Canary hero image, showcasing a phone mockup with a Canary illustration before a vibrant purple backdrop.

Canary (In Progress)

A pink and purple floral divider

About

For Spring 2024 at Cornell Tech, the theme of our INFO 5305 (User Experience and User Research) class that semester was ‘neurodivergence.’ My homework partner and I chose to focus our semester-long class project on sensory overload, which is the overstimulation of one or more of the senses. Sensory overload most commonly affects autistic people and people with ADHD, and can markedly cause a great deal of distress to the individual. As such, we sought to create a product that could help alleviate the overwhelming anxiety and discomfort generated by sensory overload.

Timeline

February 2024 - May 2024

Role

UI/UX Designer, UX Researcher

Tools

Figma, Google Suite, Zoom

Team

Sharon Lin, Nandini Proothi

A pink and purple floral divider

Skills

Collaboration
Interviewing
Qualitative Coding
Affinity Mapping
Medium-Fidelity Prototyping
Persona Design
Storyboarding
High-Fidelity Prototyping

Process

1. Create product concept
2. Interview individuals who experience sensory overstimulation
3. Begin qualitative coding using interview transcripts
4. Consolidate findings into Affinity Diagram
5. Uncover themes from user research
6. Devise a list of requirements for the product
7. Create personas and storyboards of envisioned usages for the product
8. Develop a medium-fidelity prototype
9. Receive heuristic evaluation on medium-fidelity prototype
10. Incorporate feedback to develop high-fidelity prototype
11. Create a rigorous evaluation protocol

Product Concept

Canary is an IoT product that combines a traditional phone application with a tactile object to assist neurodivergent users when they’re experiencing sensory overload. Some neurodivergent individuals may experience overstimulation when encountering loud noises, pungent smells, bright lights, or other environmental factors. This overstimulation can lead to intense anxiety and significant distress. Canary is designed to walk users through calming, meditative exercises in conjunction with the tactile elements presented by the accompanying Canary Ball, which includes various materials and textures to soothe the user. The app also offers avenues for users to distract themselves using mini-games and visual stimulation. The app's intuitive interface allows easy navigation, especially during distress. Customizable sensory profiles allow users to tailor their sensory preferences based on specific scenarios.

User Interviews

To understand neurodivergent individuals’ experiences with sensory overload — as well as their current self-soothing strategies when encountering it — we conducted a series of semi-structured interviews with 6 participants who identified as autistic or ADHD. After standard demographics questions, we asked participants about their experiences with sensory overload, their experiences with stimming (the repetition of physical actions, sounds, or other behaviors to help cope with emotions), whether they sought external help during overstimulation, and the technological interventions they used to self-soothe.

Qualitative Coding

Using the transcripts of our interviews, we began to qualitatively code our interviews in Google Sheets. This allowed us to quickly extract the core findings from our research. 

Affinity Diagrams

Gathering all our qualitative codes, we placed them all in a Miro board and began separating them into general themes. These themes included: 

• Neurodivergent experiences
• Technological interventions
• Audio sensory overload
• Visual sensory overload
• Olfactory sensory overload
• Tactile sensory overload
• Experience of sensory overload in a private setting
• Experience of sensory overload in a public setting
• Stimming
• Calming down while overstimulated
• Distracting while overstimulated
An affinity map of our interview findings, organized by multiple colors.
Our Affinity Diagram which consolidates all our qualitative codes into 11 distinct themes

Summary of Findings

The qualitative analysis helped us recognize the pain points participants faced when experiencing sensory overload. Interestingly, not all participants resorted to stimming in their past experiences of sensory overload. While some of them found themselves stimming unconsciously, others stated that it did more harm than good. When experiencing sensory overload in different settings, some participants were keen on calming themselves down, but others chose to distract themselves. Our research told us that sensory overload was triggered by different environmental factors in public versus private settings. Additionally, participants’ methods of coping with overstimulation differed based on these setting differences due to fear of public perception and stigmatization. 

Design Requirements

Given the emerging themes of our user interviews, we came up with 8 design requirements for our product. 

