Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 6 (8), e22790

Dynamic Causal Models and Physiological Inference: A Validation Study Using Isoflurane Anaesthesia in Rodents

Affiliations

Dynamic Causal Models and Physiological Inference: A Validation Study Using Isoflurane Anaesthesia in Rodents

Rosalyn J Moran et al. PLoS One.

Abstract

Generative models of neuroimaging and electrophysiological data present new opportunities for accessing hidden or latent brain states. Dynamic causal modeling (DCM) uses Bayesian model inversion and selection to infer the synaptic mechanisms underlying empirically observed brain responses. DCM for electrophysiological data, in particular, aims to estimate the relative strength of synaptic transmission at different cell types and via specific neurotransmitters. Here, we report a DCM validation study concerning inference on excitatory and inhibitory synaptic transmission, using different doses of a volatile anaesthetic agent (isoflurane) to parametrically modify excitatory and inhibitory synaptic processing while recording local field potentials (LFPs) from primary auditory cortex (A1) and the posterior auditory field (PAF) in the auditory belt region in rodents. We test whether DCM can infer, from the LFP measurements, the expected drug-induced changes in synaptic transmission mediated via fast ionotropic receptors; i.e., excitatory (glutamatergic) AMPA and inhibitory GABA(A) receptors. Cross- and auto-spectra from the two regions were used to optimise three DCMs based on biologically plausible neural mass models and specific network architectures. Consistent with known extrinsic connectivity patterns in sensory hierarchies, we found that a model comprising forward connections from A1 to PAF and backward connections from PAF to A1 outperformed a model with forward connections from PAF to A1 and backward connections from A1 to PAF and a model with reciprocal lateral connections. The parameter estimates from the most plausible model indicated that the amplitude of fast glutamatergic excitatory postsynaptic potentials (EPSPs) and inhibitory postsynaptic potentials (IPSPs) behaved as predicted by previous neurophysiological studies. Specifically, with increasing levels of anaesthesia, glutamatergic EPSPs decreased linearly, whereas fast GABAergic IPSPs displayed a nonlinear (saturating) increase. The consistency of our model-based in vivo results with experimental in vitro results lends further validity to the capacity of DCM to infer on synaptic processes using macroscopic neurophysiological data.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. Electrode Placement.
Electrode placement (silverball electrodes) in primary auditory cortex (A1) and posterior auditory field (PAF) in auditory cortex (A). The anatomical labelling of auditory fields was taken from and matched to a rat brain from our animals. The indicated scaling is in mm.
Figure 2
Figure 2. DCM and the Neural Mass Model.
A Neural mass model used to represent regions in auditory cortex. Three cell subpopulations contribute to the ongoing dynamics. These include spiny stellate cells in granular layer IV, pyramidal cells and inhibitory interneurons in extra granular layers (II & III and V & VI). Intrinsic connections link dynamics between subpopulations in each source. Dynamic states include currents, g, and membrane potentials v. Extrinsic connections enter at specific cell layers. B Functions controlling ongoing dynamics and their parameterisation. Left: Excitatory synaptic kernel, which is convolved with the input firing to produce a depolarising change in membrane potential. The function is parameterised by its height He and time constant. He is allowed to mediate the effects of isoflurane. Increases in He produce different responses, as per the arrow. Right: Inhibitory synaptic kernel, which is convolved with the input firing to produce a hyperpolarising change in membrane potential. The function is parameterised by its height Hi and rate constant κi. Both can mediate the effect of isoflurane. Increases in these parameters produce different responses as per the arrow. C Three competing hypotheses regarding extrinsic connectivity in hierarchical auditory cortex, embodied by model 1, with forward connections from A1 to PAF and backward connections from PAF to A1 (M1:FB). The reverse architecture is constructed for model 2 (M2: BF). Model 3 contains lateral connections between the regions (M3: LL).
Figure 3
Figure 3. Modelled Data.
A Time series recording from one animal in the noise condition showing increased burst suppression with increasing isoflurane dose. B Average cross-spectral density matrix representing spectral responses with prominent low frequency components for four isoflurane dose levels (Hashed line: 1.4%: green, 1.8%: black, 2.4%: blue, 2.8%: grey) as rats heard a white noise stimulus. Significant differences in spectral power are found for LFP recordings from A1 and PAF and also for their cross-spectra (off-diagonal term). Fits from model 1, averaged across animals are shown as full lines. C Average cross-spectra as per B, but for recordings and subsequent fits from the silent environment. D Log-evidence differences at the group level (relative to worse performing model M3: LL), showing very strong evidence in favour of model 1 (M1: FB) for both noisy and silent environments.
Figure 4
Figure 4. Parameter Estimates under Isoflurane.
A Average dose responses at 1.4%, 1.8%, 2.4% and 2.8% for He (green) and Hi (grey) for region A1 from white noise condition (** p<0.005,* p<0.05; error bars denote s.e.m.). Overall trial effects are positive compared to zero baseline at 1.4% for the inhibitory parameters and negative for excitatory parameters. B Dose responses for He and Hi for region A1 from silence data. C Dose responses for He and Hi for region PAF from white noise data. D Dose responses for He and Hi for region PAF from silence data.
Figure 5
Figure 5. Dose Response Curves.
A Linear components of polynomial fits for each animal individually in noise conditions for regions A1 and PAF, using a linear regression to describe the dose response of (conditional) EPSP effects (green) and using a second order function to describe the dose response of (conditional) IPSP effects (grey). B Linear components of polynomial fits for each animal individually during silence, for regions A1 and PAF obtained as per A.

Similar articles

See all similar articles

Cited by 31 PubMed Central articles

See all "Cited by" articles

References

    1. Suffczynski P, Kalitzin S, Pfurtscheller G, Lopes da Silva F. Computational model of thalamo-cortical networks: dynamical control of alpha rhythms in relation to focal attention. International Journal of Psychophysiology. 2001;43:25–40. - PubMed
    1. Jansen BH, Rit VG. Electroencephalogram and visual evoked potential generation in a mathematical model of coupled cortical columns. Biological Cybernetics. 1995;73:357–366. - PubMed
    1. Valdes Sosa PA, Sanchez Bornot JM, Sotero RC, Iturria Medina Y, Aleman Gomez Y, et al. Model driven EEG/fMRI fusion of brain oscillations. Human brain mapping. 2009;30:2701–2721. - PubMed
    1. David O, Kiebel SJ, Harrison LM, Mattout J, Kilner JM, et al. Dynamic causal modeling of evoked responses in EEG and MEG. Neuro Image. 2006;30:1255–1272. - PubMed
    1. Kiebel SJ, David O, Friston KJ. Dynamic causal modelling of evoked responses in EEG/MEG with lead field parameterization. Neuro Image. 2006;30:1273–1284. - PubMed

Publication types

MeSH terms

Feedback