Dance and movement-based artistic practices are a powerful nonverbal channel for emotional communication. The integration of robotic systems into these expressive contexts enables novel forms of performative art, while simultaneously raising concerns regarding the fidelity and perceptual accuracy of conveyed emotional information, particularly when affective content is mediated through non-humanoid embodiments. This study addresses a key limitation of robotic dance previously identified: although emotional valence and stylistic coherence were largely preserved, the resulting robotic performances exhibited diminished clarity in the perception of emotional arousal. To overcome this limitation, we investigate the role of haptic feedback as an additional expressive modality to support emotional communication in robotic dance. Visual performances by both human and robotic dancers are augmented with thermal and vibrotactile cues designed in accordance with established associations between haptic stimuli and emotional dimensions. A user study was conducted to assess how this multimodal augmentation influences the perception of emotional arousal. Results demonstrate that the inclusion of thermal or vibro-thermal haptic stimuli significantly enhances the interpretability of emotional content conveyed by a dancing robot. In particular, haptic augmentation of the audience experience improves the distinguishability of emotion intensity without altering the transmission of hedonic tone.
Orchestral conducting relies predominantly on visual gestures to convey timing, dynamics, articulation, cueing, and expressive intent, creating a significant accessibility barrier for Blind and Visually Impaired (BVI) musicians. This work investigates whether structured vibrotactile patterns, commonly referred to as tactons, can provide a non-visual channel for representing selected orchestral conducting cues, focusing on the design and perceptual evaluation of tactile encodings for conducting-related information. To this aim, a wearable sleeve equipped with vibrating motors in a simple layout was leveraged to implement established vibrotactile display principles such as spatially distributed stimulation, temporal patterning, and coarse intensity modulation. Preliminary experiments involving both sighted and BVI participants were conducted to assess stimulus localization, intensity discrimination, and the perception of combined vibrotactile features. The outcomes of these experiments informed the design of a set of tactons representing conducting intentions such as beat, entrance, cut-off, pointing, crescendo, and diminuendo. The resulting tacton set was subsequently evaluated with BVI participants through recognition tasks designed to assess its interpretability and effectiveness. Results demonstrate that the BVI cohort was able to reliably identify six tactons after limited training, achieving an average recognition accuracy of 77%, with individual accuracies ranging from approximately 60% to nearly 100%.
Thermal responses at the skin-material interface play an important role in human touch perception and material recognition. This study presents the design and characterization of a gel-based thermal contact sensor developed to approximate the transient heat-transfer responses of the human finger during material contact. The sensor incorporates a compliant, finger sized gel interface and an integrated heating system that reproduces an initial skin-like temperature condition at contact onset. The design is motivated by limitations of conventional rigid contact thermal sensors, including unstable contact conditions and discrepancies from human finger responses. Compared with a previous glass-based design, the proposed sensor improves measurement stability and achieves closer correspondence to human finger thermal responses, particularly for materials with mid-range thermal effusivity. However, the sensor exhibits faster and larger temperature changes than the human finger for high effusivity materials, highlighting the challenges of reproducing human-like interfacial heat transfer. We discuss possible con tributing factors, including thermal contact resistance, surface conformity, and the absence of fingerprint-like microstructure in the gel interface. The proposed system is positioned as a biomimetic thermal sensing platform for capturing human-like thermal contact responses under controlled conditions, with potential future applications in haptic interface evaluation, robotic material sensing, and data-driven material recognition systems.