Design Requirement #1
Interaction design requirement
Light adjustment and/or using more muted colors
Name of high-level category
Visual Sensory Overload
Requirement statement
Bright lights trigger sensory overload
Rationale for requirement
Interview participants who experienced visual sensory overload often had it triggered by bright lights. As such, we need to ensure that our product doesn’t cause or exacerbate visual sensory overload through bright lights. Given that each individual user’s threshold for defining a “bright light” might be different, there should be a simple-to-access interface for adjusting the brightness settings.
Design Requirement #2
Interaction design requirement
As few clicks as possible to get into an app feature
Name of high-level category
Technological Interventions
Requirement statement
UI for meditation apps requires too many clicks to get started
Rationale for requirement
Since our users will likely be in the midst of experiencing sensory overload, they will be in a state of heightened anxiety. As such, we want to alleviate as much friction as possible when using our app, so that they can be relieved of their overstimulation as quickly as possible.
Design Requirement #3
Interaction design requirement
IoT device should look discreet, non-medical, and not look like it was specifically designed as a device for neurodivergence
Name of high-level category
Experience of sensory overload in a public setting
Requirement statement
Perceived stigma from surrounding people regarding stimming
Rationale for requirement
Several of our interview participants were too anxious to stim in public due to fear of judgment. Therefore, our IoT device shouldn't look medical or like a stereotypical stimming device.
Design Requirement #4
Interaction design requirement
Playlist feature
Name of high-level category
Calming Down while Overstimulated
Requirement statement
Music that is familiar to calm down during sensory overload
Rationale for requirement
Some users feel comfortable listening to familiar music when they are experiencing sensory overload. We will connect Canary to popular streaming platforms such as Spotify or Apple Music using API, so users can easily access their favorite and most calming playlists.
Design Requirement #5
Interaction design requirement
An emotion wheel
Name of high-level category
Neurodivergent Experiences
Requirement statement
Non-verbal communication is preferred but writing things out is also okay while experiencing sensory overload
Rationale for requirement
When users are overstimulated and don’t want to resort to verbal communication, they can use an emotion wheel to select their mood. To make the idea less clinical and more fun, we will make it animal-themed.
Design Requirement #6
Interaction design requirement
Soft materials on IoT device
Name of high-level category
Stimming
Requirement statement
Soft textures help calm down
Rationale for requirement
Based on our user interviews, most participants found soft materials and textures to be calming. By incorporating these textures into our IoT device, we hope to address the issue of overstimulation caused by hard materials and textures.
Design Requirement #7
Interaction design requirement
Mini-games involving repetition
Name of high-level category
Technological Interventions
Requirement statement
Enjoys repetitive tasks facilitated by technological interventions
Rationale for requirement
Based on our interviews, users prefer repetitive and mundane tasks to calm themselves down while experiencing sensory overload. They resort to games and other avenues where there won’t be too many surprises or unexpected events happening. To incorporate this, we want to create mini-games in the app that are similar to Piano Tiles and Stardew Valley.
Design Requirement #8
Interaction design requirement
Customizable parameters in settings
Name of high-level category
Technological Interventions
Requirement statement
Have different options for the same feature to better accommodate neurodivergent users
Rationale for requirement
Users prefer to have the option to change certain parameters, including the speed of the game. By incorporating robust customization features, the user can tune the application to their preference by changing font size, colors used in the app, etc.

Personas

Our interview participants naturally ended up separating themselves into two different categories: those who chose to calm themselves when overstimulated, and those who chose to distract themselves when overstimulated. Those who chose to calm themselves would often wear noise canceling headphones, read articles, or listen to long form videos. Those who chose to distract themselves would often play games on their phones or fidget with an easily accessible object. Our aim with this project was to help as many users as possible when they experienced sensory overload. As such, our product had to offer different features catered to those who chose to distract themselves or calm themselves when overstimulated. 

A persona of a user who likes to calm himself down when overstimulated.
Persona of an individual who prefers to calm himself down when overstimulated.
A persona of a user who likes to distract herself when overstimulated.
Persona of an individual who prefers to distract herself when overstimulated.

Storyboards

To understand how and why individuals would use our products, we created three storyboards to illustrate ecological, interactive, and emotional design perspectives for our product. 

An emotional storyboard showcasing a user using the emotion wheel to explain to a friend how they're feeling, and then stimming by playing games on the app.An interactive storyboard depicting a user stimming through the IoT Canary Ball, music features, and game features of Canary.An ecological storyboard depicting a user stimming with Canary while in public transit.
Storyboards showcasing the ecological, interactive, and emotional design perspectives for our product.

Medium-Fidelity Prototype

We promptly set to working on a medium-fidelity prototype. This was a “T” prototype, meaning that we showed a breadth of features for our product in a superficial way, then did a deeper dive into functionality for some key features.  

Our medium-fidelity prototype showcasing Canary's core features: games, videos, music, emotion wheel, meditation, and Canary Ball.
All Canary medium-fidelity screens laid out on a Figma canvas, including the home page, games, videos, music, emotion wheel, meditation, and Canary Ball.
All the pages featured in our medium-fidelity prototype.

Heuristic Evaluation

Having completed our medium fidelity prototype, we exchanged our work with another group’s medium-fidelity prototype to perform a heuristic evaluation on each other’s work. The numbers next to the violation indicate the level of severity for the violation, with 1 being low severity and 5 being very severe.

