Predicting local field potentials from spiking activity reveals state-dependent cortical dynamics
How local spiking activity gives rise to local field potentials (LFPs) across brain regions and states remains poorly understood. We combined multi-electrode recordings from the prefrontal and premotor cortex of two rhesus monkeys with a linear autoregressive model to predict LFPs from past multi-unit activity (MUA) during wakefulness, anesthesia, and recovery. We probed the spike-to-field mapping through LFP predictability. Under anesthesia, neural dynamics were characterized by structured, low-dimensional mappings. In contrast, wakefulness exhibited higher dimensionality and reduced predictability, indicating richer, more globally coordinated dynamics, whereas awakening showed intermediate, region-specific properties. Despite a breakdown of inter-area correlations during anesthesia, inter-area reconstructions remained robust and preserved LFP correlation structure, revealing large-scale constraints on cortical dynamics. The mapping pointed to state-dependent interactions between cortical areas and local adaptation-related processes. Despite its simplicity, the linear model captures key features of mesoscale organization and provides a link between local spiking activity and large-scale network dynamics.
Authors
- Stefano Ferraina (ORCID: https://orcid.org/0000-0001-6100-2438)
- Camille Mazzara (ORCID: https://orcid.org/0000-0002-1630-7803)
- Giampiero Bardella (ORCID: https://orcid.org/0000-0002-3731-8202)
- Maurizio Mattia (ORCID: https://orcid.org/0000-0002-2356-4509)
- Sofia Raglio (ORCID: https://orcid.org/0000-0003-0500-7650)
- Andrea Galluzzi
Institutions
- Istituto Superiore di Sanità (IT)
- Sapienza University of Rome (IT)
Publication Details
- Journal
- iScience
- Published
- 2026-11-14
- DOI
- https://doi.org/10.1016/j.isci.2026.118219
- Primary Topic
- Neural dynamics and brain function
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Ministero dell'Istruzione e del Merito