The monitor holds the pressure. It does not hold that noradrenaline was started at 20:21, that the dose has not changed since, or what should have happened by now. Pairing an intervention with the response that should have followed is the argument for a second screen. The expected-response windows are stated rules, not a model — they come from the unit’s own protocol and are shown here so they can be argued with and edited.