English: (A) Network of three nodes with one driving and one modulatory input. Connection strengths in this network represent the first simulation scenario (see text for details). For the second scenario these values were reduced to half. (B) The intrinsic structures of a single brain area as described by different DCMs. This intrinsic structure was also subject to variation between the two scenarios (besides the extrinsic connections): S-DCM assumed first an excitatory self-connection σ = 1 Hz and then σ = 0.5 Hz; 2S-DCM assumed the excitatory self-connection and an inhibitory–excitatory connection (σ, μ) = (1, 0.25) Hz for the first scenario and (σ, μ) = (0.5, 0.125) Hz for the second scenario; P-DCM assumed first (σ, μ, λ) = (1, 0.4, 0.15) Hz and then (σ, μ, λ) = (0.5, 0.2, 0.15) Hz, where the inhibitory gain factor λ stayed constant for both scenarios. (C) A schematic illustration of the model specification by the means of driving and modulatory inputs for S-DCM (left), 2S-DCM (middle), and P-DCM (right) as assumed within the model inversion.