Empowering Children to Tell Stories Through Interactive Design​
​
Storytelling provides a variety of benefits to children of primary school age, and extant storytelling tools and voice AI apps for storytelling are either limited by linear storytelling methods or do not engage with players enough to inspire replayability.
A CHILD'S TALE
PROJECT OVERVIEW
TIMELINE
August 2021 - May 2022
​
PROJECT TYPE
Graduate Capstone Project - Voice AI UX
​
TEAM
UX Research / Product Design - Adam Wolf
​
KEY RESPONSIBILITIES
User interviews, secondary research, contextual observation and inquiry, expert interview, data analysis, journey mapping, rapid sketching, user testing, iteration, voice AI research, high fidelity designs, prototyping, presentation of findings and design process​
​
Long Form Documentation (PDF)
THE PROBLEM
​Storytelling is extremely valuable as an educational developmental tool for children. Voice AI is the perfect medium for encouraging collaborative storytelling among children, but it is underutilized for that purpose.
Most voice AI storytelling applications are extremely linear and require little generative imagination, and the medium as it currently exists tends to passively discourage physical activity due to the fixed nature of most voice AI devices. Additionally, voice AI storytelling tools are usually aimed at a single user, failing to reap the many benefits of collaborative storytelling.​
OVERVIEW: GOALS & CONSTRAINTS
​​
​
My goals for this project were based on research showing that engaging, collaborative storytelling teaches kids important communication skills. ​
​
The constraints vary from child to child, based on age and other factors, but they apply to most children in the target age group (5-10 years).
DESIGN SPACE: STORYTELLING
A standard story arc
I wanted to empower and encourage children and parents to interact through storytelling for:
-
Educational improvement - storytelling advances literary skills, verbal fluency,
expressiveness, fosters creativity, exercises imagination, and builds self-confidence -
Socialization - storytelling can create a social interaction between two or more people, creating bonds through interactions and shared experiences.
The design should:
-
Present a novel, engaging way to facilitate storytelling between a child of ages 5-10 and one or more peers, guardians, or artifacts.
-
Always encourage, never criticize - make the experience fun and repeatable across the target age group but also beyond, so parents also enjoy the experience.
-
Be easy to understand and low on constraints - kids like to think outside the box, so the design should encourage acting creatively (without being too abstract).
-
Be humorous, or allow for humor - serious storytelling is fine, but funny storytelling
games tend to more strongly engage children and encourage repeatability.
Stakeholders:
-
Children between the ages of 5 and 10
-
Their guardians
.jpg)
Basic storytelling journey map
Basic storytelling journey map
Basic storytelling journey map
Basic storytelling journey map
RESEARCH INSIGHTS
Secondary research in the form of an extensive literature review revealed a number of benefits to children from telling stories, especially when done in collaboration with others:
-
Kids use stories to make sense of their world and test their conceptual understanding
-
Collaboratively telling stories helps kids to learn and understand the proper context of words and phrases
-
Storytelling also teaches them better communication skills and even improves their literacy
-
Collaborative storytelling also improves children's abilities of recall and recognition through the use of repeated themes
-
It helps them to better socialize by practicing improvisation and public speaking with peers​​
Rory's Story Cubes Contextual Inquiry - P1
Primary research involving testing and observing extant storytelling tools in use by children within the target age group, both individually and collaboratively, revealed some pain points:​
-
Voice AI games such as The Magic Door were engaging but extremely linear, not promoting generative imagination and storytelling but following along with a story and making limited choices
-
Physical games or tools like Rory's Story Cubes were slightly more physical in nature, but required a level of abstract thought that was difficult for younger children
-
In all observed cases, extant tools engaged user attention for a relatively short time, no longer than 10 minutes and usually around 6 or 7 minutes total, with kids losing interest or feeling too constrained
-
The sedentary nature of the tools under review were a major pain point with the participants ages ranging from 6-8 years and having high levels of energy
-
Rory's Story Cubes was the most generative tool in these observations, but the game requires kids to make stories based on iconography and was found to be fairly constrictive of the child's imagination
FORM FACTOR
During the ideation process I considered various potential form factors for the design:
Ultimately I determined that voice AI technology, such as Amazon's Alexa, would be the ideal medium for a collaborative storytelling app. Voice AI has a variety of benefits - including the potential for mobility, ubiquitous availability on most smart devices, and being possessed of enough intelligence to act as a situated actor in natural conversational modes of communication with multiple players.