Emotional and social expression are important dimensions of socially empathic human-robot interaction. In hug gable robot contexts, the chest is not only a stable bodily contact interface with strong affective relevance, but also a natural site for presenting haptic and auditory cues together. However, existing chest-haptic designs still focus mainly on physical tactile parameters or heartbeat-like patterns. Conversely, chest haptic patterns formed by different stimulation-location combinations over time may offer richer possibilities for multimodal expression and modulation. To explore this, we implemented a parameterizable 3×3 chest-mounted vibrotactile array on a huggable robot and constructed chest haptic patterns using path type and activation area as design parameters. We conducted two experiments. Experiment 1 examined how chest haptic patterns influenced the affective perception of emotional music. Experiment 2, under a fixed activation area setting, examined how the presence of chest haptics and different path conditions influenced the social emotional evaluation of verbal expressions with empathic intentions. The results showed that, in the emotional music context, chest haptic patterns influenced the direction of emotional rating shifts and showed more pronounced modulation effects under some music conditions. In the empathic speech context, the addition of chest haptics generally improved social-emotional evaluations. These findings provide empirical support for the use of chest haptic patterns in multimodal social-emotional expression, and offer guidance for the future development of huggable robots with richer emotional and empathic expression.
Haptic feedback can convey spatial information without engaging vision or hearing, making it well suited for eyes-free and assistive applications. However, many haptic guidance systems rely on cues that require explicit cognitive interpretation, which can slow user responses and introduce ambiguity in the conveyed spatial information. To address this, we introduce fingertip deflection as a kinesthetic guidance cue via the NURing, a wearable tendon-driven device that delivers controlled fingertip deflections to guide whole-arm motion, designed to leave the fingerpad unobstructed for tactile exploration. Inspired by the instinctive response of toddlers guiding parents by the finger [1], deflection cues provide a mechanically direct and readily interpreted form of haptic guidance. Two user studies evaluated closed-loop guidance in horizontal (single-axis, Study 1) and horizontal-vertical (dual-axis, Study 2) task spaces. In singleaxis tasks, blindfolded participants reached a median endpoint error of 11.3mm in under 3 s on average, and tracked moving targets with 53.5mm mean error. In dual-axis tasks, participants accurately identified deflection cue directions, and successfully reached two-dimensional target poses with a median endpoint error of 12.6mm in under 7 s on average. Kinematic analysis revealed that participants responded rapidly (median 330 ms), with early trajectories biased within 25° toward the target heading, indicating that deflection cues provided early, actionable direction information. These results demonstrate that fingertip deflection via the NURing enables fast, readily interpreted kinesthetic guidance that produces a coordinated whole-arm response without obstructing tactile contact, supporting its potential for eyes-free interaction and assistive applications.
With the rapid advancement of humanoid robotics and embodied intelligence technologies, numerous musical instrument-playing robots have emerged in recent years, such as pianos, chime bells, and taiko drums. These robots primarily employ open-loop positional control, rendering them incapable of operating instruments requiring dexterous hands and precise tactile perception, such as a violin, guitar, and guqin. This paper describes the design and validation of a high-precision tactile-sensing finger. By mimicking the shape of the fingertip and fingernail found on a human finger, we develop a biomimetic multimodal haptic fingertip and validate it on selected guqin string-contact tasks, including open-string and stopped-note comparisons, harmonic-tuning, and tactile-triggered bimanual coordination, using the guqin, a traditional Chinese musical instrument, as a challenging validation scenario rather than as a fully demonstrated robotic performance system. This research integrates tactile sensing with robotics technology, thereby contributing to applications in world heritage conservation and cultural dissemination.
In many areas of haptics, evaluation has prioritized objective task performance, with subjective aspects of user experience (UX), such as emotional response and user preference, often being treated as complementary measures. In this work we present a unified modeling framework for UX in haptic and physical human-robot interaction that simultaneously evaluates metrics of performance, emotional response, and preference as a function of system design parameters. Our framework uses Bayesian hierarchical modeling to capture both individual variability and population-level trends. We apply this approach to a line-following task conducted with a custom-built haptic knob platform in a study with 30 participants, in which we measure task performance together with subjective ratings of emotion and preference while manipulating damping, task difficulty, and feedback given to participants on their performance. Results show that while population-level models capture consistent effects on performance, emotion and preference did not show similarly consistent population-level trends. Instead, subjective responses varied substantially across users, and examining individual models helped reveal participant-specific patterns that were not apparent from population-level summaries alone. Together, these findings highlight the importance of individualized modeling in the design and evaluation of haptic systems that aim to quantitatively account for the full spectrum of user experience.
