An Experiment that Challenges

Visual Dominance

Smell Loss Is Real, yet Unevenly Felt

Smell is an intricate bodily interaction with chemicals carrying strong emotional impact, yet in a visually dominated post-pandemic world, it has grown increasingly marginalized. A rapid six-question survey (N = 41) and on everyday smell awareness and impact and responses from a participatory session with sticky notes revealed polarized impact, high awareness among affected individuals, and persistent difficulty in describing smell.

Design Research

Cross-Modality Sensory Augmentation

Research Design: Visual Priming

I recruited 18 participants randomly assigned to either a control group viewing neutral images (e.g., campus, soda water) or a treatment group viewing smell-associated scenes (e.g., coffee shop, steak, grassland); Participants then answered open-ended mindfulness questions on their immediate sensory experiences, testing whether increasing the mental accessibility of smell cues would heighten attention to smell.

01 Thematic Coding

I binary coded each sentence into 1 / 0. I marked 1 if a smell is mentioned in a sentence, otherwise it’s a 0 (N = 50). Same procedure is applied for both treatment and control group participants (N = 18).

02 Hypothesis Testing

Chi-square test with χ²(1, 50) = 5.06, p = .025, Cramér’s V = .28 indicated that visual cues heightened subjects’ ambient smell awareness with a moderate effect, suggesting the feasibility of using vision to augment smell.

Integrated Product Design

Implementation Highlights

Informed by research, I investigated how real-time olfactory experience can be visually augmented. I used MQ-2 and DHT temperature & humidity sensors to capture ambient conditions that affect smell experiences. My implmentation on Processing maps sensor data in Processing to graphic attributes and triggers responsive motion graphics during heightened smell presence or alerts when potential hazards are detected.

01 Ambient Risk Awareness

Odorable alerts people when ambient gas intensity, humidity, and temperature is continuously off the normal baseline, offering instant assitance.

02 Pixel-Style Computer Graphic

Odorable visualizes gas molecules through pixel-style graphics. In Processing, I implemented a Matrix ADT that maps 1D binary arrays into scalable 2D pixel forms, forming a reusable visual system. This foundation is paired with a Particle ADT to animate individual elements for interactive visualization.

03 Custom Sensitivity Control

Users customize when and how the device responds by setting thresholds across time, intensity, and location, ensuring visual cues aren’t intrusive.

Iterative Form Giving

Final Design

01 Ambient-Aware Animation

(1) default state displays time & ambient conditions, while (2) “odorable” state contextualizes significant gas fluctuation via ambient-aware particle animations. If dangerous signs are detected, alert animation will override other states.

02 Mobile App: Device Adjustment & Data & Visualization

The mobile app extends the smartwatch with sensitivity controls and data views. Each smell event can be tagged, annotated & photographed, forming a spatial–chronological map of the user’s scent experience.

03 Ergonomic Wristband: Odorable, Intimate, Personal

An ergonomic, curvilinear wristband that users can trust as an extension of their senses, with a single-sided elastic strap that’s comfortable, easy to wear, cost-efficient, and ready for sensor integration.

Reflection: From Abstract to Concrete

I replicated a psychological research and translated its insights into tangible systems. Over 2 months, I moved beyond my C++ foundation by self-drilling motion graphics and UI development in Processing, prototyping embedded interactions with Arduino, sketching and modeling the physical form in Fusion 360, and crafting a stylized pixel-based visual system. Aside from the mobile companion app, every element of this project was fully implemented.