Med Check
Med Check
Unreal Engine 5 • Solo project • itch.io • PC
Developed for the Safe In Our World Mental Health Game Dev Champions 2024, Med Check is a first-person psychological horror game inspired by The Backrooms and Slenderman: The Eight Pages. Set in a procedurally generated liminal space, players must collect their medication while avoiding enemies that symbolise internal mental health struggles.
Design Goals & Objectives
Explore mental health themes through symbolism and surreal horror.
Build a dynamic environment using procedural content generation (PCG).
Create a tense gameplay loop combining stealth, exploration, and survival.
Learn and apply Unreal's Behaviour Trees for enemy AI.
Demonstration of procedural generation, enemy behaviour and collectible systems
Development Breakdown
Core Gameplay
Players navigate a looping, disorienting environment in search of eight scattered pill bottles. The goal is simple: take your medication before the enemies get you.
Enemies & Symbolism
Each enemy represents a different mental health state:
Controlled – A puppet on invisible strings, audible by the scraping of limbs.
Fragile – A near-invisible glasslike entity that whispers as it closes in.
Darkness – A smoky wraith embodying emotional suppression and masking.
Manic – Unpredictable and fast, shrieking at intervals and sprinting erratically.
All enemies share patrol, chase, and line-of-sight mechanics with ambient warnings like flickering lights and spatial audio cues.
Procedural Level Generation
Implemented using Unreal's PCG Graph, rooms and props (floors, ceilings, walls) are dynamically generated each session. Spawn points for items and enemies are integrated directly into the PCG system.
Objective System
Inspired by Slenderman’s eight-page structure, Med Check tasks players with collecting 8 pill bottles:
Objectives spawn randomly at surface-level locations using Blueprint Interfaces.
Interaction is handled via Enhanced Input and triggers a collection widget + feedback loop.
Victory and failure states are tied to either full collection or contact with an enemy (via a hidden 1-health mechanic).
Enemy AI
Built using Behaviour Trees and Blackboards for modular logic:
Enemies patrol randomly, react to sight/sound stimuli, and chase to last known location.
Each enemy is a child of a master AI class with unique properties.
AI perception includes peripheral vision/hearing and line-of-sight checks, avoiding false positives.
What I Learned
Procedural level design using Unreal’s PCG system
Behaviour Tree setup and debugging
Clean Blueprint-based interaction systems using Interfaces
Linking gameplay symbolism with personal mental health narratives
Balancing fear and clarity through lighting and sound