For me, one of the biggest mysteries of the scientific method (see e.g. this post), is that it provides such a rigid structure, yet at the same time so creative. It turns out that constraints actually lead to more global, conceptual, and maybe thus more creative thinking. Jonah Lehrer in Wired provides some excellent context for a recent study that investigated this link explicitly. Highly recommended.
At the same time, this could also explain one of my other pet peeves: why concept mapping is such a powerful method. When I introduce this, I often get the complaint that it is too restrictive to think in terms of concepts and links. However, this could actually be one part of the success of this method!