Risks and Certainty in Making Students Active Learners

A recent study looked at how people make decisions in the face of perceived risk (Shafir et al, 2008, Nature 453:917-20). This psychology study has nothing to do with teaching (students earn skekels for choosing dots on a screen), but hang with me for a minutes' explanation and then I'll get to my far-fetched speculating on the relation to science education. Apparently there is a line of inquiry showing that people tend to take a sure thing with a lower reward over a less sure thing with a higher reward. Take two options with the exact same expected reward. In one option, the chance of getting the reward is higher, but the value of the reward is lower. In the other option, the opposite is true. The value of the second reward is smaller, but with a higher chance of receiving it. People tend to take the more certain option—smaller reward, but higher chance—even though over many trials, you would get exactly the same total payoff either way. It's called the 'certainty effect'. In a previous paper, though, these researchers found a 'reverse certainty effect'. They tried to replicate the above result and instead found that people took the riskier, higher reward option. So under some conditions, people will choose less certain but higher value options. This current paper tries to figure out what conditions lead to people picking certainty, and which ones lead to picking risk. Their bottom line is that if the person knows exactly what the reward is in both cases, then the reverse certainty effect applies—they'll pick high risk, high reward. But if there is a lot of uncertainty in the size of the reward (noisy reward, or hard to distinguish between rewards), people pick low risk, low reward. So our behavior is governed by how well we can perceive or anticipate what the risks and rewards are rather than only by the payoff. (As an aside, honeybees seem to make decisions in the same way —you and a bee, not so different). Most science instructors now believe that teaching in an inquiry-based way, where students have to actively participate in their learning and in constructing their own knowledge, leads to a better understanding of science and a more useful skill set. But by and large, students still get most of their instruction in a passive way, through lectures, readings, and grades from multiple-choice style tests. So when students enter a class where the teaching style shifts towards inquiry, there is often a lot of discomfort. And here's where that paper got me thinking. Assuming the certainty and uncertainty effects operate on more complex tasks, such as in learning and getting grades (admittedly a big assumption), maybe that bias in our psyches can help explain students' discomfort when they first encounter inquiry-based classes. In those classes, students are asked to take more (intellectual) risks. Not just regurgitate memorization, but make leaps and conclusions for themselves. For a student who hasn't done that, the reward may not be as clear—a multiple choice test, you get the right answer, you get 100%. The more you memorize, the better you'll do, so you just need a high input of facts and a good technique for storing them and you'll do well. But an inquiry-based activity, how do you know what you'll get, especially if you are not sure whether you can think in the new style? There is a lot more uncertainty. This study says that, if the reward is uncertain, people tend to choose the lower risk, lower reward option. So far, that fits my intuition. I think most of us would agree that many students (and everyone else) are more comfortable with what they already know. What's interesting from the study is, if you make the reward more clear so the person knows exactly what they'll get, you can switch them from a certainty effect to a reverse-certainty effect where they prefer the higher risk situation. There should be a way to apply that in an inquiry class to shift the average student into embracing that higher risk learning style. Perhaps by being very clear up front how grading will be done, and what they have to do to get certain grades? Lots of instructors give out rubrics in advance of each assignment, for instance, or start with assignments that are intermediate between passive and inquiry styles so students have more intuition about the rewards. Well now I'm really reaching to the twigs of this idea so I'll stop, but it does seem like there's a message there, which dovetails with my common sense, about making students feel sure about the reward available in order to get them to take intellectual risks in their learning.