|
The issue of directionality of information flow between neural ensembles is a key question in the study of brain connectivity. Various statistical measures have been developed during the last decades, in an effort to assess the directional connectivity and magnitude of information flow through recorded EEGs, local field potentials (LFPs) and unit spike trains. One such statistical measure is the recently introduced Partial Directed Coherence (PDC: Baccala LA & Sameshima K, 2001) which is based on a frequency domain representation of linear Granger Causality. The aim of this work is to examine the connectivity between three brain regions, under isoflurane anaesthesia and acute neural activity, by applying a version of PDC, the generalized PDC (gPDC), on LFPs and unit spike data.
Hippocampal (medial and lateral) and medial prefrontal cortex (mPFC) LFP and unit spike activity were recorded simultaneously with a Plexon MAP system in isoflurane-anaesthetised male Lister hooded rats (200-350g; n = 4) using micro-electrode arrays placed in each structure. Following basal recordings, rats were administered kainic acid (KA; 1mg/kg, i.p.). A software package was developed in MATLAB for the data pre-processing, autoregressive modelling and gPDC calculation. Data from animals with histologically confirmed electrode placements in the dorsal hippocampus and mPFC were analysed through the software. The gPDC analysis detected a strong flow of information from lateral to medial hippocampal sites and a reciprocal cortico-hippocampal flow. In specific, the hippocampus is sending information to the mPFC while the latter is silent but this directionality is reversed during bursts of activity in the mPFC. Elevated neuronal activity, induced by KA administration, disrupted the cortico-hippocampal connectivity and partially reversed the intra-hippocampal one. A comparison of the connectivity profile acquired from LFPs with that acquired from pre-processed spike trains is also presented. |
|