CONCEPTUALIZATION AND ITERATION
Voice AI integration for recording
Initial Concept
My first concept sought to encourage group storytelling through the use of a tabletop game that incorporated voice AI and a timer with buttons for each player.
​
.jpg)
Timer to elicit excitement
Collaborative physical design
Problems
My first concept had some of the features that would end up in my final iteration, but it also contained a number of issues that made it less than ideal:
-
The timer would likely create a more competitive atmosphere - not good for collaborative storytelling
-
The concept relied on kids to tell a complete story without much help, which is expecting a lot
-
Part of this design was encouraging kids to stop telling the story and create art or other forms of creative expression, but this would also disengage them from the game
-
The nature of any tabletop gaming device is sedentary, and one of my prime goals was to incorporate activity
​
Concept Insights
After my first concept I reconsidered the objectives of my design and the insights I gained through expert interview and user testing. I revised my objectives and the features which should help to meet them:
​
​
Iteration
I iterated upon my first concepts in various ways after conducting further research into the current state of voice AI technologies. I dropped the timer and incorporated the ability to create a digital memory of the story in the form of short videos, photos, and audio clips. One of the most important iterations to my design was the addition of individual Bluetooth wearable devices in place of the single cumbersome desktop device, both to afford movement on behalf of the players and to record multimedia clips of the storytelling.
​
These devices would be worn by each player and used to facilitate both the collaborative telling of stories and the digital capture of moments in time. Each device includes a Bluetooth transmitter, speaker, camera, and microphone, and can relay voice communication between the players and the central voice AI device (such as an Amazon Echo).
They would also record video clips and other media and upload the snippets to the web, where they would be combined into an accessible memory of the story game. This serves not only to provide kids with an entertaining montage of themselves and their friends telling a story together, but it reinforces aspects of the story that was told and also encourages kids to play the game and make another fun video. ​
​
User testing revealed that some form factors of the wearables might be more popular (specifically the headband and bracelet or watch) and others less so (especially the belt), and furthermore that it would be impractical to have a camera situated at waist height or embedded into a ring. As such, I iterated on the wearables, though these might still be narrowed down to only one or two of the best form factors after further testing.
​
FINAL DESIGN
My concept will give generative creative control to the players and inspire an active, engaging experience.
The AI sets the scene and steers the overall story arc, but the control of the individual story elements is left entirely up to the players. When the players are fully through the story arc, Alexa prompts them to watch the replay on a screened device or grants them the option to start a new story.
ADDITIONAL HIGH FIDELITY INTERFACE SCREENS
The "Replay" aspect is a design choice that I believe will incentivize children to play the game and come back for more. This idea gives them not only the ability to see themselves on the screen, which kids love, but to have agency in the creation of that media. Storytelling might be engaging while it lasts, but having that multimedia memory of the game, which a child could come back to watch over and over again, is really the concept in this iteration that would most motivate replayability.
SYSTEM ARCHITECTURE
To better understand how the final design would function in a real-world environment, I created this high-level system diagram showing the flow of input, processing, and output through the AI storytelling assistant.
Below you can see the final system diagram of the prototype, with all of its component pieces. The Alexa device outputs story elements to the players, prompts them to tell a segment of the story when it's their turn, and resolves any player issues that may crop up during play. Alexa takes in player stories, questions or problems, and any media that the players capture with their wearable devices. The multimedia is uploaded and converted into a montage of clips in the order they were recorded. Players can access that media on a screened device.
PROTOTYPE VOICE AI DIALOGUE TREE
This prototype is a functional set of dialogue choices that will trigger additional dialogue options as the players move through the story. The section seen here is only a beginning, the full implemented script would be many hundreds of potential branching choices.
STORY MAP
This story map is likewise not a complete set of potential directions that players could go in the course of a session, but it does show the primary possible route through the game complete with choices the players might make, problems the players might encounter, and actions taken by the voice AI to steer the course of gameplay towards its conclusion.
LIMITATIONS
A variety of potential limitations and ethical considerations became apparent in the design of this product. These are areas that would benefit from further research before development:
DIGITAL DIVIDE / COST
WEARABLE DEVICE LIMITATIONS
VOICE AI LIMITATIONS
The cost of the wearables could be prohibitively expensive for many families. The device technology is relatively complex, incorporating camera, microphone, Bluetooth transceiver, and haptic feedback, and these are susceptible to limitations of distance, audio interference, poor camera angles, etc. Finally, voice AI is not yet at the point where it can be used for realistic contextual dialogue. As such, the AI would currently be limited to a set of expected responses and dialogue options.
