Coming into this class, when I thought about changing behavior in the context of design, my mind jumped to debatably unethical cases — like Apple using red colors to monopolize your attention, or TikTok’s infinite scroll. I had already taken 247G (Design for Play) and thoroughly enjoyed thinking through the psychology of game design. I also enjoyed watching people play out physical games in real-time and iterating my games in response. I was curious to see how behavior change would ‘fit into’ this design process and add onto my knowledge. To my surprise, this class gave me a more fundamental understanding of the design process than I ever had before. Thinking specifically about human behavior and how to nudge it in positive ways gave me a new framework to use to think through design work.
I loved learning about models of human behavior and how to change how humans make decisions. BJ Fogg’s Tiny Prompts taught me that no behavior happens without a prompt. Human memory is faulty and unreliable; we can instead design context prompts and anchor new behaviors in existing habits to prompt ourselves to act differently. For example, tying string around your finger can remind you to drink hot water in the morning. Similarly, I enjoyed the function [B = f(P+E)] which tells you to either change something internal to the person or external to the environment, and [B=mat] which tells you that motivation, ability, and a trigger create a behavior. I also enjoyed our learnings about rewards — like how intrinsic and immediate rewards are the best at motivating people to perform a behavior. I think this was my favorite part of the course because I appreciate these quantitative approaches to thinking about the nuanced, fuzzy phenomenon of human habit.
Something I thought could have been done better is how we decided on the behaviors we focused on for this quarter. There was not a great deal of conversation in class about different types of behaviors, and I felt that we chose our behaviors somewhat randomly from a single group conversation. Our group personally looked up a list of popular New Year’s Resolutions and chose one of them. I also felt that, after seeing other group projects in the Science Fair, some groups were stuck with behaviors that were not particularly fruitful for the project. I thought it could have been cool to have a more structured way of thinking about different potential behaviors that are genuine user problems.
In terms of the overall approach of this class, I thought the weeks were broken up extremely well into bite-sized pieces that assembled into a full project. I thought we could have had more guidance with regards to running our baseline study and intervention study — what are different types of data collection? Could we see some examples of behavior change studies that the teaching staff thinks are interesting? When we started our baseline study, our group was a little bit confused about what we were trying to measure. What types of compliments people give? How many they give? When they give them? How they give them? Why they give them? Who they give them to? Ultimately, we looked into all of these questions — and this was fruitful — but we were unsure about what we were supposed to get a “baseline” for since that implies we’re looking for a quantitative starting point. I think it could be helpful to reframe the baseline study as an experiment in gathering context on a behavior. I felt similarly about the intervention study — we weren’t sure what question we were asking. We ended up analyzing how many compliments were given, how people felt about them, what the compliments were about, and who people gave compliments to. While these were interesting pieces of information, they felt somewhat haphazard — it wasn’t clear to me what we were trying to extract from the study. I think there could have been some helpful framing about what the purpose of each study was.
In terms of tools and methods, I fell in love with some of the mapping and modeling techniques. I’d never used affinity grouping, frequency grouping, and a 2×2 matrix before. I found that experience very fruitful in my group; it definitely helped us draw out interesting insights! I especially liked doing the affinity grouping without a pre-set list of categories defined. It was cool to watch the categories evolve freeform from our collective minds. I also enjoyed the Connection Circle and Fishbone models. When we applied them to our own behaviors, I even felt like I learned new things about myself! Using them in the context of our project, they were also very helpful. It forces you to think of a problem at different moments and steps in the process, which can breed new insights.
I enjoyed how ethical considerations were centered in this class. I think that our behavior nudge (the BeReal mechanism) is ethical because it’s something the user can consciously participate or not participate in. I do think our app has an ethical issue waiting to happen in terms of bullying. It would certainly be easy for people to send hateful comments on our platform. We need to think more about how to combat this and not turn into askFM. Secondly, there’s a softer ethical concern that could occur if someone doesn’t get any compliments and sees an empty screen, causing them to feel bad about themselves. We chose not to include a public wall which might help by preventing a sense of FOMO or embarrassment. I still think this would make a user sad and I wonder how we might combat a sense of disappointment — perhaps we prompt users to send out compliments to particular friends. I very much enjoyed talking about potential negative outcomes from our app and how we would combat them; this is an important line of risk-averse thinking that I will carry with me to future projects!
Overall, I had a definitively positive experience in this class. I will bring a more nuanced understanding of human behavior, a more careful approach to ethics, and an excitement for good design to my future projects.