This study investigates how variations in the intensity of spatiotemporally modulated ultrasonic mid-air haptic stimulation modulate the amplitude, latency, and phase synchronization of somatosensory evoked potentials (SEPs), as well as their effect on behavioral perception. Electroencephalography recordings were obtained while stimulating the index and the middle finger of the left hand across different intensity levels, and a two-interval forced choice (2IFC) behavioral task was conducted to assess perceptual discrimination. The data revealed two dominant SEP components, a negative peak around 275 ms (N275) and a positive peak around 450 ms (P450), both showing increased absolute amplitude with higher stimulation intensity. Wavelet phase synchronization analysis further demonstrated great phase-locking stability at higher intensities. These findings indicate that mid-air haptic intensity modulated both the amplitude and temporal coherence of cortical responses. The behavioral results of the 2IFC task were consistent with the neural data. Participants reliably discriminated between intensity levels that exhibited distinct SEP amplitudes, while theirperformance declined for intensities that exhibited similar SEP amplitudes. Together, these results suggest that SEPs provide an objective neural correlate of subjective tactile perception. The study confirms that mid-air haptic stimulation can elicit intensity-dependent cortical responses, highlighting its potential for therapeutic, diagnostic, and interactive applications (e.g., virtual and augmented reality) despite its nontraditional stimulation mechanism. To our knowledge this is the first systematic integration linking intensity-dependent cortical responses to behavioral discrimination performance in mid-air haptic stimulation.
Digital Audio Workstations (DAWs) form the basis of contemporary music, audio production, and sound-based artistic practice, yet their increasingly visual interfaces present significant accessibility barriers for blind and partially sighted (BPS) sound creators. While screen reader technologies provide partial access, they are poorly suited to the real-time, embodied demands of mixing, spatial manipulation, and continuous parameter control, where serial speech feedback can interfere with listening and constrain creative flow. This paper examines how vibrotactile interaction can support accessible, non-visual engagement with DAWs by providing real-time, embodied feedback at the point of control. Working with nine BPS sound creators using non-visual participatory design methods, we developed two haptic prototype interfaces: the Haptic Audio Parameter Control Interface (HAPCI) and the Haptic Fader Audio Meter (HFAM). The interfaces were evaluated for their capacity to convey audio peak-level metering, spatial positioning, and continuous parameter changes through vibrotactile tactons and textures. Reflexive thematic analysis revealed that vibrotactile feedback enabled forms of real-time spatial and parametric awareness that are difficult to achieve using speech-based accessibility tools alone, distributed attentional demand across modalities to avoid over-burdening the auditory channel, and supported more fluid creative workflows. Participants developed personalized interaction strategies and expressed strong preferences for haptic feedback integrated into familiar hardware. The findings suggest an initial design framework for mapping DAW control data to vibrotactile feedback, organized around parameter type, with implications for supporting accessible, non-visual, and embodied engagement with professional audio production for this research population.
This work presents an inverse filter method capable of controlling complex and independent haptic stimulations for up to ten fingers interacting concurrently with a display. Our approach extends the inverse filtering framework to support continuous operation, enabling real-time streaming of multitouch vibrotactile signals while accounting for finger motion. We provide a comprehensive theoretical analysis that establishes the relationship between key design parameters—such as actuator count, display mechanical properties, and finger configuration—and system performance metrics, including vibration amplitude, spatial resolution, stability, and waveform reconstruction error. This analysis offers valuable engineering guidelines for the design of such systems. Our experimental setup showcases the seamless integration of this inverse filtering approach into a commercial off-the-shelf 15.6 inches OLED tactile display, equipped with 16 low-cost piezoelectric patches affixed to the rear. The resulting system can control broadband vibrations at moving positions or playback the vibrations of different musical instruments in a recording. The theoretical insights and practical implementation demonstrated in this work pave the way for advanced haptic feedback systems with broad applications in human-computer interaction and accessible technology.
