Interdisciplinarity and innovation catalysts

A recent post in the HBS Working Knowledge site about the book by Amar Bhidé The Venturesome Economy argued in terms of the world not being as flat as Friedman made us think in The World is Flat. Geographic specialization still plays a role, which explains why certain areas of the world tend to specialize in a given industry. In retrospect, Bhidé's argument is an extension of ideas by the now Nobel Prize winner Paul Krugman in his classical article History and Industry Location.

Specialization is an interesting phenomenon, in that it generates economies of scale, which is Krugman's main argument to explain the phenomenon of geographic aggregation. But then something doesn't seem right, since hyperspecialization tends to narrow down the scope of a field and, one would think, decrease innovation. Apparently, this contradiction can be explained by an earlier finding from Feldman, which demonstrates that although geographic areas tend to specialize in a given industry, in order to achieve innovation the region still needs to have access to a variety of methods. This multiplicity of methods will lead to innovation.

The idea of multiple methods focusing on a single proble is at the heart of interdisciplinarity, Research on Research being a good example. The question that still remains is for the role of a catalyst that make specialists in multiple methods successfuly talk to each other while focusing on a single problem. There seems to be some previous evidence that these interdisciplinary catalysts make the system work more smoothly, but at this point we don't seem to be training them, the educational system still focusing on the generation of specialists. Is it that their number is too small to deserve special attention, or are these catalysts still under the radar?