Self Discovery Versus Direct Instruction

This week’s article summary is The Effects of Telling First on Learning and Transfer.

A pedagogical conundrum for teachers is how often to directly instruct students and how much to let them self-discover?

My guess is most teachers support self-discovery as an ideal, yet in my case I often struggled to provide enough of it in my classes as a history teacher.

About 7 or 8 years ago, I visited a public high school that had totally committed to self-discovery. Teacher lecturing was frowned upon. I visited a few classes—mostly science/math combos—in which the students were asked to complete projects with minimal background information or explanation from teachers. Teachers moved around the classrooms observing student groups working on projects and often offered comments, but they didn’t share their expertise and knowledge with the students. It was pretty evident as I observed each group that the students were frustrated: they knew where the project was supposed to go (if I remember correctly, it was something to do with determining how to land a space craft on Mars) but they were adrift and puzzled with how to get there. They clearly didn’t possess the requisite math and physics background. I saw a lot more blank stares than idea generation. Finally I asked a student how she enjoyed this new type of self-discovery, DIY, project-based-learning, design-thinking teaching; she whispered conspiratorially so not to be overheard, “I just wish the teacher would call the whole class together and give us some direction. I really miss lectures. I think I might transfer to the district’s other public school.”

As teachers, we all know the importance of prior knowledge (see last week’s article summary) as a significant boon to subsequent learning, and typically it’s the teacher who supplies that background content in order for students to then acquire new knowledge.

As such, I’ve always been skeptical of project-based learning as a means of learning new knowledge. It is often fun and can be used to apply what you’ve learned, but to me it’s an inefficient and ineffective way to learn new content.

With all that said, I was surprised by the findings in the article below: self-discovery, in fact, does support long-term learning better than direct instruction.

Due to confirmation bias (it’s tough to unlearn something), it will take more than one research study for me to change my opinion.

But this article provided some healthy cognitive dissonance to make me reconsider my thoughts on self-discovery.

Joe

The most common pedagogical sequence in U.S. schools it to tell students the important principle or skill up front and then have them practice on a set of well-designed problems. This approach is a convenient and efficient way to deliver accumulated knowledge.

Nevertheless, many scholars are working on instructional alternatives, for instance, having students wrestle with a problem through a project, inquiry, or guided discovery and only then revealing the “answer” or underlying principle. The mechanics of these alternatives withhold didactic teaching at first lest it undermine the processes of discovery. The theory is that students first need to experience the problems that render told knowledge useful.

But is the experience-first approach effective?

In a recent study, researchers compared eighth-grade teachers who used telling-first and those who used an experience-first approach. Students’ initial recollection and test performance was the same in both groups, but long-term transfer was significantly better in the experience-first group. Why?

The researchers believe it’s because:

• Telling-and-practice pedagogy prompts students to apply solutions, one problem at a time, which reduces their chances of seeing similarities across cases.
• Giving students the end-product of expertise too soon short-cuts the need to find the deep structure that the expertise describes.
• Without an appreciation of deep structure, students are less likely to see the structure in new situations that differ on the surface, and they will fail to transfer.”

That said we aren’t in favor of frustrating students with uncertainty. There is definitely a time for telling.  It’s important for teachers to help students tolerate the short-term ambiguity of not being told the right answer. The effort to find and characterize the structure can improve learning and test performance in the long run.