Bandwidth

Bandwidth is the range of frequencies your recording system can accurately capture, determined by analog filters and sampling rate. It defines what types of neural signals you can observe.

Why it matters: Different neural signals occupy different frequency ranges. Your bandwidth must encompass all signal components you want to study, or you'll miss important information.

Common bandwidth configurations:

• Wide-band (1 Hz - 10 kHz): Captures both LFP and spikes in single recording
• LFP band (1-300 Hz): Local field potentials, brain rhythms
• Spike band (300 Hz - 6 kHz): Action potentials, single-unit activity
• High-frequency (1-30 kHz): Full spike waveforms, ultra-high-frequency oscillations

Determining required bandwidth: Sampling rate must be at least 2× highest frequency (Nyquist). For spikes, 30 kHz sampling (15 kHz bandwidth) is standard. Some systems use 20 kHz for economy. Higher sampling captures more spike detail but increases data volume.

Analog filtering: Most systems apply analog anti-aliasing filters before digitization to prevent frequency folding. Digital filters can further separate LFP and spike bands after recording.