Reason and emotion
In 2008, Andrew Elby got funded a new grant to study mathematical sensemaking with students in introductory physics classes for engineers and in a couple of their followup courses in engineering. Andy, Eric Kuo, Brian Danielak, Michael Hull and I started working on the project. As part of the project, we were doing interviews with students. We used to also meet regularly to look at the interview videos and do some close preliminary analysis playing with lots of different kinds of ideas.
Even kind of early, we were tuning into how working on physics/engineering problems brings up lots of different kinds of emotions. And these emotions would come up in lots of different situations — when stuck on a problem, or if the interviewer tried to probe some aspect of a problem more deeply, or at the resolution of some issue that led to an intuitively satisfactory solution. It was starting to inform a bit about how we were understanding the task of problem solving but initially it was kinda not very prominent in our thinking. (Andy and I published a case study on a student, Jim (pseudonym), where we modeled some of Jim’s feeling of “this is hard” as part of the what was making him feel stuck at a point in problem solving but kept epistemology as the central focus of our writing — this was published in International Journal of Science Education in 2011).
But it was in the preliminary analysis of some interviews that Brian Danielak did that we could not tear our attention away from emotion. Brian, being amazingly sensitive, would explore these issues with the student participants in our study, even when that wasn’t part of the initial protocol. In looking at an interview with a student, Judy (pseudonym), I distinctly remember when we were trying to model some of the data in words. We were playing with naming some candidate epistemological elements (such as seeing a divide between real-life circuits and their idealized models; or seeing limited utility in working with idealized models) and their connections. And noting in words how Judy’s annoyance was really important to understanding why these epistemological elements were stable in the moment. In a hoot, Andy (if I remember correctly), threw in annoyance as an element in the visual model we were drawing. It was a game-changer! We started to model how epistemological bits of the mind were connected to emotional bits.
At that time, we were still playing pretty firmly in the cognitive space. We had been noticing aspects of the students’ course contexts that mattered, and we would talk about it, but that still remained un-modeled except in our conversations. In 2010, I gave a talk at the first TRUSE conference, where I presented the Judy case study. I was kinda scared — people had warned me about talking about emotions — but I had the support of co-conspirators Brian and Andy 🙂 Perhaps it was my nervousness, I finished a 40 min talk in 25 min! And I was a little bit dejected at the reception of the argument. But post-session, Ricardo Nemirovsky encouraged me that this was important work that I should keep pursuing.
That encouragement has been a huge source of support for me ever since. Because the road to try to pursue research on emotion-cognition entanglement was not going to be easy. I tried to get some postdoctoral fellowship funding to study this and it kept getting rejected, again and again and again. And the Judy analysis sorta kinda languished a bit (even though we did do a conference paper in the meantime). Being on soft-money means that it is really difficult to pursue unfunded ideas! By the time the Judy-analysis was getting to a point where we would consider a journal submission, our project funding had already ended. It was also difficult to refine this analysis. We got some mixed feedback, and we second-guessed ourselves a lot — we felt that what we are seeing is perhaps too subtle to capture in a paper; we did not have good models for theory or methods to adapt (the literature on emotions is messy!); the story wasn’t quite as neat …
During these years, a few of us who were all tuning into our interest in emotions (Lama Jaber, Jessica Watkins, Vashti Sawtelle, Luke Conlin, Jennifer Richards, and a few others who joined later) started a group we called Affect Gang. We would meet regularly and discuss papers, emerging analysis etc on emotions. That group was what sustained me in keeping some hope alive in pursuing this line of inquiry.
Finally in 2014/15, (with a lot of reminders from Andy :D), we (Andy, Brian, and I) submitted an article to Physical Review PER on the Judy case study. The reviews were amazingly thoughtful and engaging, but also very critical. They questioned some of the basic ways in which we were modeling the data. As disappointment gradually gave way to renewed determination, I basically started doing an overhaul of how we were thinking about it. Instead of building a model that lives more closely in the knowledge-in-pieces ontology, we allowed our analysis to breathe more freely. We started structuring the analysis around the idea of co-occuring variations in epistemology and emotion. Instead of picking out “patterns” of epistemology or emotion that we see in the data, we started to focus the analysis on segments of data and messy stories of unfolding reasoning that involved epistemology, emotion, conceptions, experiences in past, hopes for the future. And I felt that this messiness somehow is more honest towards the lived experience of those moments for Brian and Judy.
The manuscript went a major overhaul and went back to PR-PER in 2016. The reviews were again amazingly thoughtful, but also critical — but thankfully allowed for revising/resubmitting rather than a reject. Some of the ways in which we were interpreting and thinking of responding to reviewer comments were pulling us back towards a more knowledge-in-pieces kinds of thinking and I struggled with that a LOT. I had tasted the messiness! I didn’t want to fit the data into a model where we only represent emotion and epistemology. The words, while long winded, allowed us to be honest to the class contexts, future aspirations, past experiences that were being shared in the interview. Those mattered! They allowed us to think about how Brian and Judy were both part of the interviewing experience. And anything we could think about as a ‘model,’ especially a visual one, hid away all that messiness. Made it reductionist. It felt to me like in trying to expand our arguments from conceptions and epistemology, we just reduced the emotions and emotional contexts into “resources” — which was deeply unsatisfactory.
It had been 7 years of working with this data. I wanted so badly to publish this, and yet felt like by telling a cleaner story, I would be doing violence to the data and to the people who created those moments that I had the privilege to watch and analyze. I was sharing all this with Chandra one day in her office. And Chandra was just like — why don’t you draw a more messy representation? I didn’t know how! So Chandra started showing me stuff that Michael Cole had drawn to represent the flow of time and linking events in time. And Chandra and I started playing with some drawings on her blackboard. For me, that was another game-changer. The diagram we drew on Chandra’s board changed a bit as Andy and I kept thinking and talking about it. … and finally after 8 years, this paper is now published!
It takes a village to get a manuscript done!
(Exploring the entanglement of personal epistemologies and emotions in students’ thinking. Ayush Gupta, Andrew Elby, and Brian A. Danielak. Phys. Rev. Phys. Educ. Res. 14, 010129 — Published 25 May 2018)