banner

Haptic Experience Inventory (HXI)

The first standardized questionnaire measuring Haptic Experience (HX)


An easily deployable tool for researchers and developers to assess HX in a standardized way.

Proven to be applicable to different demographcis, across different contexts and haptic modalities.

Enable quantitative comparisons across multiple products or iterations within a product.

Provide actionable insights and support informed decisions throughout the design process.

Structure

Model Image

The HXI is built around five core factors that represent distinct aspects of the haptic experience: Autotelics, Realism, Harmony, Discord, Involvement. Each factor captures a dimension of how users perceive and evaluate haptic interactions, contributing to a more nuanced understanding of the haptic experience.

Autotelics is a distinct hedonic dimension, characterized as "the touch experience as an end in and of itself" as opposed to “instrumental”. It refers to the positive experience gained from haptic stimuli alone, without considering its functional utility as in the system. In short, it means the haptic sensation itself feels good, pleasant, or pleasurable

High-Autotelic Example: A relaxation app with a soft pneumatic vest providing gentle pressure sensations for a calming experience.
Low-Autotelic Example: A 4D movie with unpleasant jabs in the back, leading users to avoid the sensation.

Realism is the extent to which sensory stimuli could simulate or represent sensations that one would expect to encounter in reality. In contrast, sensations that feel strange, unnatural or surreal are not considered to be realistic.
It encompasses not only faithful recreation but also intentional design choices that may involve incorporating exaggerated or abstract elements to evoke believable sensations that align with users’ expectations and intentions nonetheless.

High-Realism Example: An action-adventure game where haptic feedback closely mimics the tension of a bowstring and the release of an arrow.
Low-Realism Example: A messaging app with long, low buzzes that feel strange and unnatural.

Harmony emerges as a pivotal factor shaping the overall quality of sensory encounters. Within the context of haptic feedback, Harmony refers to the seamless integration of haptic sensations with other sensory modalities, resulting in a unified and cohesive multi-sensory experience. Previous research has also explored a similar concept known as congruence, described as the intuitive match between different modalities

High-Harmony Example: A game where vibrotactile footstep feedback is synchronized with visual and sound effects, matching different terrains.

Discord represents the lack of harmony in haptic experiences. It refers to the jarring, unpleasant, or disjointed aspects of haptic feedback that detract from the overall sensory experience. Discord occurs when haptic sensations are poorly integrated with other sensory modalities or when they create a sense of disconnection or discomfort for the user. Discord is the negative counterpart to Harmony. Discord may be more immediately noticeable than Harmony, acting as a "hygiene" factor.

High-Discord Example: A racing game where the controller vibrations are out of sync with on-screen collisions, creating a confusing and disjointed experience.

Involvement refers to a psychological state characterized by heightened engagement, focus and attention, as a result of the haptic sensations. Close to what was originally termed "immersion" by Kim and Schneider, involvement primarily focuses on the level of engagement experienced by users. Involvement is closely related to immersion and often regarded as a precursor to it.

High-Involvement Example: A VR storytelling experience with subtle vibrations enhancing emotional connection with the plot.
Low-Involvement Example: A virtual museum tour with generic haptic feedback that fails to enhance engagement with the virtual environment.

Related Work

The HX model proposed by Kim and Schneider provides theoretical background for HXI. HXI was finalized after several iterations and validation studies.

Kim and Schneider defined and developed a theoritical model of HX. This model identified five experiential factors of Harmony, Expressivity, Autotelics, Immersion, and Realism. This model captured unique constructs and emphasis within haptic technology.

https://dl.acm.org/doi/abs/10.1145/3313831.3376280

Based on Kim and Schneider's model, Sathiyamurthy et al. and Anwar et al. in turn developed an item pool, and used factor analysis to try to establish a HX questionnaire. Unfortunately, no reliable and validated tool was produced, highlighting complexities involved.

https://ieeexplore.ieee.org/document/9517220

https://doi.org/10.1145/3544548.3581514

Drawing insights from previous attempts, Shi and Schneider drafted new sets of items through theoretical construction, expert reviews, and cognitive interviews, and established and validated the HXI through exploratory and confirmatory factor analysis.

https://doi.org/10.1145/3706598.3713834

User Guide

HXI is designed to be broadly applicable across different haptic modalities. While only three (vibrotactile, mid-air, force-feedback) are included in the initial validation study, the contensistency demontrated by HXI supports its diverse applicability. Nevertheless, for modalities not covered in the study, we recommend exercising caution and adhering to the complete form of the HXI and report reliability metrics, as well as other relevant measures, to further examine its suitability.

HXI is intended for administration immediately or shortly after the haptic interaction. In the initial validation study, the HXI was administered within 12 hours for online participants and immediately after the interaction for in-person participants. Prolonged recall periods may compromise the reliability of results from the HXI.

This questionnaire is structured using a 7-point Likert scale, where participants will indicate their level of agreement with each statement by selecting one of the following options: Strongly Disagree, Disagree, Somewhat Disagree, Neither Agree, Nor Disagree, Somewhat Agree, Agree, Strongly Agree.

The HXI items should be shown in shuffled randomized order when possible. We also recommend explaining what 'haptic sensations' and 'others senses' mean immediately before the HXI items. The explainings provided when validating are "Haptic sensations: Anything that you feel with the sense of touch. It could be vibrations, force, temperature, pressure, or any other physical sensation. Other senses: Sensory modalities beyond touch, such as vision and hearing."

There have been cautions against doing so and study showed psychometric properties may change when only using a subset of the questionnaire. We recommend using the complete standard form of the HXI if the situation allows.

The seven options should be scored from 1 (Strongly Disagree) to 7 (Strongly Agree). We recommend against scoring items on a scale from -3 to 3, as a score of 0 does not correspond to a meaningful condition, which may lead to interpretive inconsistencies.

To calculate a factor score, each individual's responses to the four items in each factorshould be averaged.
To calculate the overall general score: reverse the score for Discord first, then add it with other four factors scores.

Thus, General Score = Autotelics + Involvement + Realism + Harmony + ( 8 − Discord )

The HXI provides a standardized tool for evaluating haptic experiences and captures distinct important aspects of haptic interactions that shape HX.

As a quantitative measurement, HXI can enable comparisons across different variations or iterations of a product, as well as between different products.

It is important to note, though, that haptic applications have different goals and contexts, and the relative importance of each factor may vary across each product. This was also echoed in expert interviews. Depending on the context, not all factors are targeted equally. For example, a relaxation app may prioritize Autotelics, while haptic design for a game may prioritize Involvement.

We strongly encourage future studies using HXI to report Cronbach’s alpha (reliability) and how the HXI relates to other measurements (e.g., other UX questionnaires, overall ratings,sychophysiological measures like electrodermal activity andeye-tracking data).

This will help further establish the validity and reliability of the HXI, provide a more nuanced understanding of how it can be used in combination with other methods to provide a comprehensive evaluation of haptic experiences in haptic design process.

Download

The HXI is free to use for all under CC BY 4.0. We hope this instrument will be useful for researchers and developers in the field of haptic technology.

Manual Items

Share Your Work

If you are using HXI in your study, consider letting us know how you used it in this quick survey. This could benefit our understanding of application and future directions of HXI.