Immersive and Wearable Thermal Rendering for Augmented Reality
Alexandra Watkins, Ritam Ghosh, Evan Chow, Nilanjan Sarkar · 2025
Read Immersive and Wearable Thermal Rendering for Augmented Reality by Alexandra Watkins, Ritam Ghosh, Evan Chow, Nilanjan Sarkar (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-20646.txt.
Introduction
Alexandra Watkins, Ritam Ghosh, Evan Chow, and Nilanjan Sarkar build a wearable device that lets augmented reality users feel heat and cold, in a 2025 paper. Feeling temperature in virtual reality is one thing, but augmented reality mixes virtual objects with the real workspace. Users still need nimble fingers and honest signals from real surfaces.
Their prototype mounts on the palm rather than the fingertips, leaving the fingers free. It can relay the true temperature of a touched object, and it paints changing heat patterns across the palm that shift over space and time. In human-subject tests, the palm placement cost a little realism but not comfort or immersion. Users could tell real from rendered temperatures. Moving patterns clearly beat static warmth for immersion and realism. The authors frame these as workable design rules for thermal touch in augmented reality.
Catalog record
| CREATOR | Alexandra Watkins, Ritam Ghosh, Evan Chow, Nilanjan Sarkar |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-20646
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Immersive and Wearable Thermal Rendering for Augmented Reality Alexandra Watkins, Ritam Ghosh, Evan Chow, Nilanjan Sarkar Abstract We present a proof-of-concept palm-mounted thermal feedback prototype addressing thermal rendering challenges specific to augmented reality (AR), where users must interact with both real and virtual objects in their physical workspace. In contrast to thermal feedback systems developed for virtual reality, AR thermal feedback must preserve manual dexterity, maintain access to real-world thermal cues, and provide coherent virtual temperature sensations without obstructing natural object interaction. We propose three AR-specific design considerations, which our prototype implements: indirect feedback to preserve fingertip dexterity, active thermal passthrough to sense and render the temperature of contacted physical surfaces, and spatially and temporally varying thermal rendering across the palm. Human-subject experiments evaluated perceptual sensitivity, indirect feedback, active thermal passthrough, spatial pattern recognition, and moving thermal rendering during AR interaction. Results showed that although indirect feedback reduced perceived realism during visual contact at the fingertips, it did not reduce immersion or comfort; active thermal passthrough supported temperature discrimination between real and rendered surfaces; and spatiotemporal rendering significantly improved immersion and realism compared with static thermal stimulation. These findings suggest that our design considerations are viable design strategies for AR thermal haptics, while also clarifying tradeoffs for applications that require precise realism versus broader immersive thermal experience. 1 Thermal sensations are an important part of everyday object interaction. Warmth and coolness help people interpret material properties, environmental conditions, comfort, and safety, such as when holding a hot beverage or touching a cold metal rail. These experiences are mediated by thermoreceptors in the skin, which detect changes in temperature and support material discrimination and behavioral responses [ 21 ] . In immersive systems, thermal feedback can therefore contribute to presence, interactivity, and realism by making virtual or augmented objects feel more consistent with the physical experiences they represent [ 39 , 13 , 33 ] . As immersive technologies are increasingly used in education, workforce training, healthcare, and industrial processes [ 20 , 43 , 6 ] , the need for effective multisensory feedback has also increased. Prior work has shown that the fidelity of immersive feedback can influence learning outcomes, task efficiency, and user satisfaction [ 49 , 31 , 1 , 15 ] . Thermal feedback is one important component of this broader challenge, particularly in AR, where virtual and augmented objects must be perceived in relation to the physical surfaces, materials, and interactions that surround them. AR introduces design constraints that differ from those of virtual reality (VR). AR systems overlay digital information onto the user’s physical environment, allowing virtual content to be perceived and manipulated in relation to real objects and spaces [ 36 , 4 ] . In contrast to fully virtual environments, AR users may need to grasp actual tools, touch real surfaces, and maintain awareness of physical material properties while simultaneously receiving feedback from virtual elements. Thermal feedback for AR therefore requires more than adding temperature