Nonequilibrium linear response for Markov dynamics. I: Jump processes and overdamped diffusions (Q846925)

The authors observe that systems out of equilibrium, in stationary and nonstationary regimes, display a linear response to energy impulses expressed as the sum of two temporal correlation functions. There are two interpretations of these function. The first makes a connection between correlation between observable and excess entropy flux yielding a relation with energy dissipation like in equilibrium state. The second one makes a connection between the observable and the excess in dynamical activity or reactivity. If one wants to know a system should known its reaction to external stimulus. Linear theory approach can be applied when the effects and the causes are comparable -- this can be done for systems in thermodynamic equilibrium basing on the fluctuation-dissipation theorem: the response of an equilibrium system to small external perturbations is proportional to its fluctuations. In the paper the authors generalize the equilibrium response formula (given by the Eq. (1)) to nonequilibrium regimes (both: stationary and nonstationary ones). This was done basing on the approach, which assumes that the generalization of equilibrium response formula can be viewed not just as the outcome of a perturbative calculation on the level of evolution operators, but rather in its relation with dissipative effects. From dynamical fluctuation theory it is known that there is more than jus dissipation or entropy production that governs the fluctuations around nonequilibrium state. A novel approach takes dynamical activity or traffic as a measure for the system's nervosity or internal reactivity -- in contrast to the entropy production this approach is time-symmetric and doesn't follow the time arrow rule. The details of this approach are given in Section 2 where the precise set-up and meaning is given. In Section 3 the reader has the extension of so far presented results into the general and usable formula to accompany it with an interpretation derived from dynamical fluctuation theory. Section 4 gives the examples and makes visible the contribution of new approach, which is based on Eq. (5), i.e., the equation that was given basing on Eq. (1) after the split of (1) in two separate terms: the entropic and the \textit{frenetic}. In Sections 5 and 6 we have the statistical mechanical interpretation of so far given deliberations and in Sect. 7 there is a relation with the ambition of effective temperature. In the Section 8 we have the connections between given paper and previous works of the same authors; the paper is crowned in Sect. 9. The paper also consists of 3 Appendixes.