Before this class, I thought behavior change was an identity shift, a belief informed by my reading of “Atomic Habits” by James Clear.
However, through the readings and building our “Hook’d” project, I learned that behavior change is much more than that – it is a complex interaction of prompts, environment, and ability. Through our ethics discussions, we unpacked how tech as an industry evolved to capitalize on these principles of behavior change, artificially incentivizing engagement on social media platforms. For example, these platforms use an average of 44 different techniques to keep users engaged, including notifications, infinite scrolling, and personalized content recommendations (Evans et al.). Although our generation are basically digital natives, the rapid growth of technology has outpaced our ability to act intentionally and thoughtfully.
Nearing graduation as a CS coterm and getting ready to move to a new city has fueled my desire to build an intervention that acts as speed bumps into our lives, to help us zero in on moments and relationships that do matter, through the 10 weeks of this class.
My project experience
I loved that other students in the class shared similar sentiments, as Team Moose was not the only team focusing on post-grad challenges. That being said, I wish we got more team time to learn from other groups’ approaches to their interventions and assumptions test, as I believe this drove the core motivation behind our prototypes. As my first 247 class, I felt somewhat unprepared without more examples of past assumption tests to guide our work. I did enjoy that we could see past projects in the Higher Common Sense blog from previous students, which only validated our core problem and highlighted how pertinent our problems as new grads were.
As we designed Hook’d, I confronted questions about whether technology should mediate relationships at all. Personally I realized I have ambivalence about tech solutions—wanting to create tools that help people connect while worrying about replacing authentic interaction with digital substitutes.
Ambient displays (interface that could notify you at a glance) were effectively our projects’ nudges. By providing constant, non-intrusive feedback by depicting fish “health”, our team believed that they can gently bring users to enforce positive behavior change. We thought this design decision was less manipulative than most products that are currently on the market – think Duolingo’s spam of push notifications or Amazon’s time-urgent copy. We intentionally thought of ways to balance encouraging authentic connection without creating more anxiety: designing the fish “death” to be more comical rather than intense, keeping the AI-generated summaries of previous conversations up-front in order to reduce social friction and maintaining a light and playful design system. Talking to loved ones should not be a chore.
I was surprised to see how familiar many classmates were with ambient displays, likely due to HCI backgrounds. However, this also led me to wonder how the term “ambient display” translates outside of the classroom, beyond the Stanford bubble. I’ve actually been working on a research project with the HCI Group that centers the ambient display and progressive visual growth in relation to fitness. I loved that our prototype could explore and experiment outside the confines of the research environment, and I wonder how this project could have fared into industry. I’m curious about the technical feasibility of implementing our concept in industry, though I believe a company like Apple, with its ecosystem-focused design philosophy, might be receptive to something like “Hook’d.”
A method I would probably not try again was the Bubble Maps. In an effort to depict the user flows within the app, our team was tasked to list out our app’s main features and visually group them in bubbles. Although this issue might be project specific, I feel like they did not offer enough insights into our app design. This may be due to the fact that our app nudges behavior outside of the app environment (i.e. leaving the app to FaceTime your friends), and would be curious how Bubble Maps would work for more traditional apps.
Ethical considerations
Now I believe that the best way to promote the well-being and engagement of the user is to design and advocate for autonomy. We ensure that the user is well-informed of the permissions needed to use the app, and how the data will be used through a quick onboarding tutorial. The user also has a choice whether or not to incorporate certain contacts, allow users to customize their interaction preferences (how often they want to reach out to their chosen fish), and create different fish bowls for different contexts. In future designs, it would be interesting to give users the ability to create a fish without needing to download the contact information.
Personal growth and future applications
I liked how we had a whole session dedicated to ethics but now I’m wondering how the product development process can bake in ethical considerations from the beginning. In addition, as a product designer who understands both the technical and human aspects, I recognize my responsibility to ask not just “can we build this?” but “should we build this?”
This project also highlighted my awareness of privilege in design. I am grateful to have friends that I can reach out to on a regular basis, but our team discussions revealed blind spots in our thinking on how this app would work for users without strong relationships.
Future directions I’d like to explore with our project is how we can make this a more immersive experience for our users while maintaining user privacy. As it stands, our prototype prompts users to reach out to their loved ones with a visual widget display that represents their relationships. Insights from our assumptions test informed us that small design decisions (like how often notifications appear) directly impact user behavior and well-being. However, could we incorporate a personalized nudge that could reflect a relationship more closely while keeping them at a minimum? What would the Goldilocks number be? Would this nudge be designed into the visual widget or show up as a prompt within the app once a user clicks on the visual widget? I imagine this feature would interfere with a user’s privacy, as this design choice would require the app scraping all the users’ interactions across social media apps. I would also love to see this behavior of reaching out to friends stacked with other positive behaviors such as going on a walk. I find that I personally reach out to a lot of friends while walking back home from classes, or running to the gym. It would be cool to group these interactions or bake this habit-stacking into the app. Could it be more context aware and suggest an ideal time on my calendar to connect with others? Could the app detect when I’m on my morning run and nudge me to call a friend? Overall, these questions show my evolving thinking on what it means to have technology that promotes behavior change without compromising autonomy nor privacy. It is a constantly evolving problem, but I’m glad to have finished this class having engaged in these conversations and carrying these lessons forward.
