Mean-field modeling of thalamocortical dynamics and a model-driven approach to EEG analysis Academic Article uri icon


MeSH Major

  • Cerebral Cortex
  • Electroencephalography
  • Models, Neurological
  • Thalamus


  • Higher brain function depends on task-dependent information flow between cortical regions. Converging lines of evidence suggest that interactions between cortical regions and the central thalamus play a key role in establishing the dynamic patterns of functional connectivity that normally support these processes. In patients with chronic disturbances of cognitive function due to severe brain injury, dysfunction of this circuitry likely plays a crucial role in pathogenesis. However, assaying thalamocortical interactions is challenging even in healthy subjects and more so in severely impaired patients. To approach this problem, we apply a dynamical-systems approach to motivate an analysis of the electroencephalogram (EEG). We begin with a model for a single thalamocortical module [Robinson PA, Rennie CJ, Rowe DL (2002) Phys Rev E Stat Nonlin Soft Matter Phys 65:041924; Robinson PA, Rennie CJ, Wright JJ, Bourke PD (1998) Phys Rev E Stat Nonlin Soft Matter Phys 58:3557-3571]. When two such modules interact via shared thalamic inhibition, multistable behavior emerges; each mode is characterized by a different pattern of coherence between cortical regions. This observation suggests that changing patterns of cortical coherence are a hallmark of normal thalamocortical dynamics. In a preliminary study, we test this idea by analyzing the EEG of a patient with chronic brain injury, who has a marked improvement in behavior and frontal brain metabolism in response to zolpidem. The analysis shows that following zolpidem administration, changing patterns of coherence are identified between the frontal lobes and between frontal and distant brain regions. These observations support the role of the central thalamus in the organization of patterns of cortical interactions and suggest how indexes of thalamocortical dynamics can be extracted from the EEG.

publication date

  • October 13, 2011



  • Academic Article



  • eng

PubMed Central ID

  • PMC3176602

Digital Object Identifier (DOI)

  • 10.1073/pnas.1012168108

PubMed ID

  • 21368177

Additional Document Info

start page

  • 15631

end page

  • 8


  • 108


  • SUPPL. 3