Craving Checkpoint
An Interactive Fridge Lock for Mindful Eating
Nov 20224 - Dec 2024
Teammate
Xiaoman Yang
Zijie Zhou
Mentor
Marcelo Coelho,
Daniel Leithinger
Product Design, Design Research, Concept Development, Design Engineering, Programming, 2D/3D Prototyping, Interaction Design, Visual Design
- A revised end-product after the usability test
- An academic Paper accepted by NeurIPS
Project Description
Traditional dietary interventions often rely on restriction, tracking, or delayed reflection, which can limit their ability to foster lasting change. Craving Checkpoint, a Large Language Object (LLO) in the form of an interactive fridge lock designed as a Just-In-Time Adaptive Intervention that supports 5 mindful eating through embodied and emotionally expressive interaction.
Background
In today’s fast-paced world, people often consume food without fully understanding its impact on their bodies, emotions, and overall well-being. Modern dining experiences lack personalization, missing opportunities to connect individuals with their unique dietary needs, emotional states, and culinary aspirations.
Discovery.
Initial Challenge
Short-term Mindfulness in Eating Habits
While popular interventions such as weight-management apps, structured meal plans, and commercial programs can produce short-term weight loss, they often fail to sustain long-term behavior change.
Emotional and Psychological Barriers
Forcing users to eat things they do not want to eat creates emotional and psychological barriers, leading to anxiety, stress, or other negative emotions.
User Research
Learn what people care about when they consume food
01 Method-Questionnaire
The questionnaire help us understand the key factors that contribute to an ideal dining experience, as well as the decision-making process behind users' food choices. A total of 60 participants completed the survey.
Food Choice :
Environmental Factor:
Social Challenges:
Food Health:
Environmental Impact:
To further expand the research, I conducted a speculative dining experiment utilizing the power of AI. In this exhibition space, I designed a smart dining setup capable of detecting food intake along with its precise quantity. Using advanced analysis, the system processed this data in real time and translated it into a dynamic, generative lighting experience. The lighting served as a narrative tool, visually illustrating the impact of food consumption on both the human body and the environment. I then asked people to vote on the stand I manually made and interviewed a few questions about their thoughts on the speculative dining experiment.
Data Collection
Food intervention needed to be embedded within a more natural and accessible medium.
Through interviews and responses from the stand, I discovered that people become more conscious of the properties of the food
when receiving relevant information during the act of eating. However, while the generative lighting experience provided an engaging visual representation, it was ultimately a disposable feature that did not seamlessly integrate into daily life. If the goal was to reinforce behavioral change or incorporate educational elements into everyday routines, the intervention needed tobe embedded within a more natural and accessible medium.
Additionally, the findings revealed that people are generally more concerned with how food impacts their bodies—such as nutritional value and calorie intake—rather than broader environmental factors like carbon footprint. In many cases, food choices were driven by immediate emotional states, or sometimes long-term health, but neither sustainability considerations. This insight highlighted the need to design a solution that aligns with personal motivations and habits rather than relying solely on abstract global concerns.
Ideation.
- Help users develop thoughtfulness about their food choices by providing intuitive feedback.
- Increase awareness of body status and the impact of food on health through real-time insights.
Target User
- Individuals who prioritize food ingredients and their impact on the body.
- Patients experiencing digestive or dietary-related issues or who need to cultivate mindfulness in their eating habits.
HMW Question
How might we change our eating routine to inspire
reflection and mindfulness when consuming food?
reflection and mindfulness when consuming food?
To identify the most effective and impulsive intervention point within the act of eating, I need to first interpret the routine of eating in the daily life and understand the users’ mindfulness in each section.
I also created a matrix outlining the properties and potential impact of each medium. Through this analysis, the fridge and utensils emerged as the most balanced option.
Along with the journey mapping, we identified that a device attached to the fridge, used at the moment of deciding what to eat, would be the most effective way to trigger mindful reflection on food choices. At the same time, to ensure feasibility, we designed a fridge lock that requires users to pause before making an unconscious food choice. This mandatory moment of reflection encourages more intentional eating habits and directly addresses the challenge of mindless consumption.
Design.
Process
Initial User Flow
The prototype encompasses a range of options that we believe are essential for fostering self-reflection. The output is designed to be diverse, engaging both visual and auditory elements to enhance the experience.Process
Physical Interface
For the final product's interface design, we went through multiple iterations to achieve a balance between functionality and the physical interaction experience across various buttons.The initial version emphasizes an OLED screen for visual output. All input interfaces are integrated into a single rectangular device, with their placement and design prioritized according to the significance of each input.
Input
Food Category (discrete): Slider
Purpose of open fridge (discrete): Rotary Knob
Hunger level (continuous): Slider
Time of Meal (continuous): Encoder
Play & Stop & Unlock: Button
Output
Dietary Story: Speaker
Entertainment lighting: OELD Screen
Servo
Goal: Use the device to self-reflect before picking up the food
Average time used: 1min21s
Frustration:
- All of the participants showed a certain level of confusion about the functionality of each button without instructions due to an overwhelming number of input options.
- All of the participants found it difficult to distinguish between a full meal and individual food items or lacked the necessary knowledge to analyze or categorize the food they were inputting
- 2 of 3 are struck at first because they don’t know where to start.
- 1 of 3 expressed his concern about the excessive size of the device to be attached to a fridge.
- 1 of 3 felt pressure when inputting the amount of the food she’d like to consume.
- 2 of 3 found the mandatory pause before the fridge would help them set up their mind and push them to craft their food choice.
Iteration 2
According to the user feedback, we switched to the emotion wheel that might impact the user food choice.
Input
Purpose of open fridge (discrete): Hall Effect Sensor + Magnet
Emotion (discrete): Encoder Knob
Hunger level (continuous): Rotary Knob
Time of Meal (continuous): Computer Data
Play/Unlock: Button
History: Button + LED indicator
Output
Dietary Story: Speaker
Servo
User feedback - Iteration 2
Goal: Use the device to self-reflect before pick up the food
Averge time used: 0:34
Frustration:
- All of the partipants are confused between the long-term goal and short-term goal of eating
- All of the partipants are unsure about the importance of each button.
- All of the partipants expressed satisfaction towards the simplifed interface and minimized button number.
- All of the partipants loved the design of the emotion wheel, which displayed emotions within metrics while keeping the other options hidden.
Final Design
Updated User Flow
The new user flow simplified the choices user need to consider and automate the process of locking. For the output result, inspired by the poetic simplicity of Haiku, our product transforms these reflections into personalized, thoughtful verses.
The Power of Artifical Intelligent
For transforming the user input into the poetic haiku, we design our LLM prompts in code.
Physical Interface
To enhance usability, the system limits users to three key data input categories. This reduction simplifies the learning curve and minimizes interaction time while still ensuring users can self-evaluate.
The following blueprint showcase the hardware layout inside the compact box.