Illusion Spaces in VR: The Interplay Between Size and Taper Angle Perception in Grasping

Jian Zhang, Wafa Johal, Jarrod Knibbe · 2025

Read Illusion Spaces in VR: The Interplay Between Size and Taper Angle Perception in Grasping by Jian Zhang, Wafa Johal, Jarrod Knibbe (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-2504-05791.txt.

Introduction

Virtual hands can trick the eye and touch sense, and Jian Zhang, Wafa Johal, and Jarrod Knibbe mapped how two object traits bend that trick, in work from twenty twenty-five. The traits are size and taper slant. Past work had probed traits one at a time, not as a pair.

In a forced-choice test, people judged virtual objects against real ones they held. As virtual size grew, users felt the slant grow too. As virtual slant shrank, users rated size too high. The team sums the effect up as an illusion space, with a math model and a tool to match. Designers of touch gear for virtual worlds can use the map to pick stand-in objects that feel right.

Catalog record

CREATORJian Zhang, Wafa Johal, Jarrod Knibbe
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE63 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-05791

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Illusion Spaces in VR: The Interplay Between Size and Taper Angle Perception in Grasping Jian Zhang, Wafa Johal, Jarrod Knibbe Abstract . Leveraging the integration of visual and proprioceptive cues, research has uncovered various perception thresholds in VR that can be exploited to support haptic feedback for grasping. While previous studies have explored individual dimensions, such as size, the combined effect of multiple geometric properties on perceptual illusions remains poorly understood. We present a two-alternative forced choice study investigating the perceptual interplay between object size and taper angle. We introduce an illusion space model, providing detailed insights into how physical and virtual object configurations affect human perception. Our insights reveal how, for example, as virtual sizes increase, users perceive that taper angles increase, and as virtual angles decrease, users overestimate sizes. We provide a mathematical model of the illusion space, and an associated tool , which can be used as a guide for the design of future VR haptic devices and for proxy object selections. Conference: CHI Conference on Human Factors in Computing Systems; April 26-May 1, 2025; Yokohama, Japan CHI Conference on Human Factors in Computing Systems (CHI ’25), April 26-May 1, 2025, Yokohama, Japan DOI: 10.1145/3706598.3714162 ISBN: 979-8-4007-1394-1/25/04 CCS: Human-centered computing Virtual reality CCS: Human-centered computing Empirical studies in HCI CCS: Human-centered computing HCI theory, concepts and models CCS: Human-centered computing Haptic devices † † cc-license: by 1. Introduction Accurate haptic feedback remains a grand challenge of virtual reality (VR). While visual and audio resolution continues toward realism, haptics remain relatively limited. One of the opportunities here is to leverage physical objects from the space around the user to provide haptic feedback for virtual objects in VR. By leveraging visuo-proprioceptive illusions, these objects do not have to be an exact match for one another. Understanding what virtual objects a physical object can represent remains a challenge. To date, research has considered the impact of individual geometric properties in isolation, and are yet to consider how different properties may impact one another and our perception. This style of illusory, nearby-object-based haptic interaction is typically referred to as a passive haptics approach. These solutions benefit from being just-in-time, device-free, ad-hoc, and, (conceptually) scalable ( Clarence et al., 2021 ; Clarence et al., 2024 ; Sharma et al., 2024 ) . At the same time, they are constrained by perceptual illusion limits that require careful design and precise application to not adversely impact the user experience ( Gani et al., 2022 ) . Much of the existing literature on passive haptics has explored object co-location (considering where the physical proxy object is, such that it may feel as though it is in the same location as the virtual object, e.g., through haptic retargeting ( Zenner and Krüger, 2019 ; Clarence et al., 2021 ; Clarence et al., 2024 ; Azmandian et al., 2016 ) ), rather than object similarity (considering what the object is, such that it may feel like the same in material and geometry) ( Nilsson et al., 2021 ) . This exploration of co-location has revealed that objects can be approx. 15° around an arc to the left or right of the virtual object ( Clarence et al., 2022 ) and between 0.87x and 1.31x as far from the user ( Esmaeili et al., 2020 ) . Explorations of similarity have provided insights into the minimum and maximum size the proxy can represent ( Bergström et al., 2019 ) , and the perception of separate factors including sizes, angles and curves on the surface of the object ( de Tinguy et al., 2019 ) . So far, these works have considered object properties in isolation. However, the properties’ impacts on hand-pose, and joint forces and torques are not isolated – size and shape together drive the finger spacing, orientations, and joint configurations of grasping, for example. How object shapes and sizes interplay to impact our perceptual limits remains poorly understood. We study the illusory limits and interplay of size and taper angle of physical proxies, using real-time, virtual remapping techniques where users’ fingers are retargeted whilst reaching for the objects . We produced physical hexahedron objects for grasping, with 3 widths (3 cm, 6 cm, and 9 cm) and 3 taper angles (0°, 8°, and 16°, between the side face and the vertical direction). We conduct a study with 40 participants, grasping the physical objects with differing virtual counterparts (6 virtual sizes \times 7 virtual angles for each physical object). By studying these object properties together, we reveal a complex, multi-dimensional illusion space across both size and taper angle. For example, our results show how increasing the virtual size of an object will strongly increase the perceived angle of that object, while increasing the virtual angle makes the perceived size smaller. We provide a set of mathematical expressions that describes a proxy object’s haptic capacity and reveals how size and taper angle interplay with each other in creating illusions. We also discuss how these results may feed into mechanical, actuated, active haptics controller designs, to facilitate super-resolution haptic devices. Our contributions: (1) we study, reveal, and describe the interplay between size and taper angle when grasping objects in VR (2) we provide visualisations of the multi-dimensional illusion space, to provide designers with guidelines on VR haptic perception (3) we provide mathematical functions that enable designers to calculate whether a given physical object can represent an intended virtual object. (4) We provide an online tool that allows designers to calculate physical-virtual object pairings within the illusion space. 2. Related Work When designing for advanced haptics in VR, researchers have to consider both co-location – ensuring the haptic feedback is delivered in the correct location – and object similarity – ensuring the haptic feedback matches the material and geometric properties of the virtual object ( Nilsson et al., 2021 ) . In this section, we begin by discussing the different approaches to increasing haptic fidelity, before presenting the current knowledge in illusory techniques and the associated perceptual limits. 2.1. Providing Haptics in VR Research has been exploring two avenues for achieving advanced haptics: (i) active haptics, and (ii) passive haptics. Some mechanically reflect the exact properties of the virtual objects (active controllers, e.g., Choi et al. (2016) ) while others use proxy objects and illusions to render them (passive approaches, e.g., Bergström et al. (2019) ). Much of the recent effort in active haptics has focused on handheld and wearable device designs. These can be portable and are most akin to the controllers we see being adopted commercially. Prominent examples here include wearable controllers for grasping rigid objects ( Choi et al., 2016 ; Choi et al., 2018 ) , axisymmetric devices for grasping pseudo-cylindrical objects ( Gonzalez et al., 2021 ) , and multiple degree-of-freedom controllers for grasping asymmetric objects ( Kelesbekov et al., 2024 ) . To date, however, these controllers have remained specialist, cumbersome, and heavy. On the other hand, grounded and encounter-type devices provide just-in-time feedback, that is only available at the exact moment the user needs it and can leave them otherwise hands-free. Examples of such devices include the commercially available Touch 1 1 1 3D Systems, USA - formerly Phantom from Sensable Technologies and

Read the complete text