What, after all, do we mean by the general term "memory"? It would seem that no event gets stored unless it first achieved affective amplification to gain attention and enter consciousness. Tomkins used the lovely term "transients" to describe the general run of events that caught our attention but were not significant enough either to trigger massive amounts of affect or to resonate with previous themes; transients are all the things that attract us for a moment but don't get remembered very much. So one of the criteria for "memory" is the intensity of the affect triggered by a thought, event, or such. Another criterion would be the degree of fit between such a thought or event and others that resemble it; this involves script formation because the brain seems to facilitate storage of things in compartments, and everything that gets into one compartment is handled pretty much the same way.
When we say that an individual doesn't have a good memory, except for the rare situation in which there is a structural defect in the system for storage and retrieval, that person most likely has an affect disorder interfering with consistent attention. ADD/ADHD would be a perfect example of this type of condition---attention flitting from one source to another, no stimulus able to hold the mind in place on a subject, no way for the individual to focus or conate or exclude from consideration other sources of attention. Attention is as much a matter of avoiding distraction as it is about focus.
Additionally, those who have difficulty focusing attention on anything will also have difficulty retrieving memory because the tasks associated with retrieval also demand affective amplification.
So I believe that disorders of attention must be associated with disorders of memory for the structural reasons just outlined.
As for the behavior and attitudes of children with ADD/ADHD, such people act as if they don't have "time" to focus on any specific memory. To the extent that they have "state-dependent memory," their memories would be of the confused state within which they acted before we started to treat them. Finally, you might remember that I suggested we use Basch's language for the difference between affect and emotion---an affect is the physiological mechanism that acts as the method for each specific form of attention, while in our formal language, an emotion is the assemblage of an affect with our memory of previous instances of such stimulus-affect-response. No affect needs memory for its action, while there can be no such thing as an emotion or a script unless there is memory of previous sequences.
I'll add to this posting later. The question is quite important, and I'd like to address the points raised by Jim Pfrommer as well.