Peering Inside the Black Box of Expertise

By: Jeffrey Bernstein, Professor of Political Science and Director of Bruce K. Nelson Faculty Development Center on October 17th, 2022

Rogers Hornsby was one of the greatest baseball hitters of all time; among other feats, he has the second highest single-season batting average in history (.424 in 1924, for the St. Louis Cardinals). Following a successful playing career, Hornsby tried managing, where he was not especially successful (side note: Tom Hanks’ character in A League of Their Own refers to being one of Hornsby’s players). Hornsby’s final job in baseball was as hitting coach for the 1962 Mets, who had the worst single-season record in the post-expansion era. He died following the 1962 season, presumably not because of how bad the Mets were.

Hornsby, it turns out, is not unique in being a great practitioner of an art (in this case, hitting a baseball) but not an especially good teacher of said art. Teaching someone to hit a baseball (or solve differential equations, or design a prosthetic limb), as we all know, requires a different skill set than actually doing any of these things. Teaching requires us to identify and break down the hidden moves we make in doing the activity; these cognitive moves likely come so naturally to experts (like we are in our disciplines) that we cannot necessarily break down and explain what we are doing.

Quick example: how do you merge onto a highway when driving? Most of us can do this routinely; we might even say we have achieved expertise with this skill. Now try explaining how to do this to a novice. Take it from my personal experience, it is not easy. Breaking down all the steps we are taking, in order, in real time, is hard. Merging onto a highway, it turns out, is much easier than teaching someone how to merge onto a highway (my older son will back me up on this point 100%!).

Taking our expertise and breaking it down so we can teach our novice students is one of the biggest challenges we will face in the classroom. The very nature of our expertise makes it hard to be aware of how we know what we know. When I put tables of quantitative public opinion data up on the screen las week, I knew exactly what the numbers meant, and how I was interpreting them. Doing so is as natural to me as walking. There is, however, a great deal of knowledge operating behind the curtain that lets me do this. Knowing, for instance, that a result is statistically significant at the .05 level is not an intuitively logical thing to explain to students, especially because much in my brain invisibly undergirds the entirely obvious (to me!) meaning of this concept. How do I, as a professor, bridge that divide with my students?

Now, think about yourself. What do you know, completely and thoroughly, that is as obvious to you as hitting a baseball was to Rogers Hornsby, but yet is not easy to teach to students (or to me)? The nature of a university – the fact that you have all studied your discipline so much more deeply than I can ever hope to do – means you possess specialized knowledge that few others have. How can experts in specific knowledge domains share their knowledge with relative novices, like our students?

My own approach to this requires “thinking aloud” about the heretofore invisible actions I am performing. For example, after struggling to teach my older son to merge onto a highway, I changed my approach with my younger one. For Bernstein Child 2.0, I would have him sit in the passenger seat, and narrate to him (“thinking aloud”) every action I was taking, every direction in which I was looking, every judgment that I was making. In doing so, I hoped he could come to see the task as I saw it. It is amazing how much is going on in your head as you are performing acts of expertise in a particular domain. Try it; you’ll be amazed.

Likewise, while I do not do this every single time I teach about quantitative data, I often try to “think aloud” what I am seeing when I work with students. This usually manifests as my putting up the data, asking students to practice their interpretation of what is on the screen (“Turn to your partner and discuss with each other what you think you see here”), and then, later in the discussion, sharing what I am seeing. For instance, “When I looked at the table, I noticed the correlations in the right-hand column are higher than those in the left-hand column. I then asked myself, what does this mean? What do these results imply for our understanding of voter behavior?”

My hope is that after I do this a few times, and they get a feel for what it looks like when an experienced data person confronts a table of numbers, they will then be better positioned to do this on their own one day. My walking them through the hidden process I use helps to unpack the black box of expert thinking, and makes it clearer to students how they can learn to perform the acts that an expert does.

Think about this: our students are often intimidated by the very same material that we find fascinating and, frankly, fairly elementary. They feel this way because they lack the comfort with the material that we have and, lacking this comfort, they cannot find the joy. When we intentionally unpack the black box of expertise that we possess, and make it clear that such expertise is not innate but instead can be learned, we help our students on the path to deep learning of the material. And that, I would assert, is one of the most critical, and rewarding, things we can do as instructors.