UX Design Case Study · Kingston University CI7801
Imagine As You Read — Designing an AI-Powered Immersive Reading Companion
Transforming passive, text-only reading into a dynamic visual experience — using Generative AI and AR/VR, with the reader always in control.
Lean UX
User Research
Generative AI
AR / VR
Mobile-First
Accessibility
🎯 The Core Problem
Reading is declining — not because stories matter less, but because text-only formats aren't meeting readers where they are cognitively and visually. Digital tools digitised text without transforming the reading experience. Readers who struggle to visualise scenes, sustain focus, or emotionally connect with narrative have no meaningful support beyond the same wall of text, now on a screen.
🧩 The Design Solution
- AI-generated visuals, on demand, under the reader's control. "Imagine As You Read" is a mobile-first reading companion that uses Generative AI to transform selected text into contextual visual scenes.
- It is user-triggered, non-intrusive, and designed to support imagination without ever replacing the act of reading itself.
Think
Research
- Lit review
- User interviews (n=8)
- Survey (n=66)
Make
Design
- Card sorting / IA
- Lo-Fi paper prototype
- Hi-Fi Figma prototype
Check
Test
- Usability testing
- Hypothesis validation
- Weighted scoring
💡 Crucial Design Decision
"The most important design decision was the one I didn't make — choosing not to make visual generation automatic, resisting the temptation to showcase AI capability. Research consistently said: give users control."
📈 Research Metrics
50%
Prefer optional visuals
🎯 Methodology
- Semi-Structured Interviews: Moved from surface reading habits to the precise moment a reader loses focus or hits cognitive friction.
- Quantitative Survey: Captured reading frequency, device usage, and openness to AI/AR features. Validated patterns at scale.
- Competitor Analysis: Mapped Wonderscope, Google Learn Your Way, Copilot, and Gemini against 8 dimensions.
🧩 Key Qualitative Findings
The Overwhelmed Student
Students hit a cognitive wall with text-heavy content — not from lack of ability, but from lack of visual scaffolding.
The Disconnected Reader
Young adults lose their reading thread mid-chapter because digital environments fracture attention. The story isn't the problem — the context is.
Control Over Automation
Users are cautious about AI-assisted reading: "I don't want something interrupting my focus." User-triggered always outperforms automated.
💡 Competitor Insight
No competitor integrated real-time text interpretation, generative visualisation, and reading-focused UX in a single product. This was the gap to fill.
"I don't want reading to feel harder — I just want a bit of help imagining things."
Goals
- Understand complex material quickly
- Stay focused longer
Pain Points
- Struggles to visualise abstract concepts
- Resents visuals that interrupt flow
"I love stories, but digital reading just doesn't pull me in enough."
Goals
- Re-establish reading habits
- Visualise scenes vividly
Pain Points
- Digital text feels flat
- Loses interest in long-form mid-way
"I understand things better when I can see them, not just read about them."
Goals
- Feel confident tackling academics
- Tools that reduce cognitive stress
Pain Points
- Dense content is difficult to retain
- Boring apps actively discourage use
🎯 Progressive Disclosure
Users should always feel in control of how deep they go. We separated the experience into clearly defined modes. Survey data showed 50% wanted "text with optional visuals." Burying visual features inside reading mode would have violated expectations.
🧩 MoSCoW Prioritisation
- Must Have: User-triggered visuals, Disable at any time, Side-panel display, Clean interface.
- Should Have: Intensity controls, Diagram visuals, Regenerate/dismiss.
- Won't Have: Always-on AR/VR, Auto visuals without consent, Gamification.
💡 Key Hypotheses
H1We believe increased engagement will be achieved if students attain improved understanding with optional AI visuals.
H2We believe longer session time will be achieved with user-controlled text-to-visual generation.
H5We believe increased user trust will be achieved with visual intensity controls and dismiss/regenerate options.
Stage 1
Wireframes
Concept sketches
Stage 2
Lo-Fi Paper
Interaction testing
Stage 3
Mid-Fi Digital
Flow validation
Stage 4
Hi-Fi Figma
Full simulation
✏️ Lo-Fi to Hi-Fi Journey
Paper prototypes created psychological safety — participants gave honest feedback because it was clearly unfinished. Grayscale digital layouts validated navigation flows. The final Figma prototype simulated full AI generation flows.
| Participant | Task | Time | Errors | Assistance | Result |
|---|
| P1 | Enable Visual Mode | 32s | 0 | 0 | ✓ Pass |
| P2 | Enable Reading Mode | 45s | 1 | 0 | ✓ Pass |
| P3 | Scan Text Feature | 40s | 1 | 1 | ✓ Pass |
| P4 | Immersive Mode Testing | 55s | 2 | 1 | ✗ Fail |
🎭 Emotional Friction Points
Relief
Participants expressed unprompted relief when visual generation was confirmed as optional.
Control
The "dismiss visual" control was used enthusiastically across all participants. Users needed the override.
Before — The Problem
Immersive Mode was accessible via the mode switcher with no contextual prompt. P4's task failure revealed users didn't encounter it naturally.
After — The Fix
Immersive Mode entry was repositioned as a progressive enhancement — surfaced after a user generated at least one visual, with a gentle prompt.
Before — The Problem
The initial three-state toggle (Off / Low / Medium) felt binary. Participants wanted to communicate how much visual support they needed.
After — The Fix
Replaced with a continuous slider with labelled reference points (Off → Subtle → Moderate → Rich), supplemented by a preview thumbnail.
🏆 Validated Hypotheses
H1 — Validated ✓
Optional AI visuals improve perceived engagement without raising cognitive load.
H5 — Validated ✓
User control over visual intensity and dismissal is essential for trust.
H9 — Validated ✓
A text-first architecture supports diverse contexts smoothly.
💡 Closing Reflection
"Imagine As You Read is not a reading app that uses AI. It is a reading experience that uses AI only when the reader asks it to — and that distinction is everything."
GJ
Ganesh Javhare
UX Researcher & Designer · CI7801 Major Project
