This episode has such huge implications that I didn't know what to call it. Efficiency and Innovation in Transfer, the actual name of the book chapter, seemed far too dry to put across the fundamental shifts in thinking about pedagogy, assessment, education research design, and cognitive theory that this article suggests (at least to me).

The authors suggest that the current literature on transfer of learning has too negative a view of the possibilty of transfer, and suffers from too many internal contradictions. They propose a new perspective on transfer called Preparation for Future Learning (PFL), as opposed to the generally accepted standard which they label Sequestered Problem Solving (SPS). In short, when you ask people to solve an unfamiliar problem and grade them on whether they get the right answer (SPS), they universally do badly; but when you ask them how they would approach solving the problem, including what questions they would ask (PFL), then you get a completely different perspective - not only do people do much better on the whole, but more educated people tend to do better, showing that education is apparently teaching some more broad kinds of problem solving (which is otherwise very hard to demonstrate). In addition, when you allow students a bit of time to invent solutions to problems at the start of the class before teaching the standard solution, not only do they perform just as well in a test as those who spend the whole time learning the solution, but they far outperform them on measures of transfer of learning, getting in one case almost triple the score of the control groups. Mind-blowing!

When I discovered this article I was slightly shocked and somewhat whiplashed, as I felt forced to change my mind again on discovery learning, to a third position (it has its place at the start of a class learning a new concept, where it can dramatically improve transfer of learning for students). I was sort of ready for it though, as I had been thinking recently that while I had seen a lot of research on discovery learning and more didactic approaches (such as Direct Instruction), I hadn't yet seen anything where the two were used within the same lesson. Well now I have, and the results are astonishing.

Enjoy the episode.

***

Link to paper: https://aaalab.stanford.edu/assets/papers/2005/Efficiency_and_Innovation_in_Transfer.pdf

RELATED EPISODES:

Mention of generative learning: 127. Necessary Conditions of Learning by Ference Marton

Transfer of learning: 98. Range by David Epstein; 108. Expert Political Judgement by Phillip Tetlock; 110. Transfer of Learning by Robert Haskell

Discovery learning: 88. The Failure of Constructivist, Discovery, Problem-based, Experiential, and Inquiry-based Teaching; 90. Discovery learning, the idea that won't die