The Limited Utility of Multiunit Data in Differentiating Neuronal Population Activity

Area of Science:

• Neuroscience
• Electrophysiology
• Computational Neuroscience

Background:

• Single neuron recordings are the gold standard for monitoring neuronal populations.
• Multiunit activity (MUA) is used as a surrogate when single neuron recordings are not feasible.
• MUA allows monitoring of many neurons but lacks specificity for neuronal subtypes.

Purpose of the Study:

• To investigate if knowledge of single unit waveforms can enable MUA to distinguish the activity of specific neuron types.
• To determine if MUA components can monitor medium spiny neurons (MSNs) and fast-spiking interneurons (FSIs) in the mouse dorsal striatum.

Main Methods:

• Experimental and computational modeling approaches were employed.
• Analysis focused on MUA components at frequencies >100Hz.
• Investigated the relationship between MUA and single unit activity under varying isolation conditions.

Main Results:

• When spikes are well-isolated, MUA at >100Hz correlates with single unit spiking and reflects neuronal timing and spectral signatures.
• MUA's dependence on waveform allows for distinguishing neuron types under ideal isolation.
• In typical recordings with non-isolated spikes, MUA lacks sufficient information to predict specific MSN and FSI population activity.

Conclusions:

• Knowledge of spike waveforms is necessary but not sufficient for MUA to reliably predict specific neuronal ensemble activity.
• MUA can monitor specific neuron types only under ideal spike isolation conditions.
• Distinguishing differential activity of neuronal subtypes using MUA remains challenging in standard electrophysiological recordings.