An alternative where design is seen, not read. GestALT is an alternative, perception-driven learning platform for UI/UX education. It addresses a fundamental limitation in how design is commonly taught: beginners are often introduced to UI/UX through theory-heavy documentation and static explanations, even though design itself is inherently visual, experiential, and interactive.
GestALT replaces passive, theory-first learning with an interactive system where users learn by observing, manipulating, and experiencing the consequences of design decisions.
Traditional UI/UX learning relies heavily on:
- written theory
- design laws explained in isolation
- examples without interaction
This approach often overwhelms beginners and fails to reflect how design decisions are actually made in practice. GestALT exists as an alternative—one that treats perception as the primary learning medium and interaction as the primary teacher.
At its core, GestALT is a game-based learning system, but not in a superficial or gamified sense. The platform is built on the idea that meaningful design understanding emerges from working within constraints and responding to feedback, rather than memorizing principles. Users are presented with intentionally flawed or incomplete interfaces and asked to improve them. Instead of being taught rules upfront, users:
= make design decisions first
= observe the outcomes of those decisions
= reflect on why those outcomes occurred
Learning happens through cause and effect, not instruction alone.
The central technical component of GestALT is a perception simulation engine.Rather than evaluating designs subjectively, the system models how users are likely to perceive an interface based on measurable properties, including:
-
element size
-
visual contrast
-
spacing and alignment
-
placement and hierarchy
Each UI element is represented as structured data rather than static visuals. When a learner modifies an interface, the engine recalculates metrics such as:
-
attention flow
-
visual hierarchy strength
-
cognitive load
This transforms feedback from opinion-based critique into deterministic, explainable evaluation.
GestALT is designed as a stateful system, not a collection of isolated lessons. For each user, the platform maintains a learning profile that tracks:
design decisions made recurring mistakes time taken to respond reliance on hints or guidance
Using this data, the system adapts future challenges by:
- increasing difficulty in areas of demonstrated strength
- reinforcing concepts where users struggle
- adjusting feedback depth and verbosity
- This adaptive behavior reflects real-world system design concerns such as state management, personalization, and data-driven decision making.
From a software engineering perspective, GestALT emphasizes clear domain modeling and separation of concerns. Core domain concepts include:
- Users
- Challenges
- UI Components
- Perception Metrics
- Decisions
These entities are explicitly defined and connected, allowing the system to scale in complexity without becoming brittle. The frontend functions as a visualization and interaction layer, not as the source of business logic.
Explainability by Design Explainability is a first-class goal of the system. For every evaluation, GestALT exposes: which visual principles were violated ; which rules were triggered ; how specific changes affected perception metrics This mirrors real engineering systems where understanding why a system behaves a certain way is as important as the result itself.
Accessibility is treated as a core system constraint, not a checklist.
GestALT enforces rules related to:
- color contrast
- hit target size
- keyboard navigation paths In advanced challenges, learners must balance aesthetic choices against accessibility requirements, reflecting real-world product tradeoffs faced by designers and engineers.
GestALT is an alternative to theory-first design education. It treats perception as something that can be simulated, tested, and refined through interaction, rather than explained solely through documentation.
As a project, GestALT demonstrates:
- perception-driven UX thinking
- system-level software design
- rule-based evaluation
- adaptive learning logic
- explainable decision systems It is intended to function not only as a learning platform, but also as a serious software engineering portfolio project, showcasing depth beyond surface-level UI design.
GestALT treats visual design as a functional part of the learning system, not decoration. Typography and color choices are intentionally restrained to support perception-driven, practical learning while minimizing cognitive load.
Merienda is used for headings and key instructional moments to introduce a human, perceptual quality into the interface. Its calligraphic form slows reading slightly and encourages reflection, aligning with GestALT’s emphasis on learning through observation rather than memorization.
Work Sans is used for body text and UI elements due to its neutrality and high legibility. It recedes into the background, allowing users to focus on interaction and decision-making rather than typography itself.
JetBrains Mono is used sparingly for metrics and system feedback to signal precision, structure, and evaluation.
The interface uses a deep charcoal and soft off-white base to reduce eye strain and maintain visual calm during extended learning sessions.
Muted indigo is used for focus and active states, guiding attention without urgency, while warm sand tones add subtle contrast and maintain approachability.
Color is used primarily for feedback and state changes, not decoration, ensuring that visual emphasis always carries meaning.
Together, the typography and color system creates a calm, cognitively safe environment that encourages experimentation, iteration, and sustained focus—supporting GestALT’s goal of being an alternative to theory-first UI/UX education.