Okay, so what's knowledge complexity? Here's a bit from the methods section:
That is, a schema may be considered complex if it contains a variety of domain-specific elements that share conceptual linkages; thus, a measurement of cognitive complexity should seek to measure the number of elements and the degree to which they are associated.There are a lot of results. Go through them yourself if you want specifics, but a bunch of variables in the multiple regression significantly predict knowledge complexity. Education, obviously. A variable called Understanding (how much you understood info about it). Information seeking styles, yes, the effort one puts into it, no surprise, and a variable called media sources (the number of sources used). All those, significant. Not significant? Individual media exposure items like newspapers, TV, and the web. Not sure why they are significant but not the overall measure of number of sources, except perhaps for some multicollinearity issues.
In a final table the authors break down the complexity variable into parts (elements and connections). You don't learn an awful lot from this, but helpful nonetheless if one wants to explore this approach to knowledge.
It is interesting to see media sources is a positive predictor of the "elements" dependent variable but not the "connections" dv. In other words, the more sources you use, the more "things" you remember about global warming but you don't see any more connections than someone with fewer sources. That's interesting and deserves deeper thought.