Research Themes and Projects

Picture illustrating the combination of computational models (color snapshots) with in vivo intracellular recordings (yellow trace) to study the oscillatory behavior generated in a neuronal structure (see A model of spindle rhythmicity in the isolated thalamic reticular nucleus, Journal of Neurophysiology 72: 803-818, 1994).

The research conducted in this laboratory stands at the interface between several disciplines, such as biophysics, physics and neuroscience. The themes investigated (see below) range from the microscopic (single neurons) to the macroscopic (networks or populations of neurons) aspects of the central nervous system function. We use theoretical methods and computer-based simulation techniques to explore the complex behavior of single neurons and understand their basic integrative properties (see The integrative properties of cortical neurons in vivo, The integrative properties of thalamic neurons). This task requires to integrate details about experimental measurements of these neurons, their morphology, their biophysical properties, as well as the properties of their synaptic inputs (see Biophysical models of synaptic transmission). There is also a need for a constant and continuous exchange with experimentalists recording single cells (intracellular measurements).

At the network level, we try to understand the collective behavior of neuronal populations, which in many cases cannot be simply deduced from single-cell behavior. In cerebral cortex and thalamus, neurons are characterized by complex intrinsic properties (see above) and also influence each-other through many different types of synaptic interactions involving different classes of receptors. These networks are therefore highly complex and computational methods can be particularly pertinent in predicting their behavior. This approach was followed for the case of oscillatory behavior in thalamus and cortex (see Network models of thalamic oscillations and Network models of thalamocortical oscillations). Models can also be used to understand the genesis of pathological behavior such as epileptic seizures (see Network models of epileptic discharges). Here again, a tight relation with experimental data is needed.

A lot of attention has been focused recently on the population level modeling of large cortical networks. To do this, one needs to design population level models, using mean-field techniques (see Mean-field models of neuronal populations). This type of model is appropriate to investigate large scales, such as for example the propagating waves seen over millimeter distances in primary visual cortex, or slow-wave dynamics over centimeter distances in the whole human brain. The mean-field models can integrate biophysical features like the synaptic conductances, spike-frequency adaptation, the different cell types and their excitability, and even the heterogeneity of neurons in cortex. It can also include the differential gain of excitatory and inhibitory neurons, and its emergent properties at the mesoscopic or macroscopic scale.

Finally, another aspect of computational neuroscience is to directly provide methods to analyze experimental data. Single- or multi-electrode recordings often reveal complex behavior which may not be easy to analyze. Such complex signals can be analyzed in many different ways with the help of theoretical approaches (see Spatiotemporal analysis of electrophysiological data). In some cases, the theory can help analyzing complex, apparently random signals. This is the case for intracellular recordings of "synaptic noise", from which many useful information can be extracted (see Stochastic analysis of synaptic noise).

These different approaches have been summarized in the following review papers by Destexhe's group:

    Destexhe, A.. Intracellular and computational evidence for a dominant role of internal network activity in cortical computations. Current Opinion in Neurobiology 21: 717-725, 2011 (see abstract and PDF).

    Destexhe, A., Hughes, S., Rudolph, M. and Crunelli, V. Are corticothalamic 'up' states fragments of wakefulness? Trends in Neurosciences 30: 334-342, 2007 (see abstract and PDF).

    Destexhe, A. and Contreras, D. Neuronal computations with stochastic network states. Science 314: 85-90, 2006 (see abstract and PDF).

    Destexhe, A., Rudolph, M. and Paré, D. The high-conductance state of neocortical neurons in vivo. Nature Reviews Neuroscience 4: 739-751, 2003 (see abstract and PDF).

    Destexhe, A. and Marder, E. Plasticity in single neuron and circuit computations. Nature 431: 789-795, 2004 (see abstract and PDF).

    Destexhe, A. and Sejnowski, T.J. Interactions between membrane conductances underlying thalamocortical slow-wave oscillations. Physiological Reviews 83: 1401-1453, 2003 (see abstract and PDF).

See also the Research Grants page for more details about current funding and on-going research projects, as well as possible PhD or postdoc opportunities.

Research themes of the laboratory and overview of publications

(each link below gives, for each topic, a summary, a list of the published work in the laboratory, and PDFs copies of the corresponding publications)

1. Modeling ion channels and extracellular potentials

1.1 Biophysical models of synaptic transmission

1.2 Model of the hyperpolarization-activated current Ih and its regulation by calcium

1.3 Models of local field potentials

1.4 Models of neuronal magnetic fields

2. Modeling the integrative properties of single neurons

2.1 The integrative properties of cortical neurons in vivo

2.2 Integrative properties of thalamic neurons

2.3 Model of hyperpolarization-activated persistent activity

2.4 Generalized cable equations for neurons

3. Network models

3.1 Models of thalamic oscillations

3.2 Thalamocortical oscillations

3.3 Role of sleep in memory consolidation

3.4 Models of epileptic discharges and absence seizures

3.5 Networks of "silicon" neurons

3.6 In vivo like states in cortical networks

4. Mean-field and large-scale models

4.1 Macroscopic models of neuronal activity

4.2 Mean-field models of neuronal populations

5. Computational methods to analyze experimental data

5.1 Spatiotemporal analysis of electrophysiological data

5.2 Analysis of synaptic noise from intracellular recordings

5.3 The Active Electrode Compensation (AEC) method for high-resolution intracellular recordings

5.4 Analysis of multi-electrode and ensemble recordings

5.5 Power-law and frequency scaling in electrophysiological data

6. Dynamic-clamp experiments

6.1 The Dynamic-clamp: real-time interaction between models and living neurons

(see also The Active Electrode Compensation (AEC) method above)

7. Theoretical and numerical methods

7.1 Conductance-based integrate and fire models

7.2 Event-based integration algorithms for conductance-based integrate and fire networks

For more information, please contact:

Department of Integrative and Computational Neuroscience (ICN),
Paris-Saclay Institute of Neuroscience (NeuroPSI),
CNRS, Bat 33,
1 Avenue de la Terrasse,
91198 Gif-sur-Yvette, France.

back to main page