Anyone know of any usability or cognitive psychology research done in the optimum number of choices for menus/navigation? I need some evidence to argue a case against a top level category menu of hundreds of options
There is a lot of research in social Psychology, HCI and even marketing that supports your case. I think that if you are having difficulty finding it, it is because your limiting your search to faceted navigation. In addition, IA suggests that having that many top level choices may mean that you need to reevaluate the structure and classification of your navigation. Faceted navigation is a system that encourages way-finding and exploration so keep that in mind. But in short, having too many choices causes the user to freeze up and be unsure where to go next. In some cases the user wil be unsure or even dissatisfied with their choices.
Here is a link to the Wikipedia article on B. SChwartz's "Paradox of Choice..." that may get you started.
Most of the news articles on this subject focus on having too many choice in a cultural context : too many coffee shops, too many shirts etc. But the same principles have been applied to web and application design. There are many HCI and other design examples of the problem of too many choices. That won't change just because you are using faceted navigation.
Be aware that it is difficult sometimes to know how many is too many, but your example of hundreds is clearly too many.