Their heuristic evaluation of our application, as well as our proposed fixes for the violations, were as such:

Violation
Reason of Violation
Proposed Fix
Match between system and real world (1)
The Canary Ball page has a relatively intuitive interface for adjusting the haptics speed and intensity, but the feature itself might be unfamiliar and confusing to a first-time user.
We would add documentation on how the Canary Ball is used. We might also change the word “Haptics” to “Vibration.”
Error Prevention (2)
In the Music Settings page, there could be problems when linking a music streaming account. What if a user doesn't have access to a music streaming service? 
It will be made clear that the Music feature is completely optional. Users are not required to have access to both Spotify and Apple Music to use the feature. They can disconnect from either feature at their own discretion, and do not have to use the feature at all if they don’t have music streaming services or do not wish to connect their accounts to our app.
Help and Documentation (2)
The Home Screen could benefit from some documentation to help users understand how to use the Canary app.
We will add an onboarding tutorial for new users to get to know the features of our app.
User Control and Freedom (2)
Users should be able to access other features (such as Meditation) from any page (such as Music) without having to return to the Home Screen.
We will add a bottom navbar so users can quickly switch between features without having to go back to the Home Screen.
Recognition Rather than Recall (1)
In the Videos feature, it is hard to capture the overall feeling of the video with just the title. It might be better to add a brief description, a few keywords with hashtags, or intuitive emojis to capture the atmosphere of the video.
We will add some hashtags of the main content featured in the videos so that users have a better idea of what to expect in these videos. This will also lessen any fear of uncertainty for users who may be affected by that when overstimulated.
User Control and Freedom (3)
The Videos feature could benefit from being divided into categories (such as "Soothing"), similar to the Music page. All pages could benefit from Search Bar functionality.
We will add search functionality for videos, games, and music. Additionally, we will incorporate category tabs so users can seek videos and music based on their individual preferences.
Error prevention (2)
For Music, it is unclear what would happen if a user were to accidentally connect or disconnect a streaming platform.
We will add an error popup if something is improperly connected to our app.
Help and Documentation (1)
The title "Emotion Wheel" may not be immediately intuitive to users on how to interact with the feature. There is also no help icon, tooltip, or documentation within the feature. Users may not understand what to expect upon selecting an emotion.
We will add a brief explanation on how to use the Emotion Wheel feature on the page.
Help and Documentation (1)
The Meditation page lacks a help icon or tooltip explanation for the different meditation categories. Users new to meditation may be confused and unsure which type of meditation would best suit their current needs.
We will be adding an explanation of each type of meditation — including a detailed overview of what different types of activities entail (time, effort, outcome, etc).
Help and Documentation (3)
User unfamiliar with the Canary Ball feature may need additional information on what "Haptics Intensity" or "Haptics Speed" mean, as well as how such settings would affect their user experience.
Will will provide documentation on the Canary Ball page about how such features would impact users' IoT device. We will also be adding a walkthrough so first-time users can test different haptics settings before saving their preferences.
Flexibility and Efficiency of Use (2)
The Meditation page should allow repeat users to mark favorites or create shortcuts to preferred meditation exercises. Otherwise, the page is inefficient, despite being designed for quick stress relief.
We will add a button that allows the users to favorite certain meditation episodes. By displaying these at the top, they will be easily accessible.

Rigorous Evaluation Protocol

To test the efficacy of our product, we developed a rigorous evaluation protocol to test on users.

Product Description
Canary is an IoT product that combines a traditional phone application with a tactile object to assist neurodivergent users when they’re experiencing sensory overload. Some neurodivergent individuals may experience overstimulation when encountering loud noises, pungent smells, bright lights, or other environmental factors. This overstimulation can lead to intense anxiety and significant distress. Canary is designed to walk users through calming, meditative exercises in conjunction with the tactile elements presented by the accompanying Canary Ball, which includes various materials and textures to soothe the user. The app also offers avenues for users to distract themselves using mini-games and visual stimulation. The app's intuitive interface allows easy navigation, especially during distress. Customizable sensory profiles allow users to tailor their sensory preferences based on specific scenarios.
Goals of the Evaluation
Metrics
• Task completion: Are the users able to navigate through the application without the need for additional information?
• Speed: This will only be measured when they are navigating to a certain feature i.e., how long did it take them to go to meditation mode or games mode?
• Emotional reaction: Depending on the ease of the use of the app, and the activities involved in the app, how does the user feel after using certain features? (Example: “It took too long to open connect to the music service and it made me feel frustrated.” This will be self-reported by the user.
Protocol
Overview
This will be a one-on-one session we hold with each participant to conduct a UX evaluation of Canary. The goal of this session is to examine the efficiency, task completion rate, and user satisfaction when using Canary.

