Source-linked AI summary
ScentEcho: Exploring Adsorbent Materials for Accurate Odor Collection and Playback
Chih-Hung Lee, Yuchi Sun, Rui Zhang, Suhang Wei, Qi Lu
TL;DR
Realistic olfactory interaction is limited by the difficulty of reliably capturing, preserving, and replaying real-world scents beyond predefined sources. This study presents ScentEcho, a portable modular system that evaluates adsorbent materials through user ratings and interviews. Results show odor-specific material performance, a strong association between perceived intensity and similarity, and graded recognition from broad categories to specific sources.
Problem
Current olfactory systems rely on predefined odors or fixed source materials, limiting personalized and contextually authentic scent interaction.
Method
ScentEcho evaluates adsorbent materials for collecting and replaying lemon, coffee, mint, and lavender, using similarity and intensity ratings plus qualitative interviews.
Results
Different adsorbent materials performed better for different odors, while perceived similarity and intensity showed a Pearson correlation of 0.8295.
Takeaways & Limitations
Odor fidelity goals can account for the interaction between similarity and intensity and the graded progression from category-level to source-level recognition.
Takeaways & Limitations
The study used 12 participants aged 23–35, and reproduced odors were generally weaker than the originals.
Abstract
from arXiv · showhide
Delivering odors that feel realistic and recognizable remains a core challenge for olfactory interaction systems, particularly in applications that demand precise scent delivery. A key limitation lies in the difficulty of capturing, preserving, and playing back real-world scent sources in a reliable and scalable manner. This study explores the potential of adsorbent materials for supporting realistic scent playback. We present ScentEcho, a portable system that enables modular scent collection and release. Through user evaluations, we identify which adsorbent materials tend to perform better for specific odors, and observe that perceived intensity strongly influences similarity ratings. In addition, odor recognition follows a graded pattern, with users moving from broad category identification to more specific source recognition as similarity increases. These findings offer practical insights for designing olfactory interfaces that are both expressive and perceptually aligned with user expectations.
1 INTRODUCTION
Olfactory interfaces need more personalized and contextually authentic scent inputs than predefined fragrances and fixed mixtures provide. ScentEcho addresses this gap by collecting and replaying real-world odors while evaluating adsorbent materials and user perceptions.
- Predefined fragrances and fixed source materials constrain personalization and contextual adaptability in olfactory interaction.
- Existing scent-collection concepts often lack validated material or system performance, while the smell camera remains bulky, difficult to reuse, and unable to store or switch among multiple scents.
- ScentEcho is a compact, portable device for personalized scent collection and playback using modular storage and a simple pneumatic system.
- The study evaluates adsorbent materials for lemon, coffee bean, mint, and lavender, with participants rating replayed-scent similarity and intensity.
- Participant ratings and interviews revealed a correlation between perceived intensity and odor similarity, informing understanding of olfactory realism.
2 RELATED WORK
Related work shows a need to move beyond convenient commercial fragrances toward scent collection and playback methods that better reflect lived experiences. Existing alternatives remain limited by inaccurate descriptors, practical burdens, or insufficient technical validation.
- Commercial perfumes and synthetic fragrance compounds are convenient, but their labels or descriptors may not accurately reflect olfactory perception.
- Realistic and semantically aligned olfactory cues are especially valuable for evoking emotion, triggering memory, and supporting interpersonal communication.
- Systems should support more diverse, personalized scent inputs that reflect real-life contexts and meaningful interactions.
- Solvent extraction and hydrodistillation can capture odors accurately but are time-consuming, equipment-intensive, and require substantial raw materials.
- Prior physical scent-collection systems introduce creative interaction concepts but often lack rigorous technical evaluation, and the smell camera has limited practical deployment because of bulkiness.
3 DESIGN AND IMPLEMENTATION
ScentEcho uses separated control and storage compartments with pumps, valves, and modular odor units to collect and replay multiple scents. Its compact construction supports one-handed operation, maintenance, and adsorbent-material testing.
- The device separates an upper electronic control compartment from a lower compartment containing four independent odor storage units.
- Two air pumps and four individually controlled solenoid valves direct airflow for odor collection and playback.
- The modular lower compartment connects magnetically to the control compartment, enabling inspection, replacement, and future expansion of odor storage units.
- Users press self-locking buttons to select storage units and toggle a switch to activate the collection or playback pump.
- ScentEcho measures 127×68×47 mm and weighs 248 g, supporting compact, one-handed operation and maintenance.
- The implemented device was used to test how different adsorbent materials capture and replay various odors.
