Abstract

Entropy production, or dissipation, characterizes irreversibility and nonequilibrium states deviating slightly from equilibrium conditions. However, far from equilibrium, dynamical fluctuations also depend on nondissipative, time-symmetric aspects such as the average jumping rate or the diffusivity of the system. These features, which in the years acquired names such as “traffic,” “dynamical activity,” or “frenesy,” are kinetic quantities measuring the degree of agitation and are essential for developing a statistical mechanical description of generic systems far from equilibrium. Examples show how frenetic aspects help understand the nonequilibrium linear response. Writing such a response as a difference between a dissipative and a frenetic term allows for identifying homeostatic regimes in biological conditions, where both terms cancel each other, or even regimes with a negative differential response. Frenesy also constrains the precision of currents (in a “kinetic uncertainty relation”) and even provides a lower bound to the entropy production in entirely irreversible conditions. This lower bound empirically may be more precise in estimating dissipation than directly applying the theoretical expression for the entropy production in Markov jump systems.