Total Time of Session: 1 hour
Logistics
The study will be held in person. It is preferred that the participant use this application on their own mobile phone for this activity to rule out any bias that might arise due to lack of familiarity with the interface.
Introduction and Asking for Consent (5 minutes)
Hello <Participant>, I am <X>, and this is <Y>. We are graduate students at Cornell Tech researching the usability of Canary — an IoT product that combines a traditional phone application with a tactile object to assist neurodivergent users when they’re experiencing sensory overload.

Thank you for agreeing to participate in this study. The purpose of this study is to examine how efficiently users are able to use Canary. For this study, we will start off with a short demo, followed by demographic questions before we begin the product evaluation. 

For one of the tasks, we will be playing loud noises which are meant to replicate an overstimulating environment — similar to that of a subway station during rush hour. This task is optional. In order to participate in this task, we require you to 1). have a commonly used stimming device currently on your person in case this user evaluation still leaves you feeling overstimulated and 2). fill out this consent form stating that you would be willing to participate in this exercise. If you decide to participate, you can still ask us to stop the experiment if the environment is too overwhelming. You can do so by pressing this button. By consenting, you acknowledge the potential for discomfort during this task. Do we have your consent for participating in Task 3?

<Participant answers and signs consent form>

We will be demonstrating certain features of the application now. Do you have any questions about the study before we begin?

<Answer any questions that come up>
Demo (5 minutes)
We will be demonstrating how to connect the Canary Ball (an IoT device) to the phone app. We will be walking the users through the process step-by-step, highlighting key actions such as Bluetooth pairing and adjusting settings. There will also be time for the users to ask questions and interact with the app and device during the demo.

<Perform demonstration>
‍
During the product evaluation portion of this study, we will be timing your task completion time. Additionally, you’ll be asked to report your emotional reaction to the task after completing each task. 
Demographic Questions
1. What is your age?
2. Which gender do you identify with?
3. How do you identify in terms of ethnicity (Hispanic or non-Hispanic)?
4. How do you identify in terms of race?
5. What is the highest level of education you have completed?
6. Have you experienced sensory overload in the past?
7. Have you used any technologies to distract or calm yourself down while experiencing sensory overload in the past?       
     a. If so, which ones?
task 1
Navigate through the app and launch a game. Additionally, adjust the speed and haptic settings of the game to that of your liking.
task 2
Navigate through the app and launch the meditation feature. Choose an activity of your choice and complete the guided meditation activity.
task 3
Note: For this task, we will be playing a series of loud noises that can cause you to be overstimulated. As mentioned in the consent form, you can ask us to stop the study at any point and you are encouraged to bring a familiar item for stimming in case the exercise leaves you feeling overstimulated.

As loud noises are playing to simulate a subway station during rush hour, navigate through the app in an attempt to calm or distract yourself. Additionally, you can make use of the Canary Ball for stimming.
Data Collection
Time taken to complete tasks (Speed)
• For most tasks, we want the users to navigate to each mode in roughly 30 seconds. If they are unable to do that, we want to take input from them and redesign the app in a way that each mode (games, meditation, etc) is accessible from the home screen. Our rationale is that when users are experiencing sensory overload, they would want to quickly launch the app and use a feature.
• How we will collect this data: The tasks will be timed.
Emotional reaction to tasks
• Our goal is to have the user feel calm after using any feature on the app. If the user feels overstimulated, frustrated, or angry while using the app, we want to take their feedback and redesign the features.
• How we will collect this data: This will be self-reported by the user.
Task Completion
• We want the users to be able to navigate through the app without any additional help or information.
• How we will collect this data: We will be noting down their ability to complete the task without any errors or additional information as they are completing the tasks.
Post interview
1. How would you describe your experience using Canary?
2. How easy or difficult was it to learn how to use Canary?
3. How easy or difficult did you find the tasks to be for you?
4. On a scale of 1 to 5 (with 1 being not at all efficient and 5 being very efficient), how efficiently did you feel you were able to complete the tasks?
5. Did the use of Canary leave you feeling calmer than before using the app?
6. Would you use Canary again?
Wrap Up
• Do you have any questions? 
• Is there anything else you would like to share with us?
Thank you for your participation today.
A pink and purple floral divider

Takeaways

This project was a rather thorough walkthrough of traditional UX research methods in a relatively short period of time. I discovered that I quite enjoy conducting semi-structured interviews, especially when I get to probe into some answers to gain deeper insight into the participants’ rationale. While the process of qualitative coding can be grueling (especially when it comes to correcting machine-automated transcripts), I found it to be an incredibly helpful method for compiling our findings and uncovering what tied them all together.

A pink and purple floral divider

Next Steps

If I were to continue with this project, I would recruit at least 16 neurodivergent participants to test our product. I would then work on the next set of iterations for Canary based on the feedback.