Opening the black box toward a modular approach to spike sorting.

A modular framework improves spike sorting accuracy and efficiency, outperforming Kilosort4 on dense recordings and enabling detailed assessment of individual algorithm steps.

• Why it matters: Current spike sorting tools are often monolithic, making it difficult to evaluate and optimize each step, which hampers progress with increasingly complex high-density recordings.
• What they did: The authors developed a modular, benchmarked framework for key spike sorting components—peak detection, feature extraction, clustering, and template matching—using biophysically realistic ground truth data.
• The result: This approach enables precise performance evaluation, leads to a superior component-based spike sorter, and highlights probe motion as a major bottleneck, fostering community-driven improvements.