Mnemo Magic App
At-a-Glance
Mnemo Magic is a UX design project exploring how generative AI can reduce the effort required to create and study mnemonic devices. Mnemonic devices have been widely used as a way to easily recall abstract or dense information. Creating effective memory aids requires users to translate information into something personally meaningful. This process can be difficult and time-consuming when studying under pressure.
I designed Mnemo Magic as an AI-powered study tool that generates personalized mnemonic devices while giving users control over how they customize, organize, and practice them. I incorporated cognitive psychology principles such as chunking, retrieval cues, and associative learning to support meaningful recall.
The project included primary and secondary research, competitive analysis, iterative design, usability testing, and an interactive high-fidelity prototype.
Timeline
4 months
Problem
Learners often struggle to retain abstract information, especially under time pressure. Mnemonic devices are a proven memory aid, however, users struggle under cognitive load to find the time to create their own. Existing mnemonic device generators either lack personalization or output random content with no structure. As a result, learners lack an efficient study tool that generates meaningful mnemonic devices that can be easily recalled and studied with repetition.
Solution
Mnemo Magic uses generative AI to create personalized mnemonic devices based on the information users want to remember.
Rather than simply generating a mnemonic and ending the interaction, the experience allows users to:
-
Generate and customize mnemonic devices
-
Save memorable results to favorites
-
Organize content into study decks
-
Review information through flashcards and quizzes
-
Reinforce learning through study games
The goal was to reduce the cognitive effort involved in creating memory aids while supporting the repeated recall and practice necessary for learning.
Research
I used a combination of primary and secondary research to understand both the user problem and the cognitive principles behind mnemonic learning.
Primary Research
User Survey
I conducted an online survey to understand how people currently use mnemonic devices, their confidence in creating them, and what they would want from a tool designed to assist with mnemonic creation.
Secondary Research
Cognitive Psychology
I reviewed research surrounding memory retention and mnemonic learning, focusing on principles including:
-
Chunking: Breaking information into smaller, manageable pieces.
-
Retrieval Cues: Creating triggers that help users recall stored information.
-
Associative Learning: Connecting new information with existing knowledge or concepts.
These principles helped inform how the AI-generated mnemonic experience could support meaningful recall rather than simply generating arbitrary words.
Survey Insights
The survey revealed several important opportunities.
Users lacked confidence creating mnemonics themselves
Participants indicated that creating mnemonic devices was not always intuitive. Even when users understood the concept, they could struggle to develop something memorable on their own. This reinforced the opportunity for AI to reduce the effort involved in creating the initial memory aid.
Personalization mattered
Users expressed a strong preference for being able to personalize and customize their mnemonic devices. This suggested that a successful AI tool should not simply generate a single answer and expect users to accept it. Users should have the ability to influence the output and make it more personally meaningful.
The tool needed to support more than generation
The research also informed the broader product experience. If users were going to generate memory aids, they would need an efficient way to save, organize, revisit, and practice them. This led to the inclusion of study decks, favorites, quizzes, flashcards, and study games as part of the broader experience.
Competitive Analysis
I compared Mnemo Magic against existing tools to understand where an AI-powered mnemonic experience could provide additional value.
Quizlet
Quizlet provides effective flashcard-based studying but does not focus on generating personalized mnemonic devices.
ChatGPT
ChatGPT can generate mnemonic devices when prompted, but mnemonic generation is not its primary purpose. It also lacks a dedicated structure for organizing, refining, and studying generated mnemonics.
MnemonicGenerator.com
MnemonicGenerator.com provides mnemonic generation through a web-based experience, but offers limited customization. Generated outputs could also feel disconnected from the information the user was trying to remember and did not make sense logically. (See photo below)
Opportunity
The competitive analysis revealed an opportunity to combine AI generation + personalization + organization + studying within a single focused experience.
When asked to generate a mnemonic for the order of the planets, the tool used letters from my prompt rather than the information I wanted to remember. The result was a collection of random words that lacked logical meaning or a memorable connection to the planets.
Persona
To translate the research findings into design requirements, I developed a persona representing the needs, behaviors, frustrations, and goals identified through research.
User Journey Map
Paper First
Process begins with my initial idea wireframes and phases into mockups, and finally the high fidelity prototype
Design System & Rationale
Cognitive physiologists have suggested findings in several articles that point toward colors having an impact on memory retention and recall. The color palette used for this project includes shade ranges of blues, greens, and yellows. A survey conducted showed that users often associate the color green with a calming emotion, which may in turn reduce mental fatigue. Yellow has been shown to produce a greater impact on memory and grasping the attention of users. Blue has been studied to find that the color improved sustained attention and creative performance.
Paper to Digital
Then, I brought my idea to an interactive digital prototype to bring to ten potential users to gather feedback.
User Testing & Design Iterations
I translated the research findings into an initial prototype and evaluated the experience through usability testing.
Participants completed tasks involving mnemonic generation, saving content, navigating study materials, and interacting with favorited items. I used a combination of task success data, click/heat maps, and think-aloud feedback to identify areas where the interface created friction or where the interaction model did not match users' expectations.
Design Change 1: Usability Testing provided valuable insights and highlighted opportunities for design refinements, specifically in regard to labeling and icon usage. Users experienced difficulty associating the oblong star icon with the concept of “favorites,” with several participants indicating a preference for a more conventional symbol, such as a traditional star or heart.
Design Change 2: When tasked with how they would share or delete a favorited item, users appeared uncertain about how to access these actions. Heat map data showed that participants frequently tapped directly on the card or icon rather than the overflow menu, suggesting the interaction pattern was not clearly communicated. In response to these findings, the icon was revised to include arrows rather than the previous dots to more clearly signal a swipe action.
Awards
iSchool 2026 Best Student Paper
Before
Before
My Role
UX Designer | Innovator
After
After
Platform
Figma
Prototype
What I Learned
Through this rewarding project, I gained a better understanding of how cognitive psychology principles can directly inform UX design. Background research into cognitive psychology, memory retention, and color theory allowed me to better inform my design decisions. I also learned the importance of clear iconography and reducing ambiguity to reduce friction.
Usability testing was insightful in revealing both quantitative and qualitative data. Heat and click maps were able to capture success and failure rates per task, while users thinking aloud allowed me to understand their navigation process and where they would expect actions to be placed.
From this experience, I also had the opportunity to navigate a situation where users were not receptive to certain design choices. As the designer and strategist, it was a challenge when my initial design decisions did not resonate with users. This allowed me to set aside my personal assumptions and mental model to align with the user needs and goals. Ultimately, if a design choice is a distraction or hindrance to the user completing their tasks and goals, it must be reconsidered. I was able to gather user feedback and make informed adjustments to support task completion.