4 SCENT PLAYBACK PERFORMANCE EVALUATION
The evaluation compares adsorbent materials for reproducing four everyday odors through similarity and intensity ratings, supplemented by interviews about odor fidelity. Controlled procedures and Pearson correlation analysis examine both material performance and the relationship between perceptual dimensions.
- Experiment Setup: The study evaluates adsorbent materials for lemon, coffee beans, mint, and lavender using user experiments focused on reproduced-odor similarity and intensity.
- Experiment Setup: The four odorants were sourced from local markets, with storage conditions adjusted for refrigerated lemon and mint samples before adsorption.
- Experiment Setup: Experiments were conducted at 25±2◦C and 50±10% relative humidity with participant informed consent.
- User Evaluation Procedure: Participants rated replayed odors against target odors on a 1–7 Likert scale, where 7 represented an identical match.
- User Evaluation Procedure: Pearson correlation analysis related similarity ratings X and intensity ratings Y using their deviations from the respective means.
- User Evaluation Procedure: Participants could re-smell targets, rated within 10 seconds, and answered scoring and exploratory questions about odor perception.
5 RESULTS
ScentEcho evaluations showed that adsorbent-material performance varied by odor and that perceived intensity was closely associated with similarity judgments. Participants also exhibited graded odor recognition, moving from broad categories toward specific sources as similarity increased.
- Material A achieved a 4.5 average similarity score for mint, while material D achieved 5.3 for lavender.
- Coffee similarity scores ranged from 4.2–4.3 and lemon scores from 3.9–4.1 across the four materials.
- Material D reached an intensity score of 5 for lavender, while materials A, B, and C each exceeded 4 for coffee.
- 0.8295 was the Pearson correlation coefficient between perceived odor similarity and intensity across odors and participants.
- Among 12 participants, 7 had difficulty separating similarity from intensity, especially for weak reproduced scents.
- Similarity scores of 3–4 supported broad category associations, whereas scores of 5–6 supported specific source identification.
6 DISCUSSION
The discussion identifies intensity as a major factor shaping odor similarity judgments and recognition. It also frames odor recognition as graded and highlights practical constraints affecting ScentEcho’s refinement and deployment.
- 6.1 The Interdependence of Similarity and Intensity in Odor Perception: 0.8295 correlation between intensity and similarity indicates that users often conflate these dimensions when evaluating odors.
- 6.1 The Interdependence of Similarity and Intensity in Odor Perception: Reduced intensity can obscure odor characteristics and impair recognition, while sufficient intensity allows more independent similarity evaluation.
- 6.1 The Interdependence of Similarity and Intensity in Odor Perception: 8 out of 12 participants distinguished auditory timbre from volume, contrasting with the more entangled odor dimensions reported for olfaction.
- 6.1 The Interdependence of Similarity and Intensity in Odor Perception: Playback systems should ensure sufficient intensity alongside odor-profile similarity, especially when precise recognition is required.
- 6.2 Hierarchies of Odor Recognition and Design Implications: Odor recognition follows a graded spectrum: scores of 3–4 evoke categories, while scores of 5–6 support specific sources.
- 6.3 Practical Considerations and Future Directions: The study was limited by 12 participants, weaker reproduced odors, possible metallic interference, and an air-pump design that may affect lifespan and airtightness.
7 CONCLUSION
The conclusion presents ScentEcho as a portable, modular platform for personalized scent collection and playback. The system connects adsorbent-material evaluation with perceptual findings and supports future expansion and application in contextual olfactory interaction.
- ScentEcho is a portable system for personalized scent collection and playback that uses modular hardware and evaluates adsorbent materials.
- User studies showed that scent similarity and intensity interact in perception, underscoring human-centered evaluation in material selection.
- The system currently supports four odor channels, while its modular structure is designed to accommodate future expansion.
- ScentEcho’s flexible design opens possibilities for targeted scent delivery, memory recall, and multisensory storytelling.
I. Personal Information
The section records participants’ perfume-use habits and self-rated sense of smell, alongside instructions for scoring odor similarity and intensity against a seven-point standard.
- Personal Information: Participants reported how frequently they used perfumes or fragrance-related products.Responses ranged from never to daily.
- Personal Information: Participants rated whether they considered themselves to have a keen sense of smell.The response scale ranged from strongly disagree to strongly agree.
- Evaluation Instructions: Similarity and intensity ratings used the standard sample as a seven-point reference, from completely dissimilar or no odor to identical.Samples A–H were evaluated relative to this reference.