Connection Essay Sample 2

Dominowski defines insight problems as those that require a change in representation or problem space, where "warmth ratings" show an abrupt change just prior to solution. They require that learners overcome the interpretation fostered by the problem presentation to adopt one that is more structurally appropriate. That seems like a reasonable definition, though it would encompass more than one type of experience. Yet the examples of problems Dominowski offers that have been used to examine insight qualify, I believe, as "puzzle problems." Dominowky cites Perkins as suggesting that while these puzzle-like insight problems have no significance themselves, the processes used to solve them are much the same as those used in producing more important insights. Perhaps the literature on this is rock-solid and I am simply unaware of the evidence that would convince me, but it seems to me that the type of insight needed to solve puzzle problems is very different, and in fact a poor cousin of, the type of insight that produces the meaningful connections, parsimonious solutions, and revolutionary perspectives useful in real life.

There are two main differences I can see between the types of insights required to solve puzzle problems and those useful in dealing with complex, ill-structured problems that are found in real life. First, the two types of problems elicit different cognitive strategies. Puzzle problems often possess subtle cues (through their wording or length or other characteristics) that trigger a particular type of thinking, which I can only describe as the general strategy of "Look for a trick." At this point, the learner calls up the list of tricks he or she has learned (often through the embarrassment of having failed to solve a puzzle), such as alternative word meanings, unusual uses of an object, tricky questions, etc. Real life problems call up different sets of strategies. One example might be expressed as "Look for connections." This summons a search for implications, analogies, and commonalities. Another might be "Look for unlikely sources of help," and would elicit brainstorming, examination of the problem context for alternative possibilities, etc. The difference here might be subtle, but I think it is significant.

A second difference lies in the number of helpful insights possible. For a puzzle problem, there is only one key fact to realize, and once this one possible flash of insight occurs, the solution is obvious. But it is the same key for everyone, the same insight is needed to solve the problem. I agree with Baronwski that this represents a novel idea for the learner, but because the learner quickly realizes that everyone must have this identical novel idea to reach solution, the problem takes on the feeling of a "trick." In contrast, for complex problems found in real life, though an insight may not be essential for solution, several different insights are possible. In other words, in puzzle problems one specific insight is necessary for solution, while other problems can elicit zero to many insights.

Perhaps it is merely a prejudice, but I think the second type of insight is more valuable than the first. For an example, I'll slip into writing in the second person to comment on these connection essays. Their purpose, I believe, is to encourage insight of the second type. If you collected a set of connection essays that all detailed the same insight, I believe you would be not only dismayed, but disappointed and perhaps angry. After all, you don't want students to learn the trick involved in applying theory to design; you want them to come up with their own unique, personally meaningful understanding of the implications of theory for instructional design. The first would be a mind game, the ultimate example of "guess what's in my pocket," and would be a frustrating and pointless exercise for your students. The second, with any sort of luck, might help to shape a personal philosophy of how learning occurs and how design can foster it.

My point in arguing that the type of insight required by puzzle problems is actually different from the type required by complex problems is to say that I am uncertain that we can really learn much about how insights of the second type occur and how to facilitate them by studying insights of the first type. I fear we might in fact trivialize this second type of insight, and that it might give rise to the belief that we, as educators, have some control over its occurrence. We're simply not that powerful. We may be able to identify characteristics of environments and problems that have produced insights, but we cannot summon them by command, the way that puzzle problems would lead us to believe. Epiphanies are wonderful not only because they are rare, but because they are unscheduled and completely personal.

This helps me to reflect in a new way on something I've always found a bit unsettling in anchored instruction, at least as far as the Jasper series illustrates it, but which I've never been able to articulate. Jasper problems have some of the characteristics of puzzle problems. First, they require specific, prespecified insights, and in each case, only one insight will actually help to solve the problem. Second, I think students who have worked on several Jasper problems might develop a bag of tricks, strategies that work well to figure out Jasper problems but which have little to do with solving the complex problems of real life. These strategies might include, "Look for hidden clues in the story," "Don't be fooled by irrelevant information," and "The most obvious answer is probably not the correct one." These are not the most useful strategies in real life learning and problem-solving, though they do perhaps have their value. I wonder if it is not this element, rather than their unfamiliar contexts, that most undermines the claim that Jasper problems are authentic. Jasper problems stand in contrast to the earlier type of anchored instruction environment Bransford, Vye, Kinzer and Risko describe. In the response-contingent stimulation project, students could have a variety of insights (or none at all), and because there was no one right answer, they would not feel there is a "trick" to the situation. This activity would trigger strategies like "observe," "make predictions based on my personal theory," and "support my claims with evidence." These are strategies that work well in real life learning and problem-solving.

I don't make these comments to criticize the Jasper series; I really do think it's a marvelous educational tool. Nor do I think that convergent insights are necessarily a bad thing; rather they are simply not the only or best type of insight. I think the real importance of this distinction is to me as a designer. For me, it's tempting to slip into designing for the types of insight involved in puzzle problems. After all, as designers and educators, we want to know what it is our students will learn. We want to be certain we are not wasting valuable educational time, and to assure ourselves we are not, we want to prespecify the insights students will have and test to see if we have been successful. Yet we need to want something more than that. We need to want what Eleanor Duckworth called the "having of wonderful ideas." The only way I see of helping students to have these insights is to provide them stimulating environments and take the chance that they will have their own insights, revelations, and epiphanies, or they will not.

Duckworth, E. (1987). The having of wonderful ideas and other essays on teaching and learning. New York: Teacher's College Press.