I am interested in the design and results of the recent Purdue study about students remembering more if they take an open-ended test right after being introduced to something, compared to those who do a lot of studying, or even those who make concept maps. NY Times article, January 20. - http://tinyurl.com/4ul9u8a

Two of the keys for me in the design, if I understand it correctly, are that the “test” is apparently (1) a free-form essay in which one writes whatever one can remember (2) without using the instructional material. Then one goes back to review the instructional material, followed by doing the test again.

The NY Times article quotes researchers who wonder how this works, and “constructivists” who feel challenged by this result. I think it reinforces the concept we’ve been promoting in our problem solving programs around the importance of kids connecting to their own thinking, and in particular it is aligned with why our Notice and Wonder approach works on the front end of problem solving in which students learn that they can gain a lot by paying attention to what is already happening in their minds.

My explanation would be that this test is accomplishing the same thing on the back end of the learning process, Assuming that students have encountered new ideas and information in a meaningful, engaging way in a given activity or lesson, they need to sit down without access to the instructional material and articulate their thoughts about what they just learned and were doing. The point being that they are connecting the ideas that they have, rather than constantly shifting their attention externally. This has much more power at this point in the learning cycle than studying, making concept maps, etc. in which the students don’t spend enough time articulating and making sense of their own thinking.

The problem with the way concept maps or graphic organizers are usually done is that they are presented as additional stuff or tasks to learn, and the students make them with the learning resource at hand, so it doesn’t do the work that the kind of test being discussed does where the students learn to be left along with their thoughts to gather and articulate them.

Many students who struggle with math have learned that their own thinking is irrelevant, and they discount it. They either give up or they struggle to find out what they should be thinking and write it down, without ever actually thinking it. I think that learning requires significant moments when one mulls something over, trying to make sense of it, putting one’s thinking together, and asking questions to which one wants the answer, keeping one’s attention internal, on the development and connecting of thoughts, rather than going to external sources. This, by the way, does not have to be an individual act. I think it is possible to do this in a group as well.

I would also assert that, like other problem solving and learning strategies, it is possible to get better at doing the kind of thinking the test calls for and important to make this improvement over time. It’s not simply about being put in that situation (I’d call it reflection, and it is, but there is a fair amount of other stuff around about what reflection is, so I am hesitant).

I would expect there to be other conditions than an open-ended test essay that accomplish this internal focus and organization of thought. One of our research projects here is the Virtual math Teams project (http://mathforum.org/vmt/) and colleagues were wondering if the collaborative discourse we promote there might be even more effective. I would suggest that it is not necessarily accomplishing the same thing, nor addressing the same phase in the learning. I think it could, but more importantly I think the two approaches are compatible and part of the same system and outlook.

I would say that collaborative discourse of the problem solving sort that we’ve been doing in VMT is more on the front end of creating a learning environment in which the students are similarly “left with their own thoughts”. They explore scenarios together, following their collective thinking and the questions that emerge, often without reference to any external resources beyond the ideas of the group (and we are interested in “group cognition”).

It is a different phase of the learning process. They have mostly not been working immediately after exposure to new content. To the extent that they recall and connect prior knowledge and experience, there is certainly a similar sort of thinking process but it is much farther away from the original acquisition and it is going to have less of an impact on those previous learnings, I think. More important is the impact on forming new ideas because the VMT participants initiate ideas and questions, discover, etc.—which is thinking work they could do in the test situation but are less likely to do there. I think this VMT process “leaves them along with their thoughts” on the front end of learning and makes it more likely that they will be interested and engaged, already connected to their own thinking, and thus have something rich to reflect on if they were to then do the open-ended test essay. These two learning moments go well together. Our challenge in the VMT moment is optimizing the productivity of those early interactions so that new concepts emerge.

In addition, we are concerned in VMT with learning to learn, particularly in groups. In that context we are focused on productive disposition and on collaboration skills and practices, both of which provide a new sort of fodder for a test to then solidify. There was another interesting study result recently around the value of writing about feelings about tests right before taking them and the positive impact this had on performance (University of Chicago study, EdWeek, January 13 - http://tinyurl.com/4ec8mhj).

Interestingly, I think other work that our colleague Ellen Clay is doing to help teachers develop effective instructional explanations also speaks to this model. She has them producing videos of their explanations, which could amount to a very quick set of iterations through the two-phase cycle: (1) try to express ideas just learned – (2) assess, review, lookup, and revise”. As you produce the video I think you are more likely to be “left with your own thoughts” for that brief moment, similar to the test condition.