Integrate-and-fire network model of activity propagation from thalamus to cortex
Introduction
Complex biological systems such as cells, organisms and neural networks are characterized by outstanding abilities of self-organization and information processing (Kauffman, 1993). In the field of neuroscience, the possibility to understand mechanisms underlying encoding and transmission of sensory information across brain structures such as the thalamus and the cortex is a challenging perspective.
The thalamocortical system constitutes the vast majority of mammalian brain. It is composed by the thalamus, a symmetrical structure located near the center of the brain, and the cerebral cortex, in particular the primary sensory areas, which are direct recipient of thalamic outputs. It contributes to several cognitive functions such as consciousness, sleep, decision making, alertness and so on (Sherman, 2016). In particular, the thalamocortical system is a processing center of the information coming from the external world. Most of sensory inputs coming from the peripheral system are projected to the thalamus, which in turn sends them to specific cortical areas (Yoshioka et al., 1994). Therefore, it is important to investigate the role of the thalamus in the mechanisms of transmission and processing of sensory information. For decades, the thalamus has been considered simply a relay station lacking mechanisms of computing and processing of sensory inputs (Sherman and Guillery, 2001). Instead, recent experimental and theoretical studies have revealed that the thalamus performs a pre-processing of information and modulates cortical areas through thalamocortical projections even without sensory inputs (Steriade et al., 1993a, Constantinople and Bruno, 2013). In fact, the thalamus behaves as a gatekeeper selecting which information to project to the cortex. This is consistent with many experiments confirming thalamic influence in attention and in the transition between sleep and wakefulness states (Sherman, 2001). Moreover, several thalamic nuclei have the key role of transmitting information between different cortical areas (Sherman and Guillery, 2002). A deeper investigation on thalamic functions is therefore needed to understand the pathways that external information crosses throughout the thalamocortical system.
Detailed experimental descriptions of thalamic activity are infrequent due to the difficulty of performing large scale recordings of single units in this area. Given this limitation, computational and numerical analyses are important because they offer a detailed perspective of neural network dynamics. Therefore, consistent and robust models of the thalamocortical system are crucial for connecting experimental findings with numerical results. During the last 20 years, several thalamic models have been developed for different applications (Golomb et al., 1996, Traub et al., 2005, Vijayan and Kopell, 2012). Cortical networks, in turn, have been widely studied and various mechanisms of cortical activity are now known in detail. However, little has been done regarding the interplay between these two systems. In particular, models able to describe dynamics at single neuron scale together with mesoscopical collective phenomena are necessary for a complete comprehension of the thalamocortical system. Here we present a novel computational model focusing on the description of the connectivity from the thalamus to the cortex and accounting for the spread of thalamic activity into the cortex. Our model takes into account single neuron dynamics by describing evolution of state variables for every neuron in the network model. Furthermore, from this microscopic dynamics we derived mesoscopic variables which we specifically related to experimentally measured LFPs. During the last twenty years, LFP has gained increasing attention because it provides an insight into a wide range of coordinated cellular activities (Harris Bozer et al., 2017). Our scaled computational model describes mesoscopic activity of both thalamus and cortex through simulated LFP signals. In particular, we were interested in investigating the propagation of mesoscopic rhythms of the thalamus into the cortex due to thalamocortical connections.
Section snippets
Methods
We considered a thalamic T = {TTC, TRE} and a cortical Γ = {ΓPY, ΓINT} neural networks, both composed by an excitatory and an inhibitory population (respectively pyramidal (PY) neurons/thalamocortical relay (TC) neurons and interneurons (INT)/reticular (RE) neurons for cortical/thalamic network). We based our work on two previous papers, Refs. Mazzoni et al. (2008) and Barardi et al. (2016), describing a cortical and thalamic network model, respectively. We extended these works by adding
Results
The presence of two different dynamical regimes in the thalamus is known (Steriade et al., 1993b). However, how these thalamic activities affect cortical dynamics is still an open issue. We developed our model for an efficient numerical and theoretical analysis of thalamocortical dynamics. We studied the simulated dynamics exhibited by our model during both asleep and awake states. We expected different types of rhythmic activities at the mesoscopical level due to excitation-inhibition
Conclusion
To achieve a better understanding of information routing through the mammalian brain, we presented a novel model of a local thalamocortical circuitry. Our model is able to reproduce typical dynamics of the thalamocortical system. We focused on a detailed description of the system at the single-neuron level, by means of both well-known and recently observed features of cortical and thalamic cells. Differently from rate or neural mass models, spiking network models allowed us to compute
Conflict of interest
All authors declare that they have no conflict of interest in this research.
