Positive feedback is ubiquitous in biochemical networks and can lead to a bifurcation from a monostable to a bistable cellular state. One such example is the immune self/foreign decision, which is manifested in a complex biochemical cascade. We consider a coarse-grained view of this cascade, its output and response to inhibition, and propose a theory which maps the stochastic dynamics to thermodynamic state variables. Our mapping allows experimental data analysis without curve fitting and highlights previously unknown features in the data. Beyond steady state, our theory reveals critical slowing down and a universal scaling of the response to changing stimulus. Finally, motivated by the immunological synapse, where a pair of cells communicate with each other through molecule exchange, we consider a pair of sensing/responding cells and how they might communicate effectively.