In the 60s and 70s, Thomas Allen of MIT studied how information is discovered, interpreted and brought into an organization. He particularly studied how information was brought into the US national labs. What he discovered was a network of people, whom he called "gatekeepers" who each were responsible for some facet of knowledge. They took external knowledge, interpreted it in terms of the needs of the organization and made it available to others. These gatekeepers were set up within an ad hoc network. None of them would know the complete extent of the network but each would know his/her local connections within it. ordinary people within the organization would be connected to one or more of the gatekeepers to obtain the relevant interpreted knowledge that they required.
One of his results was counter-intuitive to me until I understood just what he was getting at. He discovered that the success of a project at US national labs of that time was inversely proportional to the amount of preliminary research in the literature done by the participants. That is the more that a research team consulted the published literature, the less likely that they were to succeed in their project. The reason for this is that these were the people who did not take advantage of the tacit knowledge of the organization which was generated by the gatekeeper network. The most valuable knowledge within an organization was the constantly updated and refined knowledge contained within the minds of of the people within it. That is the best most useful knowledge is within Popper's Second World. The knowledge with the Third World of the published literature but that knowledge is refined and made more valuable within the Second World.
There was a lot of work done on this in the 90s in sociology and engineering departments. People wanted to understand how this worked and what tools could be used to make it work better. One name for this was the "Knowing Organization". I worked on this in the 90s but haven't done much of any since so things must have changed but I think that the basic result must still hold. For any organization in which creativity and innovation are important, the most important thing to do is to create tools that allow for serendipity and informal contact and awareness. Taking but one example, the Perimeter Institute is a world leading physics lab. Its labs are designed to be open so that people can see each other and the hallways are lined with blackboards. The idea is to encourage people to informally meet and to share ideas. The hallway blackboards allow discussions to spring up and serendipitous associations to be made. Annual meeting and conferences can be seen in the same way. The formal paper presentations are nice but the best work happens in the hallways and the receptions.
I see this research as confirming the observations made earlier in this chain about the importance of browsing and serendipity. However the workings of this in informal human Second World knowledge must be emphasized.