Scaling Up and Scaling Down: Toward a Two-Way Street Between Science and Practice

The translation of basic human learning theory into better educational practices is a research endeavor in its own right.  The low success rate of large-scale randomized field trials in education, the majority of which have yielded no difference between treatment and control, is an indication that stronger scientific results and theories are needed to lead up to such trials.  Translational research in education is complex not just because of the millions of plausible instructional options that theory offers, but also because of the many social constraints of schooling in which theory must live and survive.  Our NSF-funded Pittsburgh Science of Learning Center (PSLC) is supporting a research method called "in vivo experimentation" that is a stepping stone between laboratory-based theory and randomized field trials.  This method involves embedding medium-scale experiments that test general principles of learning and instruction within the specific content and operation of academic courses in schools and colleges.  These experiments require the engagement of both domain experts, like chemistry instructors, and learning scientists, like cognitive psychologists.  They fuel the development of theories of learning, theories of knowledge, and theories of practice, all of which are needed to reliably improve education.   With the support of PSLC, interdisciplinary teams have used advanced learning technologies, like simulations and intelligent tutors, to run in vivo experiments to improve theory and practice in mathematics, science, and language learning.  I draw from these projects to illustrate the critical interplay between insights from the lab and insights from the field.