Acknowledgements
J.G.O. was supported by the Spanish Ministry of Economy and Competitiveness and FEDER (project FIS2015-66503-C3-1-P and Maria de Maeztu Programme, project MDM-2014-0370), the Generalitat de Catalunya (project 2017 SGR 1054), and the ICREA Academia programme.
E.C. was supported by the grant “PANACEE” (Prevision and analysis of brain activity in transitions: epilepsy and sleep) of the Regione Toscana – PAR FAS 2007–2013 1.1.a.1.1.2 – B22I14000770002.
A.M. was supported by internal funds of Scuola
References (60)
- et al.
Intrinsic firing patterns of diverse neocortical neurons
Trends Neurosci.
(1990) Modelling corticothalamic feedback and the gating of the thalamus by the cerebral cortex
J. Physiol.-Paris
(2000)- et al.
Maintenance of persistent activity in a frontal thalamocortical loop
Nature
(2017) Networks of spiking neurons: the third generation of neural network models
Neural Netw.
(1997)- et al.
Cortical dynamics during naturalistic sensory stimulations: experiments and models
J. Physiol.-Paris
(2011) - et al.
Sleep slow oscillation and plasticity
Curr. Opin. Neurobiol.
(2017) - et al.
Transition between functional regimes in an integrate-and-fire network model of the thalamus
PLoS ONE
(2016) - et al.
Model of thalamocortical slow-wave sleep oscillations and transitions to activated states
J. Neurosci.
(2002) - et al.
Low-frequency local field potentials and spikes in primary visual cortex convey independent visual information
J. Neurosci.
(2008) - et al.
Corticothalamic feedback controls sleep spindle duration in vivo
J. Neurosci.
(2011)
Information theory and neural coding
Nat. Neurosci.
Functions of gamma-band synchronization in cognition: from single circuits to functional diversity across cortical and subcortical systems
Eur. J. Neurosci.
Adaptive exponential integrate-and-fire model as an effective description of neuronal activity
J. Neurophysiol.
Fast global oscillations in networks of integrate-and-fire neurons with low firing rates
Neural Comput.
Dynamics of sparsely connected networks of excitatory and inhibitory spiking neurons
J. Comput. Neurosci.
Neuronal oscillations in cortical networks
Science
Mechanisms of gamma oscillations
Annu. Rev. Neurosci.
The origin of extracellular fields and currents – eeg, ecog, lfp and spikes
Nat. Rev. Neurosci.
Comparison of the dynamics of neural interactions between current-based and conductance-based integrate-and-fire recurrent networks
Front. Neural Circuits
Deep cortical layers are activated directly by thalamus
Science
Electrophysiological properties of cat reticular thalamic neurones in vivo
J. Physiol.
Control of spatiotemporal coherence of a thalamic oscillation by corticothalamic feedback
Science
Mechanisms of long-lasting hyperpolarizations underlying slow sleep oscillations in cat corticothalamic networks
J. Physiol.
Dual function of thalamic low-vigilance state oscillations: rhythm-regulation and plasticity
Nat. Rev. Neurosci.
Self-sustained asynchronous irregular states and up-down states in thalamic, cortical and thalamocortical networks of nonlinear integrate-and-fire neurons
J. Comput. Neurosci.
Cortex, cognition and the cell: new insights into the pyramidal neuron and prefrontal function
Cereb. Cortex
Synchronization properties of spindle oscillations in a thalamic reticular nucleus model
J. Neurophysiol.
Propagation of spindle waves in a thalamic slice model
J. Neurophysiol.
Extrapolating meaning from local field potential recordings
J. Integr. Neurosci.
Cited by (3)
Editorial
2020, BioSystemsThalamocortical Spectral Transmission Relies on Balanced Input Strengths
2022, Brain TopographyComputational Models in Electroencephalography
2022, Brain Topography