MIT Press

Figure 1:

$Neural oscillations depend on the refractory period. (A). A network representation of the model. Left: Neural activity represented as a Markov chain. Here the rates (pQA, pAR, and pRQ) determine the occupation probabilities of the three states: Q, A, and R. Right: The fraction of neurons in each state evolves over time. Active neurons (yellow) can either excite (as shown) or inhibit neighboring quiescent neurons (blue) by modulating the average firing probability pQA. Complex oscillations emerge when these neurons must wait to fire again (red). (B). Without the refractory state (left), the fraction of neurons in each state does not evolve over time (right).$

Neural oscillations depend on the refractory period. (A). A network representation of the model. Left: Neural activity represented as a Markov chain. Here the rates (⁠ $p^{Q A}$ ⁠, $p^{A R}$ ⁠, and $p^{R Q}$ ⁠) determine the occupation probabilities of the three states: $Q$ ⁠, $A$ ⁠, and $R$ ⁠. Right: The fraction of neurons in each state evolves over time. Active neurons (yellow) can either excite (as shown) or inhibit neighboring quiescent neurons (blue) by modulating the average firing probability $p^{Q A}$ ⁠. Complex oscillations emerge when these neurons must wait to fire again (red). (B). Without the refractory state (left), the fraction of neurons in each state does not evolve over time (right).

This Feature Is Available To Subscribers Only