Fig. 1.

Flowchart demonstrating the analysis pathway for both Experiments. SAM beamforming is used to identify the peak location of gamma responses for each participant and each session (a). This allows a virtual sensor recording to be estimated at this location, and spectral density estimates constructed for the stimulation time (green columns). These spectral density functions are pre-whitened and then normalised so that the global mean spectra have an area-under-the-curve of 1, but individual relative differences in amplitude are preserved. Model spectra are then generated by the canonical microcircuit model (CMC) shown in (b) and the DCM fitting procedure generates model spectra (shown in red) that best fit the true spectral density for each participant/session, shown in blue.