Wearable haptic systems for dance are often designed around normative bodies and prescriptive mappings that treat touch as an instructional signal. We present a disability centered haptic system designed to support dance improvisation through open-ended vibrotactile interaction. The system comprises a modular, flower-shaped wearable and a real-time graphical interface, without assuming fixed placement or inter pretation. This enables dancers and choreographers to author, record, and remix affective vibrotactile sequences during practice and performance, supporting improvisation and remote participation. Design decisions were grounded in a four-session co creative process with disabled dancers, emphasizing embodied and community-centered research practices. We evaluated our system through a usability study, an artistic qualitative evaluation with two disabled dancers, and a public participatory performance, demonstrating how shared vibrotactile sensation can support Disability-centering approaches to dance creation, relational sensing, and affective co-presence.
The Alien Between Us is an interactive art installation created by multidisciplinary artist Laura A Dima. It explores how technology can facilitate intimate co-located interactions beyond traditional (audio-visual) communication systems. This art project uses haptic technologies and biofeedback to simultaneously connect two audience members. Two identical sculptures serve as interfaces that simulate biometric data, such as heart and breathing rates, to create emotional feedback loops between participants. The project investigates how mediated social touch (MST) and physiological mirroring can attune participants to each other's affective states. The installation offers a somatic language of mutual care to be incorporated alongside consent protocols. It challenges the disembodied nature of digital culture by emphasising bodily awareness and interconnectedness. The sculptures are not merely passive haptic devices; rather, they play an active role in the interaction. Their agency is expressed through a self-defence mechanism that deploys transcutaneous electrical nerve stimulation (TENS) when boundaries are breached, signalling discomfort. In this way, the technological bodies actively participate and assist human participants in navigating the ethics of touch. This paper documents the conceptual framework, technical design, and public reception of the work using a research-through-design approach. It contributes to discussions on digital intimacy, embodied interaction, and the role of artworks as haptic interfaces in cultivating empathy towards the human and non-human agents alike.
This work addresses the lack of standardized, com parable psychophysical evaluations for shape-changing haptic interfaces (SCHIs). Given the complexity and diversity of dynamic shape-change stimuli, we developed three high-resolution, stimulus-targeting test devices that predominantly evoke one of the following three: translational or rotational surface misalignment (both cutaneous perception), or expansive motions (kinesthetic perception). Psychophysical experiments measured their Point of Subjective Equality (PSE), Just-Noticeable Difference (JND), and Weber Fraction (WF) across various transformation magnitudes, directions, and grasp types. These quantitative results were supported by nonparametric bootstrapping, providing robust statistical estimates, and enriched with qualitative user feedback collected through questionnaires. The results show that larger transformations consistently yielded significantly lower WFsfor all three devices, indicating improved relative perceptual sensitivity at larger magnitudes. This suggests non-constant WF behavior over the tested range, rather than the approximately constant WF across stimulus levels expected under Weber's law. Several device- and interaction-level factors may plausibly contribute to this, including changes in contact geometry and greater joint involvement at larger shape transformations. Furthermore, no directional bias was observed for either surface misalignment device; however, a non-significant preference for expanding motion emerged in the expansion device. Together, these findings provide an applied psychophysical characterization of the present device implementations, offer study-grounded design parameters for devices leveraging the three investigated stimulus types, and motivate extending similar investigations to additional shape-change-evoked stimuli.
Temporal judgments vary across sensory modalities due to differences in information processing. Auditory intervals, for instance, are typically judged to be longer than visual intervals of the same duration. However, comparisons between tactile and visual duration judgments are scarce, particularly in VR where unique characteristics of the environment can affect the temporal judgements. Here, we compared visual and tactile duration judgments in a virtual reality (VR) environment. Participants performed a cross-modal comparison task, judging whether tactile stimuli delivered to their hands were longer than visual stimuli presented on their matched hand representations in VR. We found that tactile stimuli were consistently judged to be longer than visual stimuli of the same duration. Furthermore, the behavioral data conformed well to a model from the Scalar Expectancy Theory (SET) of timing- with higher variability in visual as compared to tactile judgments. Our findings suggest that duration judgements in VR may depend on factors such as saliency contrast and unique conditions of VR. These results can also help in the design of VR applications.
