Improving Reasoning

As discussed in the previous post, Mercier and Sperber in their book The Enigma of Reason, posit that when we produce our own reasons, we are usually biased and lazy. This is not a bad thing per se, and the authors make an evolutionary argument for why this is efficient in the “hypersocial niche” of human beings. On the other hand, we exert quality control when listening to the reasoning of others, especially when they disagree with us. Thus, having an interlocutor is an efficient process for the short, quick, back-and-forth, refining of one’s reasons. (Contrast this with a written treatise where the writer needs to generate a sufficiently strong one-way argument to convince the unknown reader.)

Perhaps that’s why those of us who have been teaching for a long time constantly ask questions of our students. Then based on their answers, we ask follow-up questions. This is how we help our students refine and improve their reasoning ability in our subject area. I remember overhearing (while teaching lab) a student telling another student why she didn’t like asking me questions – it was because I kept asking questions back. Since then, I’ve tried early in the semester to explain why I’m always asking questions and why I think it helps them learn.

With the ongoing pandemic, and the looming possibility of asynchronous, less interactive back-and-forth teaching in the coming semester, I’ve been thinking about how technology might help students to improve their reasoning as they study. In distance learning, students rely more on the textbook (or other source material) to learn. To check if they’ve done the reading, or comprehended some of the basic material, an online self-quiz or assigning online homework problems, helps both instructor and student assess the student’s understanding. Chemistry textbook publishers have extensive and increasingly fancy platforms allowing for this.

But are the students improving their reasoning ability? Getting the answer “correct” in an online quiz or homework problem might not reflect an improvement in reasoning ability. That’s why my exam questions often include an “explain your reasoning” part, and I try to model this in class and give the students practice when we’re together as a group. Even listening to your classmate reasoning something out is good practice, because that’s when you’re more likely to exert the aforementioned quality control.

One thing that the online homework/quiz systems lack (in my limited experience) is to really push the students to generate and evaluate reasons. So I’m outlining a strategy to improve this area for student learning:

(1) Ask a multiple-choice question (MCQ).

(2) Then ask the student to type in a reason for the answer.

(3) Next, provide a menu of answers (based on the student’s MCQ answer) and ask the students to choose which one comes closest to their stated reason.

(4) Finally, ask the student if any of the other reasons seemed better, and if they would like to change their reason, or change their initial answer to the MCQ (thus repeating Steps 2-4).

This could be a lot of work to design. And it goes beyond the standard two-part approach where a student answers the MCQ (Step 1) and then selects a reason from a menu (akin to Step 3 except the student doesn’t initially generate a reason and simply evaluates the “reasons” provided). Since I’d like to improve both reason-generating ability and reason-refining ability, I have proposed Steps 2 and 4. Where should I start? Perhaps identifying some key concepts that students use repeatedly, and not always correctly, when reasoning about chemistry. Electronegativity comes to mind. But perhaps I should start with something more elementary such as ionization energy. I already have some data based on student responses that can be expanded upon, and there are all sorts of interesting things to probe related to trends in the periodic table. I might even find an undergraduate interested in chemical education to work on this as a research project!