Figure 1.

Proposed scheme of subnetwork participation of deep layer cells by amplitude variations of beta oscillations. (A) Canolty et al. (2012) have shown that in deep cortical layers the firing rate of single cells show a highly robust sigmoidal relation to the amplitude of beta oscillations in the local field potential. Some cells (cell 1 and 2) fire strongest during high beta amplitudes (gray shading, left panel), while other cells (cell 3 and 4) fire strongest during suppressed beta amplitudes (gray shading, right panel). The amplitude-to-rate mapping is consistent across recording sessions, suggesting that a high beta amplitude cortical state (left panels) indexes the activation of a selected subnetwork of deep layer cells. When beta amplitudes change over time, e.g., during the planning and execution of a movement, the subnetwork of cells with high firing switches (right panels). (B) In addition to the overall state-change, the slope of the amplitude-to-rate mapping switches reliably and reversibly when subjects engage in different tasks. This finding shows that the rank-ordering of a subset of cells according to their firing rate switches with the specific task. This task-specific re-ordering of cells could signify the formation of task-specific subnetwork (subnetwork for task A and task B in the example sketch). (C) The selection of subnetworks illustrated in (A) and (B) could be realized by a recurrent winner-take-all (WTA) circuit in deep cortical layers. It is open, however, whether subnetwork selection is mechanistically caused by beta rhythmic input to deep layer cells (left panel), or whether beta rhythmic modulation follows from cell activity after they have been selected by task- and state-specific inputs (right panel).