cues to the hand; it requires strategies that balance physical interaction, real-world thermal perception, and coherent virtual sensation. Although recent AR head-mounted displays provide increasingly sophisticated visual and auditory feedback [ 2 ] , haptic and thermal feedback remain important for supporting the perceived presence of virtual objects and interactions [ 12 , 27 , 41 , 5 , 46 ] . Compared with force, vibration, and tactile feedback, thermal feedback remains relatively underexplored in immersive interfaces [ 39 ] . Existing systems demonstrate that thermal stimuli can improve realism and presence [ 21 , 17 , 8 ] , and recent wearable or multimodal devices have incorporated thermal cues into gloves, rings, handheld devices, and actuator arrays [ 48 , 38 , 58 , 19 ] . These approaches show the value of thermal feedback, but they also reveal a design conflict for effective use in AR. Devices that cover the fingers or require handheld controllers [ 3 ] may interfere with grasping, manipulation, or natural thermal sensing of physical objects. Such compromises may be acceptable in fully virtual contexts, but they become more consequential when users must interact fluidly with both physical and virtual entities in the same workspace. We believe these AR-specific constraints motivate three design considerations for thermal feedback, as demonstrated in the use case of Figure 1 . First, indirect feedback can relocate thermal actuation away from the fingertips to reduce interference with manual dexterity. Second, active thermal passthrough can address the loss of natural thermal sensing caused by device occlusion by actively measuring the temperature of contacted surfaces and rendering that temperature to the skin. Third, spatiotemporal thermal rendering can support thermal sensations that vary across the contact surface and over time, allowing virtual thermal events to better correspond to dynamic AR interactions. Together, these considerations frame thermal feedback for AR as a problem of balancing real-world touch, virtual object feedback, and mixed physical–virtual coherence. Figure 1: Haptic-mediated thermal interactions in AR enabled by the proposed design considerations. ( a ) Physical interaction. Leaving the fingers unobstructed supports fine manipulation of physical tools or objects and preserves direct thermal sensing at the fingertips. ( b ) Active thermal passthrough. Temperature sensors on the external surface of a haptic device allow collocated thermal actuators on the interior surface to render the thermal properties of contacted surfaces. ( c ) Indirect feedback. Thermal sensations applied to nearby occluded regions, such as the palm, can provide thermal cues during virtual interactions while avoiding actuation on the fingertips. ( d ) Spatiotemporal rendering. Independently controlled and spatially arranged thermal actuators enable sensations that vary across the contact surface and over time, such as the walls of a virtual cup being covered and uncovered by warm liquid. 1.1 Present Work In this manuscript, we present a novel proof-of-concept palm-mounted thermal-rendering prototype that implements these design considerations for AR (Figure 2 ). The palm provides sufficient surface area for multiple thermal actuators while leaving the fingertips available for grasping, manipulation, texture perception, and direct contact with real objects. This form factor allows the device to explore indirect thermal feedback, active thermal passthrough, and spatially distributed thermal rendering without fully occupying the contact surfaces most often used for precise manual interaction. The system combines independently controlled thermoelectric modules (TEMs) with outward-facing contact-detection and temperature sensors to provide warm and cool sensations, active thermal passthrough, and time- and space-varying temperature sensations rendering on the palm using only thermal actuators. We provide supporting evidence for our design considerations and their implementation in a proof-of-concept device using five controlled human-subject experiments. These experiments confirm the perceptual sensitivity to warm and cool stimuli on the palm as an appropriate location for haptic feedback, the user experience tradeoffs of indirect feedback, the ability of active thermal passthrough to support object temperature perception, recognition of spatial thermal patterns, and the effect of moving thermal rendering during AR interaction. This work contributes an AR-specific framing for thermal feedback, a novel wearable prototype implementation of three design considerations, and empirical evidence that these considerations can support thermal interaction while preserving the physical sensing and manual dexterity that distinguish AR from fully virtual environments. In particular, we demonstrate that the designed AR thermal interaction prototype for AR applications