The papers in this special section present approaches to either tracing or improving the prediction of outcomes or effects of R&D activities in programs or organizations. The effects of the creation of knowledge are notoriously difficult to predict and, when effects are observed, very difficult to trace back to specific research activities or R&D investments. Both of these problems, one at the front end of the R&D process and the other at the back end, are crucial concerns in research evaluation. The first is always required of good planning and the justification of new programs and investments. The second is inherent in demonstrating that investments made and past activities have either resulted in the desired outcomes or, if not, what changes would have to be made for future success. In sum, these are problems at the heart of evaluation as the ‘nervous system’ of policy learning. Copyright , Beech Tree Publishing.