Computational model of thalamo-cortical networks: dynamical control of alpha rhythms in relation to focal attention
Introduction
EEG oscillations that occur during arousal, sleep and absence seizures are generated in the thalamic and cortical networks that are mutually connected (Steriade et al., 1993, McCormick and Bal, 1997). The alpha rhythms (8–13 Hz) are seen in the EEG mainly during relaxed wakefulness. It is well known that when a subject focuses attention on a particular sensory modality the corresponding alpha EEG activity decreases, as was classically demonstrated for the alpha activity of the posterior regions of the brain by Berger (1929) and thereafter studied in detail (see review in Niedermeyer and Lopes da Silva, 1999). Furthermore, it is also well known that the rhythms within the alpha frequency range of the central areas of the scalp (the so-called mu rhythms) also decrease when the hands are stimulated or a movement is executed (Chatrian et al., 1959). It has recently been shown that the (de)synchronization of the central mu rhythm or the occipital alpha rhythm is most evident in the upper frequency band between 10 and 13 Hz (Pfurtscheller et al., 2000). Pfurtscheller and Neuper (1994) made an observation that throws new light on these phenomena of (de)synchronization of alpha band rhythms, and that has led us to formulate a novel hypothesis regarding the corresponding neurophysiological substrate (Suffczynski et al., 1999). This observation is that desynchronization of the alpha activity, i.e. a decrease of spectral power within the alpha band, at a given brain location (focal area), related to a specific sensori-motor event where the subject has to make a cue-guided movement of the hand or the foot, does not occur in isolation; rather it is accompanied by an increase of spectral power of certain frequency components, i.e. an increase of synchronization, in neighboring areas (surround areas) that correspond to the same or other modalities of information processing (Fig. 1). We call this phenomenon here focal ‘event-related desynchronization’ (ERD)/surround ‘event-related synchronization’ (ERS) of rhythmic activities within the alpha frequency range (alpha and mu rhythms). For example, a hand movement results in desynchronization of the hand area mu rhythm and in synchronization of the foot area mu rhythm as well.
The cellular activity during alpha/mu rhythms is still unknown, but we hypothesized that it may share basic mechanisms with other 7–14 Hz rhythmic activities (Lopes da Silva et al., 1980), although specific types of rhythmic activities recorded from the cortex and/or the thalamus can differ qualitatively and quantitatively. Our experimental results showed that alpha rhythms of the visual cortex are generated by cortical populations centred in layers IV and V (Lopes da Silva and Strom van Leeuwen, 1977), and depend on the interaction between thalamo-cortical and cortico-cortical mechanisms (Lopes da Silva et al., 1980). The relative contribution of these different mechanisms is related to the state of sleep/wakefulness. The alpha rhythmic activities are characteristic of the relaxed awake state and are generated in intra-cortical networks, but are also under the influence of thalamo-cortical circuits. The dynamical states of these networks depend on a number of neuromodulatory systems. With respect to attentional processes and arousal the modulating role of the cortical cholinergic inputs from basal forebrain is well documented as reviewed by Sarter and Bruno (2000).
Here we focus on a part of the system that is responsible for the generation and the modulation of alpha rhythmic activities, namely the thalamo-cortical circuits. We do not model the intra-cortical networks, although the latter also certainly play an important role, for two reasons:
- 1.
Our main aim is to test whether cross-talk at the thalamic level would be sufficient to account for the cross-talk found at the level of the scalp EEG corresponding to the hand and to the foot cortical areas, as found experimentally by way of the ERD/ERS phenomenon, based on the previous anatomical and physiological work of many people regarding the functional organization of reticular–thalamic–cortical modules.
- 2.
We did not choose to make a model including a detailed description of the cortico-cortical circuits because the basic anatomical and physiological experimental data regarding the cortex is, as yet, less comprehensive than that of the thalamic–cortical circuits. Nevertheless an interesting model of neocortical networks that can account for spatial synchrony of alpha rhythms has been recently reported (Jones et al., 2000). This question is further considered in Section 4.