Wearable haptic devices increasingly use vibrotactile feedback for user movement guidance, yet fundamental questions remain about how cue sequences impact induced user motion. This study systematically evaluates how cue sequencing in a wrist-worn vibrotactile device induces wrist rotation from a static starting position. Sixteen participants responded to ten different cue sequences (five clockwise/counterclockwise pairs, presented at random) while manipulating a haptic device in a controlled virtual environment. Results reveal three critical findings: first, inducing wrist rotation requires at least one vibrotactile stimulus on the opposite side of the wrist from where the sequence begins; second, overlapping tactor actuation intended to create smoother sensations counter-intuitively increases confusion and produces more lateral translation than sequential actuation; and third, clockwise and counterclockwise cue sequences elicit symmetrical user responses. Statistical analysis using Friedman tests and posthoc Wilcoxon signed-rank tests confirmed significant differences between cue sequences in both rotational and translational movement (p < 0.05). These findings provide evidence-based guidelines for designing vibrotactile cue sequences in wearable haptic systems where reliably inducing rotational motion in users can reduce cognitive load and improve task performance.
As haptic interfaces integrate more seamlessly into wearables and everyday environments, they increasingly require actuators that are soft, thin, silent, and energy efficient. However, conventional motors and temperature-responsive polymers often struggle to deliver these properties due to their bulky form factors and high power consumption. High-Voltage Electrostatic Actuators (HVEAs), which generate force by applying an electric field to localized charge concentrations using high voltages and ultra-low currents, have recently emerged as a compelling alternative due to their fast, silent, and low-power operation within highly customizable and compliant form factors. This paper presents a focused review of HVEAs for haptics, examining four major classes: electrostatic switchable adhesives, dielectric elastomer actuators, soft electrohydraulic actuators, and electrokinetic pumps. For each class, we describe their mechanisms that enable haptic output; characterize their band widths, force densities, and spatial scalability; and evaluate their versatility for rendering cutaneous and kinesthetic feedback across wearable and world-grounded interfaces. Through this cross-technology analysis, we identify common design constraints and emerging strategies for improving ergonomics, streamlining fabrication, and integrating self-sensing. We conclude by out lining where HVEAs are uniquely positioned to advance haptic interaction and highlighting key research directions needed to translate these technologies into practical systems.
Haptic user interfaces that provide both kinesthetic and vibrotactile feedback have strong potential to enhance a wide range of medical training simulations. Yet, designing such devices remains challenging due to constraints related to integration, workspace, cost, and the need for diverse feedback types. To address these issues, we investigated the applicability of a novel type of brake based on the active lubrication principle. This actuator is compact, low-cost, and able to deliver both kinesthetic and vibrotactile feedback. To demonstrate its advantages, we focused on maxillofacial drilling in surgical training. We designed, implemented and evaluated a one-degree-of-freedom drill mock-up embedded into a tabletop prototype able to render axial forces up to 4.5 N. To assess both the relevance of combined feedback and the suitability of this actuator for surgical simulation, we conducted two studies: one with fourteen non-surgeon participants and another with six maxillofacial surgeons. Results confirm the potential of this actuation approach for surgery simulators and highlighted the importance of combined kinesthetic and vibrotactile feedback to improve realism, user preference, and task performance. Based on the surgeons' feedback, we then proposed a redesigned brake, with an improved control method, implemented in a new hand-held version of a drill replica. Preliminary characterisation shows that the new prototype can render forces up to 5.5 N and produce higher vibration amplitudes than the initial design.
