/car/band settings and compare waveform/snr and record raw/processed channels, noise level, filter/reference state.

  • Measure, calculate or compare manipulate noise/hum/car/band settings and compare waveform/snr using an explicit operational rule.

  • Interpret the result and state why filter frequencies require a valid sample rate; real amplifier/electrode artefacts are broader

Roadmap.

1. Initial prediction or classification.

2. Configure the board, GUI, simulation or dataset and record metadata.

3. Manipulate noise/hum/CAR/band settings and compare waveform/SNR.

4. Acquire or inspect raw/processed channels, noise level, filter/reference state.

5. Produce processing-chain diagram and before/after qc.

6. Compare conditions or models and justify the chosen measurement.

7. Answer a limitation question: Filter frequencies require a valid sample rate; real amplifier/electrode artefacts are broader.

Inputs → outputs. Students receive a configuration/protocol prompt, variable definitions and any required starter data. They manipulate manipulate noise/hum/car/band settings and compare waveform/snr. The reusable output is processing-chain diagram and before/after qc.

Sources / connections / priority. Sources: [P3; P4; O2; O4; G7]. Natural follow-ons: EXT-04, EXT-05. Development priority: Advanced specialised module. Boundary: Filter frequencies require a valid sample rate; real amplifier/electrode artefacts are broader.

EXT-04 — Threshold detection and waveform features

Concept / theme Spike detection trades missed events against false detections and feeds feature extraction.
Audience / context Master’s. Formal practical, workshop or modular course. Prior modules: EXT-03, DAT-03.
Logistics 60 min; 1–3; boards: 0–1; software: Spikeling extracellular GUI; Jupyter/SpikeInterface; prepared dataset: GUI-generated or prepared timestamped traces. Equipment: Standard USB cable; worksheet/protocol card.
Mode / stages GUI + Jupyter. Stages: 6–11.
Spikeling relationship 6 — GUI simulation workflow; 5 — Recorded-dataset analysis; 8 — External computational comparison

Learning outcomes.

  • Identify and predict the principal behaviour described in spike detection trades missed events against false detections and feeds feature extraction.

  • Configure or document vary threshold/refractory under controlled noise and extract peak/energy/channel features and record detections, true events, waveform features.

  • Measure, calculate or compare vary threshold/refractory under controlled noise and extract peak/energy/channel features using an explicit operational rule.

  • Interpret the result and state why threshold crossings are events, not automatically well-isolated units

Roadmap.

1. Initial prediction or classification.

2. Configure the board, GUI, simulation or dataset and record metadata.

3. Vary threshold/refractory under controlled noise and extract peak/energy/channel features.

4. Acquire or inspect detections, true events, waveform features.

5. Produce detection roc-like table and waveform panel.

6. Compare conditions or models and justify the chosen measurement.

7. Answer a limitation question: Threshold crossings are events, not automatically well-isolated units.

Inputs → outputs. Students receive a configuration/protocol prompt, variable definitions and any required starter data. They manipulate vary threshold/refractory under controlled noise and extract peak/energy/channel features. The reusable output is detection roc-like table and waveform panel.

Sources / connections / priority. Sources: [P3; P4; O2; O4; G7]. Natural follow-ons: EXT-05. Development priority: High-priority extension. Boundary: Threshold crossings are events, not automatically well-isolated units.

EXT-05 — Clustering, spike sorting and quality metrics

Concept / theme Tetrode features support clustering, but unit identity remains an inference with contamination and incompleteness.
Audience / context Master’s/doctoral methods. Formal practical, workshop or modular course. Prior modules: EXT-02, EXT-04, STA-02.
Logistics 60 min; 1–3; boards: 0–1; software: Spikeling extracellular GUI; Jupyter/SpikeInterface; prepared dataset: GUI-generated or prepared timestamped traces. Equipment: Standard USB cable; worksheet/protocol card.
Mode / stages Prepared-dataset/Jupyter. Stages: 6–11.
Spikeling relationship 5 — Recorded-dataset analysis; 6 — GUI simulation workflow; 8 — External computational comparison

Learning outcomes.

  • Identify and predict the principal behaviour described in tetrode features support clustering, but unit identity remains an inference with contamination and incompleteness.

  • Configure or document cluster synthetic waveforms, compare with ground truth and calculate refractory/snr/quality metrics and record feature vectors, labels, true unit ids, quality metrics.

  • Measure, calculate or compare cluster synthetic waveforms, compare with ground truth and calculate refractory/snr/quality metrics using an explicit operational rule.

  • Interpret the result and state why ground-truth synthetic success does not guarantee performance on biological recordings; avoid treating refractory checks as proof of isolation

Roadmap.

1. Initial prediction or classification.

2. Configure the board, GUI, simulation or dataset and record metadata.

3. Cluster synthetic waveforms, compare with ground truth and calculate refractory/SNR/quality metrics.

4. Acquire or inspect feature vectors, labels, true unit ids, quality metrics.

5. Produce sorting report with error trade-off and curation decision.

6. Compare conditions or models and justify the chosen measurement.

7. Answer a limitation question: Ground-truth synthetic success does not guarantee performance on biological recordings; avoid treating refractory checks as proof of isolation.

Inputs → outputs. Students receive a configuration/protocol prompt, variable definitions and any required starter data. They manipulate cluster synthetic waveforms, compare with ground truth and calculate refractory/snr/quality metrics. The reusable output is sorting report with error trade-off and curation decision.

Sources / connections / priority. Sources: [P3; P4; O2; O4; G7]. Natural follow-ons: EPI-01, DAT-08. Development priority: Advanced specialised module. Boundary: Ground-truth synthetic success does not guarantee performance on biological recordings; avoid treating refractory checks as proof of isolation.