01 · N-Back
Three metaphors failed.
Lifting cards, stacking objects and prettier visuals that still leaned on text instructions all failed.
Cognitive Games
Four clinical memory and attention tests turned into mobile games for people with mild cognitive impairment.
- Context
- Neurowyzr's science team needed cognitive therapy sessions people would stick with.
- Problem
- The existing assessment flows held attention for 3–4 minutes, therapy needs about 20, and the core tasks were clinically approved and couldn't change.
- My role
- Experience strategy, interaction design and gamification: turning each task into a game, with scoring and feedback, inside the existing backend and a React Native web app.
- Outcome
- Four games designed and prototyped, each wrapping an approved clinical task in its own story and world.
- Client / Org
- Neurowyzr, Singapore
- Timeline
- 4 games, multiple iteration cycles
- Year
- 2025
Product DesignGamificationInteraction Design
The problem
Therapy needs 20 minutes. The tasks held 3 to 4.
For cognitive therapy, engagement is part of the product. The science team needed around 20 minutes of sustained engagement; the existing clinical tasks held attention for 3 to 4.
- 3–4 minattention the clinical tasks held
- 20 minneeded for therapeutic benefit (the science team’s brief)
The constraints
Change the experience, not the science.
The task logic was clinically approved, the backend already existed, and the platform was a web app with no game engine. The design had to live inside all three.
- Approved logic
The task mechanism could not change.
- Existing backend
A new skin on the same engine.
- Web app
No heavy game engine, no sprites.
The change in experience
Wrap the science in a story.
Keep each scientific task intact and redesign everything around it. PVT became Mystic Jungle Quest, Flanker became Signal Switch, N-Back became Space Cards Memory, and Maze became Maze Walker.
Fix 01 · Too complex? Simplify.
N-Back: same memory task, a clearer turn.
Drag to compare the original clinical screen with the game.
Fix 01 · Too complex? Simplify.
Watch people play, then remove what confuses them.
Complex tasks needed less, not more. Informal tests with two extremes, someone who reads everything and someone who reads nothing, showed what to take away.
- 01N-Back
- 02N-Back
- 03Flanker
02 · N-Back
Sliding cards worked. The Match button went.
The separate Match button was redundant: the card itself was the target. Sliding cards made each new turn visible.
03 · Flanker
A character built from parts.
The Flanker character had to adapt across screen sizes without a game engine, so it was built from separate, configurable parts.
Fix 02 · Too simple? Add a story.
Give simple tasks a reason to keep going.
Simple tasks needed meaningful friction: a story, levels and a timer so fast players do not race through fixed rounds.
- 01PVT
- 02Maze
- 03Maze
- 04Maze
01 · PVT
Colours became treasures and traps.
Users stopped responding to abstract colours and started collecting treasure and dodging traps. Same response logic, real motivation.
02 · Maze
Every level, its own world.
The Maze mechanics were already validated, so the work was differentiation: each level gets its own world.
03 · Maze
Dozens of characters, one hero.
Character explorations, narrowed to one hero people could connect with.
04 · Maze
The final character, in every state.
The final character, its poses and the tile states it walks across.
The UI
Four games. One family.
Four worlds that share one structure.
Every screen
Learn, choose a level, play, finish.
Every game follows the same path: how to play, choose a level, play, and finish.
Animation
Motion tells you what just happened.
Animation earns its place by explaining consequences: a treasure collected, a new card arriving, a tile cracking.
The design system · shared
One set of components, reused in every game.
Buttons, the top bar, score and timer, level banners, type and feedback were built once and skinned per game, each with its own palette.
Mystic Jungle Quest
- Ember#F54919
- Canopy#659813
- Deep jungle#0E2F1D
Signal Switch
- Signal blue#1F72CB
- Sky#E1FAFC
- Asphalt#786E69
Space Cards Memory
- Nebula violet#695FCF
- Midnight#0B0A36
- Alert red#E04D5D
Maze Walker
- Royal purple#6C429B
- Sunflower#E9E055
- Marigold#FCB537
The design system · new per game
And the parts each game needed of its own.
Each game also needed parts of its own: safe and unsafe objects, a policeman and hand buttons, card states, and the maze grids and tiles.
Hide the complexity.Keep the science.
DBFS taught me systems. The games taught me story.
