There has been a tendency in research into order recall to favor a time-based model. Here, instead of simply positing the existence of a series of control nodes, it is posited that they (the control nodes) have an internal nature. They code for the passing of time. The basic idea is that the first item on a list is presented, and the next item necessarily follows it at a later time point. In the same way, the third item follows the second at a yet later moment in time. A function that keeps track of this temporal progression could serve to provide the context or state of the system (the control node) to which each item becomes associated. Here, Time X (State 1) would be most strongly associated with Item 1, and X + more time passing (this being State 2) would be most strongly associated with Item 2, and so on. In temporal models, there is of course no claim that an actual time is recorded by the memory function (corresponding for instance to our concepts of “three o’clock,” etc.), but what is coded is simply the passing of time, beginning at some starting point.