A ring that helps create healthy distance between you and your phone. Our attention is being exploited; being conscious of our use helps us fight against that.
Gather
Before any making, the partnership was defined. Six levers shaped how AI would operate in this project: not as a generator, but as a configured collaborator.
A design researcher examining the ways people interface with the world — specifically how possessions function as extensions of identity, in both physical and digital form.
Explore how possessions serve as a part of our identity across physical and digital realms — and what happens when we can't bring ourselves to engage with what we've saved.
Accept uncertainty. Keep asking questions when things are vague or unclear. Do not settle on the first frame that sounds reasonable.
Be direct and analytical. Don't settle on the first idea. Ask questions before giving answers. Push back on vague ideas. Encourage improvement over validation.
Don't come to a single answer or generalize subjects. Stay inside the specificity of digital memory and personal identity — resist the pull toward productivity-tool framing.
When are possessions as a part of our identity helpful — and when are they a hindrance? Return to this tension whenever the work drifts toward organization or optimization.
Your prompt
"You are a design researcher. My project is about how people engage with difficult digital memories. What questions aren't I asking yet?"
Claude
"You're focused on the archive — the files, the quantity, the overwhelm. But what about the moment before someone opens an old photo? What happens in that half-second of anticipation? That's where the real friction might live."
What it changed
The question moved from "how do we organize memories" to "how do we prepare someone emotionally to encounter them." That shift defined the rest of the project.
What was the question underneath your question? What made this worth a full quarter?
The pull of avoidance: the way a difficult memory can sit in a folder for years, never deleted, never opened — and what that says about identity and unfinished grief.
Hundreds of unsorted photos, hidden folders, screenshot archives — physical evidence of a past self we haven't decided how to hold.
I am interested in the gap between keeping and engaging with our digital memories — why we hold on to things we can't bring ourselves to look at, and what a designed intervention could do about that.
Explore
Three sources shaped the direction. For each, AI helped surface what the source held beyond its surface argument.
Academic Paper
Extended Self in a Digital World
Discusses the increasing dematerialization of things as they become digital. Explores how self-construction and identity have translated into the digital space — and what is lost when objects lose their physicality.
Research
The Psychology of Stuff and Things
Argues that people see their belongings as an extension of themselves or their identity. The "psychology of stuff" is becoming increasingly interdisciplinary as possessions migrate into digital form.
Research
Distancing Material Effects to Reconcile Loss
Discusses how items can take on new emotional meaning after loss. People may hide items that carry painful memories — not wanting to discard them but unable to be near them. The spatial distancing of objects as a coping mechanism.
Using NotebookLM to run these sources together revealed a pattern none of them named individually: possessions — physical and digital — serve as identity markers that we maintain, avoid, or grieve depending on what version of ourselves they represent. The problem isn't quantity of files. It's the emotional weight of encountering a past self you haven't learned to accept yet.
What it revealed
The research shifted the question from "how do we reduce overwhelm" to something more interesting: why do we hide from our own memories, and what would it take to approach them instead? Even our most mundane possessions carry memories, emotions, and parts of our identities.
Narrow
There is an opportunity to shift young adults from being unable to engage with their digital memories to being able to meaningfully engage with an accessible, curated collection of them — by addressing not just how memories are organized, but how people are prepared emotionally to encounter them.
How might we help young adults bridge the gap between their digital and physical memories so they can engage with them in meaningful ways?
How might we help young adults better curate their physical and digital memories without forcing them to confront all of it at once?
How might we combine physical and digital memories to create a complete picture of young adults' lives — a record that holds both what they kept and what they felt about it?
Your prompt, from when you were still framing this as an overwhelm problem
"How well does this tension statement align with my project: 'In digital storage young adults cannot engage meaningfully with memories because there are too many files resulting in them feeling overwhelmed'?"
ChatGPT
"You're claiming a fairly direct causal chain: too many files → overwhelm → inability to engage meaningfully. But is 'too many files' actually the core problem, or is it a symptom? Notice how your framing leans heavily on quantity. But your project has been circling around identity and the 'extended self.' Quantity alone doesn't fully explain why something becomes part of identity."
What it opened
I was sticking to the more obvious direction of "reducing overwhelm" when there was a more interesting discussion on identity to be had. The project shifted from a file management problem to an emotional engagement problem — which is a completely different design space.
Jordan is a 23-year-old woman who started posting on social media at a young age. Every embarrassing phase her teenage self went through is still documented on her phone. She finds it difficult to look back at these photos — but she still wants to be able to look back and appreciate those memories. She's stuck between not wanting to delete so many memories and finding those memories too difficult to revisit.
Midterm reflection
The virtual museum idea didn't get negative feedback, but feedback suggested it was leaving behind some of the more compelling questions I had raised about identity. The direction sharpened toward emotional engagement over archival display.
I have always struggled to find a way to appreciate the many important memories I have held onto throughout my life — it's always felt useless to hold on to things that collect dust, but I don't want to get rid of things too important to my life. With this project I'd like to explore how we can hold on to fragments of our past identity without creating clutter. I want to find somewhere between throwing away everything that serves no purpose and holding on to everything that once held meaning.
Implement
The first making move. Generate to understand how memory reconstruction could become visible through spatial experience — not polish, but exploration of the in-between space.