This paper proposes and evaluates a texture mixing system for user-driven texture creation in virtual environments. The system is based on our previously proposed two-layer regression model. The first layer serves as a texture model that maps user actions to vibration features, whereas the second layer estimates the first-layer model parameters from texture features. In this study, we improved the dataset and training strategy to enhance extrapolative performance and implemented the resulting model as an interactive texture mixing system. Objective spectral similarity and subjective similarity evaluations revealed that the reproduction performance of the proposed model was comparable to that of a data-driven reference model for approximately 60% of the textures. We also showed that, under simplified single-peak conditions, nonexpert users could adjust texture-feature sliders to approach target textures within approximately one minute and with reasonably high accuracy.
When improving an upper-limb prosthetic system, choosing whether to prioritise visual design or haptic performance becomes crucial and offers valuable insights for advancing prosthesis development. This paper investigates the influence of multisensory feedback on prosthetic embodiment, focusing on key components of embodiment, including ownership, multisensory, and agency, through two Virtual Reality (VR) experiments. Our first study involved 24 participants without limb loss in a target-reaching task using different visual renderings (glove, realistic, prosthetic hands) and haptic feedback modalities (none, vibrotactile, pressure, combined). Results showed that visual appearance was a powerful determinant across all components. Realistic hands produced the highest embodiment scores, whereas the prosthetic hand representation significantly degraded them. Haptic feedback substantially affected ownership and multisensory scores; pressure feedback improved ownership compared to no feedback, and all haptic conditions elevated multisensory scores. A subsequent experiment with 12 new participants explored the effect of haptic feedback location (wrist, forearm, upper arm, contralateral forearm). This experiment found no significant performance differences among locations, though co-located feedback was preferred, where the visual feedback in VR appeared at the same spatial position as the perceived tactile contact. Finally, a pilot study with a prosthetic user provided preliminary support for the relevance of realistic visual appearance and multimodal haptic feedback. Overall, results suggest that visual realism and haptic feedback support different dimensions of prosthetic embodiment, with visual realism mainly enhancing ownership and haptic feedback strengthening multisensory. Importantly, concurrent vibrotactile-pressure feedback emerged as a promising and well-accepted solution, providing richer interaction cues without compromising embodiment.
This paper examines emergent haptic phenomena in distributed mechanically-actuated sound installations through a structural acoustic analysis of Babbling Brook, a large-scale installation comprising forty solenoid-actuated computer keyboards (eighty solenoids in total) and an eight-channel audio keystroke layer. Most haptics research in the arts has focused on direct-contact interfaces and wearable devices designed for individual users. Environmental phenomena produced from haptics have received comparatively little attention, in part because they are difficult to measure directly in walk-through gallery settings. The installation is organized around four discrete density modes that span a perceptual continuum: at low densities listeners perceive discrete mechanical keystrokes, while at high densities the acoustic texture approaches a continuous, broadband noise-like regime reminiscent of rain and water streams. Because direct vibrometric and perceptual measurement were not possible in the gallery setting, we analyze the installation's audio as an acoustic proxy for the underlying haptic transformation, using DSP metrics associated with signal stationarity and auditory-texture perception (envelope smoothness variance, spectral flux, crest factor, waveform kurtosis, spectral spread, and low-frequency modulation). Three vintage computer keyboards with distinct switch mechanisms (rubber dome, Topre capacitive, and Alps clone) are analyzed to examine how mechanical profile shapes the resulting texture, and the same pipeline is applied to three natural water recordings as an external reference. Across modes, four of seven metrics differ with large effect sizes (Kruskal-Wallis $H > 8.6$, $p < 0.035$, $\eta ^{2} > 0.70$). Keyboard-specific profiles that are distinct at sparse densities converge at high densities as waveform statistics approach Gaussian values. Mode IV swarm statistics overlap the natural-water reference on five of seven metrics and are smoother than water on envelope and spectral-variability measures, indicating that the swarm enters the broadband stationary regime of naturalistic water sounds, but lacks water's structured macroscopic fluctuations. Together, these results provide quantitative acoustic evidence supportive of an emergent perceptual transformation from a distributed systems approach to environmental haptics in the arts. Direct vibrotactile measurement and perceptual validation are identified as future work.
