Neural Mass Models of Cortical Circuits

Development of neural mass models to account for observed neurophysiological phenomena (e.g., resonance, spontaneous and evoked activities, auditory habituation) and explain cognitive functions (e.g., stimulus gating/priming, working memory, regularity formation, and change detection).

Cognition emerges from hierarchically related cortical functions that are supported by non-specialized neural circuits. The understanding of cortical functions and neural circuits has to be established from limited information from measurements (e.g., behavioral data, structural and functional brain imaging, MEG/EEG, cellular-level recordings). In this context, modeling is helpful for investigating the structure-and-function relationship in the brain (as illustrated in Fig. 1). Our studies cover three subsections. First, the dynamics through the interaction among neural populations are investigated (i.e., bifurcation analysis) [1]. Second, neural mass models are used to account for observed neurophysiological phenomena, such as resonance [2], spontaneous and evoked activities [3,8], and auditory habituation [4]. Third, the models are also linked to basic cortical functions (e.g., stimulus gating/priming, working memory, regularity formation, and change detection) as building blocks that support cognition [5,6,7]. Hence, our models are aimed at not only generating simulated data fitting observed MEG/EEG or fMRI signals, but also providing biological realizations of cortical functions for further extension and experimental verification. The models now can be easily implemented in PyRates [9], a Python-based simulator that enables parallelization to speed up simulation time.

[1] A. Spiegler, S. Kiebel, F. Atay, T.R. Knösche: Bifurcation analysis of neural mass models: impact of extrinsic inputs and dendritic time constants. NeuroImage 52(3), 1041-1058 (2010)

[2] A. Spiegler, T.R. Knösche, K. Schwab, J. Haueisen, F.M. Atay: Modeling brain resonance phenomena using a neural mass model, PLoS Computational Biology 7(12) (2011)

[3] M.N. Trong, I. Bojak, T.R. Knösche: Associating spontaneous with evoked activity in a neural mass model of visual cortex. NeuroImage 66, 80-87 (2013)

[4] P. Wang, T.R. Knösche: A realistic neural mass model of the cortex with laminar-specific connections and synaptic plasticity – evaluation with auditory habituation, PLoS ONE 8(10) e77876 (2013)

[5] T. Kunze, A.D. Peterson, J. Haueisen, T.R. Knösche: A model of individualized canonical microcircuits supporting cognitive operations. PloS ONE, 12(12) (2017)

[6] T. Kunze, J. Haueisen, T.R. Knösche: Emergence of cognitive priming and structure building from the hierarchical interaction of canonical microcircuit models. Biological cybernetics, 113(3), 273-291 (2019)

[7] S.C. Chien, B. Maess, T.R. Knösche: A generic deviance detection principle for cortical On/Off responses, omission response, and mismatch negativity. Biological cybernetics 113(5-6), 475-494 (2019)

[8] H. Finger, R. Gast, C. Gerloff, A.K. Engel, P. König: Probing neural networks for dynamic switches of communication pathways. PLoS Computational Biology, 15(12) (2019)

[9] R. Gast, D. Rose, C. Salomon, H.E. Möller, N. Weiskopf, T.R. Knösche: PyRates—A Python framework for rate-based neural simulations. PloS ONE, 14(12) (2019)