Therefore we consider here the basic mechanisms underlying the rhythmic activity, within the frequency range 7–14 Hz, that has been studied in more detail, the so-called sleep spindles. Sleep spindles are defined as waxing and waning waves between 7 and 14 Hz, grouped in sequences that last for 1.5–2 s and that recur periodically with a slow rhythm of 0.1–0.2 Hz (Steriade et al., 1990). The thalamic origin of the spindle waves is well known (reviewed in Steriade and Deschenes, 1984). Experimental studies in vivo (Steriade and Deschenes, 1988, Steriade and Llinas, 1988), in vitro (von Krosigk et al., 1993, Bal et al., 1995a, Bal et al., 1995b) as well as computational modeling (reviewed in Destexhe and Sejnowski, 1996) clarified the cellular and network mechanisms underlying this kind of rhythmicity. It is currently considered that, although the thalamic reticular nucleus is considered ‘the pacemaker’ for spindle oscillations (Steriade et al., 1987), in intact brains these oscillations result from reciprocal interactions between thalamo-cortical relay (TCR) and thalamic reticular nucleus (RE) cells. The RE cells receive excitatory input from TCR cells and project back to relay nuclei via inhibitory synapses. TCR cells can fire occasionally rebound bursts of spikes following recovery from hyperpolarization induced by inhibitory postsynaptic potentials (IPSPs) of RE origin. RE cells tend to fire bursts of action potentials in response to excitation from thalamo-cortical and corticothalamic cells. In both types of cells the ability to generate bursts is provided by a low-threshold (IT) calcium current (Jahnsen and Llinas, 1984a, Jahnsen and Llinas, 1984b, Mulle et al., 1986, Avanzini et al., 1989, Huguenard and Prince, 1992, Bal and McCormick, 1993, Contreras et al., 1993). A number of detailed, distributed models of thalamic and thalamo-cortical networks were recently developed (e.g. Wang et al., 1995, Golomb et al., 1996, Destexhe et al., 1998). These models have enhanced our insight of basic neuronal mechanisms and, particularly of the interactions between distinct types of thalamic neurones. The modulation of rhythmic activities, however, is most clearly reflected in the dynamic properties of neuronal populations at the macroscopic level.
The aim of the present investigation is to make a computational model of interacting neuronal populations, with the emphasis on thalamo-cortical networks along the same line as the models indicated above, in order to test hypotheses concerning the neuronal mechanisms underlying the phenomenon of focal ERD/surround ERS. We should note that there is experimental evidence indicating that the cortex plays an important role in the generation of cortical alpha rhythms (Lopes da Silva et al., 1980). Notwithstanding this fact also the occurrence of alpha rhythms characteristic of the relaxed awake state is well documented in several thalamic nuclei both in dog (Lopes da Silva et al., 1980) and cat (Bouyer et al., 1983). Nevertheless we focus here on thalamic mechanisms that are responsible for the modulation of these rhythmic activities. The main hypothesis is that the phenomenon of focal ERD/surround ERS depends on the existence of cross-talk between adjacent thalamo-cortical modules provided by the chain of mutually inhibitory reticular nucleus neurons. In the present study we approach this problem at the intermediate level between the distributed neuronal network and lumped circuit levels. That is, we don't simulate the explicit behavior of individual neurons but rather model the populations of interacting neurons integrating neuronal and network properties. Such approach may provide novel concepts regarding the neurophysiological mechanisms that play an essential role in the generation of normal and pathological rhythms in thalamo-cortical neuronal networks.
Section snippets
Materials and methods
The basic model is an extended version of the lumped model of the thalamus initially proposed by Lopes da Silva et al. (1974). The latter model was based on two interacting populations of neurons, namely thalamo-cortical relay (TCR) cells that excited reticular nucleus (RE) cells while the latter fed back on the TCR cells, inhibiting them with fast GABAA receptor mediated IPSPs. Each population was described by the time courses of postsynaptic potentials and sigmoid transfer functions, which
Results
First, we describe the results obtained by way of the single module model, and second, those of more complex models including more than one module. By way of the latter we investigated the relative roles of cortical and thalamic modules and how the interaction between different modules influences their modes of behavior.
Discussion
The main result of the two modules model is that a chain of modules of the TCR-RE neurons interconnected by mutual inhibitory synapses is capable of displaying dynamic changes in synchronization that reproduce the essential properties of focal ERD/surround ERS of the rhythms within the alpha frequency range, experimentally demonstrated by Pfurtscheller, 1992, Pfurtscheller et al., 1997 and Pfurtscheller and Neuper (1994). Indeed, these authors reported antagonistic behavior in the alpha
Acknowledgements
Piotr Suffczynski was partly sponsored by grant 8T11E 014 16 from the KBN (Polish Committee for Scientific Research).
References (81)
- et al.
Thalamic rhythms in cat during quiet wakefulness and immobility
Electroenceph. clin. Neurophysiol.
(1983) Neural aspects of anticipatory behavior
Acta Psychol.