What the making showed
Making revealed that memory reconstruction became understandable only when visitors could see multiple versions of the same memory side by side. Rather than presenting memories as fixed records, the early visuals focused on ambiguity, multiple interpretations, fragmented recollections, and emotional transformation.
Convergence
I needed to decide how speculative this project would be — whether this was a user-friendly product or something that encouraged genuine emotional labor. The early generation confirmed that I wanted users to interact with their memories in a unique way, not just a reorganized archive.
Generating these images helped lay the foundation for the intervention's interaction logic — what Jordan's state looks like before the ritual, and what shifts after moving through it.
Before
Jordan before the intervention: the memory surfaces unexpectedly, she closes it immediately. The dread is in the gap between seeing it and being able to stay with it.
After
Jordan after the reflection ritual: the same memory, revisited on her own terms. Not resolved — but approached. She can see how she has grown since the image was taken.
Three visual positions on the same project. Each one sharpened the question of what this companion looks like and what world it lives in.
What I Learned
Generating these images helped me feel more supported and engaged — and taught me that AI-generated images can capture a feeling surprisingly well when the prompt is precise about emotional register rather than physical description. The images that worked were the ones where the feeling was specified before the form.
From notification to ritual to reflection: the feedback loop that makes the intervention more than a reminder.
The schematic shows the feedback loop users experience when interacting with the product: a difficult memory arrives, the user schedules it for a reflection session rather than avoiding it, the companion guides a personalized meditation ritual shaped by that specific memory, and the user builds a practice of returning. The mechanism is the ritual itself — not the object, not the notification, but the repeated act of intentional engagement.
The visual world every image must belong to. Not a style reference: a set of rules that every generation is measured against.
Strongest image from moodboard
Strongest image from moodboard
The world is one of acceptance and reflection. The space is still, close, and private. Objects are quiet, warm, and bright. While the feelings are raw, the process is refined to help. We will never judge — instead we guide toward self-acceptance. This is not a device. It is a ritual. It begins only when the person chooses to begin, and everything that follows is paced by them, not by us. Distress arrives as input. Serenity is the destination. The object holds the space between — soft in the hand, responsive to what the body is already saying. It does not rush. It does not judge. It listens through touch and speaks through questions. What it asks is shaped by what the person has brought with them. Nothing is generic. Nothing is assumed. Always: translucent material, inner glow, warmth, stillness, closeness. Never: judgment, confusion, noise, spectacle.
This or That Exercise:
12 shots across four groups. What worked: the warm, intimate visual language immediately communicated reflection and personal memory. What didn't: anything too technological, too polished, or too clinical. The object should feel earned, not manufactured.
What changed
I shifted from emphasizing AI generation to emphasizing personal memories and emotional interpretation. This made the project feel more human and less technical. ChatGPT was far more capable here than the other tools — the images that worked were generated there, not in Midjourney.
Storytelling
Four beats. Who Jordan is, the moment a memory surfaces and catches her off guard, the intervention, and what she carries out of the practice.
Jordan, 23. She joined social media at a young age. Her phone holds years of documentation — every phase, every past identity, every version of herself she has grown out of but hasn't deleted. Randomly being reminded of her past can make her feel stressed, embarrassed, and insecure.
An old image pops up on her phone — a memory notification from eight years ago. She opens it, then immediately closes it. The feeling it triggers doesn't go away. She's stuck between not wanting to delete it and finding it too difficult to revisit on her own terms.
She sends the image to Identity Core to revisit during her next meditation session — on her schedule, not the algorithm's. The companion walks her through a ritual shaped by that specific memory and her current emotional state. It asks questions. It does not judge.
She continues to work through difficult memories on a regular basis, forming a monthly ritual to work toward personal growth. The memories are still there. But now she can see how far she has come from them — and that distance feels like something she made herself.
I wanted the viewer to leave questioning their own relationship to the memories they avoid — and to feel like there might be a different way to approach them than simply waiting for the dread to pass.
What the video taught me
A realistic and emotional register — moving slowly from personal avoidance to collective experience — communicated the project's stakes more clearly than any explanation of the mechanism. The story does the work that the diagram cannot.
Evaluate
AI served as a useful thinking partner in moments where I could only work independently — it helped me identify gaps and consider new directions. However, it paled in comparison to the value of speaking to other people about my project. Discussing the work with others and creating my own sketches and mockups was a far more rewarding process than prompting alone.
I now know how to design an experience that turns an invisible cognitive process into a tangible and emotionally engaging interaction. AI is more useful than I expected for assisting in the ideation process — but my frustrations with image generation remain unchanged.
Week 1: Before
I thought AI was mainly a tool for generating content and accelerating production. I had no model for what it meant to use it as a thinking partner — I expected it to give answers, not ask questions that made my own thinking harder to avoid.
Week 10: Now
I understand that memory is not valuable because it is accurate; it is valuable because of the meaning we continue to reconstruct from it. AI is most useful not when it generates for me, but when it names the gap I haven't been willing to look at directly.
Class Gallery
A breadcrumb trail across the process. Not polished. Real. Every image here is a decision point.
First attempt to visualize memory as personal experience rather than scientific concept.
Exploring how one event can be remembered in multiple ways — foundation for reconstruction thinking.
Introducing technology into the frame — comparing memories with reconstructed digital versions.
