Kant’s position is that that model assumes that drawing a causal connection between the fire and the paper is purely passive. If I experience event A followed by event B enough times, my mind just draws the conclusion that A is the cause of B. Kant’s point is that if my experience is only passive like that, no knowledge is possible (as Hume proved). What really happens, then, (and this is a horrible simplification that would make real Kantian’s cringe) is that the mind comes equipped with certain formal prerequisites that a perception has to meet to even be experienced at all. It’s kind of like: if a house only has a 3-foot door, you’re not going to fit a 5-foot TV into it. Only things that fit can get in. For Kant, the mind requires things like: existence in space and time, conforming to rules of causality, and things like that. If something is going to fit into my experience—and therefore, if it’s going to be something I can know—it must already exist in space and time and adhere to rules of causality. So, knowledge has a kind of active role in the making of experience—at least insofar as it determines the kinds of things it lets in.
I hope that makes some sense.