(1999)- et al.
The blocking of the Rolandic wicket rhythm and some central changes related to movement
Electroenceph. clin. Neurophysiol.
(1959) - et al.
Paying attention to the thalamic reticular nucleus
Trends Neurosci.
(1998) Global inhibition for selecting modes of attention
Neural Networks
(1996)- et al.
Enhancement of Rolandic mu-rhythm by pattern vision
Electroenceph. clin. Neurophysiol.
(1975) Functional topography of the human mu rhythm
Electroenceph. clin. Neurophysiol.
(1978)Neural mechanisms underlying brain waves: from neural membranes to networks
Electroenceph. clin. Neurophysiol.
(1991)- et al.
Relative contribution of intracortical and thalamo-cortical processes in the generation alpha rhythms, revealed by partial coherence analysis
Electroenceph. clin. Neurophysiol.
(1980) - et al.
Alpha rhythms: noise, dynamics and models
Int. J. Psychophysiol.
(1997)
Event related synchronization (ERS) an electrophysiological correlate of cortical areas at rest
Electroenceph. clin. Neurophysiol.
Event-related synchronization of mu rhythm in the EEG over the cortical hand area in man
Neurosci. Lett.
Post-movement beta synchronization. A correlate of an idling motor area?
Electroenceph. clin. Neurophysiol.
Foot and hand area mu rhythms
Int. J. Psychophysiol.
Functional dissociation of lower and upper frequency mu rhythms in relation to voluntary limb movement
Clin. Neurophysiol.
Event-related EEG/EMG synchronization and desynchronization: basic principles
Clin. Neurophysiol.
Tonic and burst firing: dual modes of thalamocortical relay
Trends Neurosci.
Involvement of primary motor cortex imagery: a neuromagnetic study
Neuroimage
Regulation of slow potential shifts in nucleus reticularis thalami by the mesencephalic reticular formation and the frontal granular cortex
Electroenceph. clin. Neurophysiol.
Dynamics of the human alpha rhythm: evidence for nonlinearity?
Clin. Neurophysiol.
The thalamus as a neuronal oscillator
Brain Res. Rev.
Basic mechanism of cerebral rhythmic activities
Electroenceph. clin. Neurophysiol.
Excitatory and inhibitory interactions in localized populations of model neurons
Biophys. J.
Model of biological pattern recognition with spatially chaotic dynamics
Neural Networks
The Berger rhythm: potential changes from the occipital lobes in man
Brain
Intrinsic properties of nucleus reticularis thalami neurones of the rat studied in vitro
J. Physiol.
Ionic mechanisms of rhythmic burst firing and tonic activity in the nucleus reticularis thalami, a mammalian pacemaker
J. Physiol.
Synaptic and membrane mechanisms underlying synchronized oscillations in the ferret LGNd in vitro
J. Physiol. (Lond.)
Role of the ferret perigeniculate nucleus in the generation of synchronized oscillations in vitro
J. Physiol. (Lond.)
Event-related desynchronization related to the anticipation of a stimulus providing knowledge of results
Clin. Neurophysiol.
A network of electrically coupled interneurons drives synchronized inhibition in neocortex
Nat. Neurosci.
Über das Elektrenkephalogramm des Menschen
Arch. Psychiat.
Waiting in readiness: gating in attention and motor preparation
Psychophysiology
Electrophysiological properties of cat reticular thalamic neurones in vivo
J. Physiol.
The function of the thalamic reticular complex: the searchlight hypothesis
Proc. Natl. Acad. Sci. USA
Synchronized oscillations in thalamic networks: insight from modeling studies
In vivo, in vitro, and computational analysis of dendritic calcium currents in thalamic reticular neurons
J. Neurosci.
Mechanisms underlying the synchronizing action of corticothalamic feedback through inhibition of thalamic relay cells
J. Neurophysiol.
Mass Action in the Nervous System
A network of fast-spiking cells in the neocortex connected by electrical synapses
Nature
Cited by (194)
Disruptions in brain functional connectivity: The hidden risk for oxygen-intolerant professional divers in simulated deep water
2024, Biocybernetics and Biomedical EngineeringSpatial characterization of functional neural activity during lower limb motion through functional connectivity
2024, Biocybernetics and Biomedical EngineeringMethod for quantifying arousal and consciousness in healthy states and severe brain injury via EEG-based measures of corticothalamic physiology
2023, Journal of Neuroscience MethodsPrimary somatosensory cortex sensitivity may increase upon completion of a motor task
2023, Neuroscience Letters