Learn languages organically. No more abstract phrases, learn the way that helps you grow.
People who learn through language learning apps find themselves using incorrect grammar while speaking. Most apps promote memorization over understanding. Skill Script uses experiential learning to make learning more task oriented.
10 months
6
Figma
After Effects
Skill Script LLC.
Most platforms follow an interaction pattern of teaching through object-word association. This method superficially helps learn about the object. It fails to anchor that association due to the lack of context and educate users about the components and parts that make up the object.
We tested out our hypothesis with participants and found that they found this problematic, but a more pressing issue emerged.
LLPs utilize memorization and revision as their foundation to reinforce learning, which is slower, in terms of recall than the way humans learn via conversation. The primary function of language is to communicate ideas and actions. Following this notion, understanding the pros and cons of rote learning and experiential learning provided us great insight.
Introduce a conversation based learning process to improve the quality of knowledge and recall in users?
We took a look at other platforms to see how they taught and found a typical pattern of Rote learning. Skill script's learning process focuses on experiential learning, mimicking the fundamental human behavior of gaining knowledge through direct communication.
%201.png)
%201.png)
%201.png)
%201.png)
Shifting the focus to completing a task allows us to use language as a medium of communication and not as an end goal. Users can enter different rooms which house tasks of their own. The way to complete the task is by utilizing the concepts(letters, words, phrases, and sentences) taught in the exercise.
3D models allow for more flexibility and coherence than static images. Using models enables component level learning and the creation of more intricate tasks for advanced users.
While clicks were impactful for short text inputs such as letters, we found that drags require more time and conscious effort from the users. We use drags for longer selections, which we found to be more effective post-completion.
Before we started designing, it was critical to understand who we were designing for and their goals and problems. We divided our research into surveys and contextual inquiries. Surveys helped us gain initial insight into users, provided quantitative evidence to prove our design direction, and helped inform our interview questions.
We conducted surveys to identify our design direction, gain quantitative data on the user needs. Through Pollfish, we recruited 250 participants who learn or have learned Spanish.
We observed the participants using two platforms: DuoLingo and LingoDeer. We broke down the contextual inquiry session into three phases: Survey-centric interview, Think-aloud session, and Final interview.
A multi-faceted approach worked best when learning. A combination of audio, visual, and textual information helped build a concrete relation of objects.
When asking for nuanced answers, for example, identifying lexical gender in Spanish, understanding the scenario primes the users cognitively for their response. This also helps better understand where they went wrong and reinforces the logic based on the context.
Using a minimalistic UI helped to focus on the task at hand, improving the quality of learning.
Through our findings, I was able to create iterations of wireframes. Testing with 5 participants showed promising results in terms of learnability, but the usability aspect of our design needed to be more understandable.
We took into account the user pain points, needs, and opportunities and brainstormed ideas that would bring the best value proposition to our users.
So, what did we need in our design?
After identifying the features required, we needed to bring them together in a cohesive manner. After rounds of iterations, we created a conclusive user flow that would allow users to go into the exercise, knowing what they would learn based on the task.
Through our findings, I was able to create iterations of wireframes. Testing with 5 participants showed promising results in terms of learnability, but the usability aspect of our design needed to be more understandable.
In the first iteration, participants did not find enough distinction between draggable and clickable elements. We remedied this problem by changing the color, shape, and target of the CTA. For draggable elements, we showed a hollow indented space where the CTA would fit, like a jigsaw puzzle, whereas for click events showed underlines.
Before
After
Before
After
I created a design system to create consistency in our design. It was essential to have clear communication between the users and the system due to distinct input mechanisms in place.
Bringing a structure and flow to the task helps users build a relation to the concepts they learn while also assisting in the narrative, from object discovery to implementation. Each phase of the exercise acts like a module that used to customize the task.
We created a template that allowed users to learn concepts in detail.
Skill Script simulates a user journey that begins with identifying a task by entering rooms and ends at completing it. The goal is to learn and then communicate your way to completion.
From the very get-go of things, users have control and, more importantly, context, of they are learning. We introduce a narrative to learning, which is imperative to experiential learning.
Bringing a structure and flow to the task helps users build a relation to the concepts they learn while also assisting in the narrative, from object discovery to implementation. Each phase of the exercise acts like a module that used to customize the task.
We created a template that allowed users to learn concepts in detail.
Skill Script simulates a user journey that begins with identifying a task by entering rooms and ends at completing it. The goal is to learn and then communicate your way to completion.
Using dynamic models allows us to bring further granularity to learning by studying objects on a component level, opening new avenues for tasks, and improving the quality of objectives.
Skill Script simulates a user journey that begins with identifying a task by entering rooms and ends at completing it. The goal is to learn and then communicate your way to completion.
We went back to our participants to test our prototype and found overall positive responses from all 5 participants. From a learnability perspective, participants appreciated Skill Script's experiential learning. We also received helpful critique and input for our future scope.
It took participants some time to figure out the differences between drags and clicks, but once they understood it, the problem did not resurface. Participants asked for a mini-tutorial for onboarding when the patterns came up for the first time.
Due to time and resource constraints, we made the conscious choice not to involve speech input into our platform for the MVP. Our primary focus was on experiential learning and involving speech recognition, and validation had a different set of challenges that we could not handle at the time.
Making our platform more accessible means that it should be available for a wide variety of users and devices. Expanding to mobile options will be the next step, which will bring more challenges and opportunities.
Using audio and speech to communicate more freely and practice pronunciation will be fundamental to speaking any language more fluently. To do so, discovering strategies that allow users a form of digital validation of their speech will be instrumental.
Making our platform more accessible means that it should be available for a wide variety of users and devices. Expanding to mobile options will be the next step, which will bring more challenges and opportunities.
Using audio and speech to communicate more freely and practice pronunciation will be fundamental to speaking any language more fluently. To do so, discovering strategies that allow users a form of digital validation of their speech will be instrumental.
Working on this project has pushed me to question and think about every single choice that I make. Whether it be from a perspective of usability, learnability, interaction, linguistics, and much more, it has allowed me to take in multiple perspectives and distill them into a cohesive thought. I intend to keep this going and improve upon my design process with this knowledge in mind.
Head on over to my other projects to check out more of my work!
