Scales are commonly employed in Human–Robot Interaction (HRI) research, yet due to its multidisciplinary nature, many in this community lack direct training in psychometrics. This poses challenges for appropriate scale selection, accurate assessments of reliability and validity, and use. We provide a tutorial to empower researchers without scale development expertise to assess scale quality efficiently. We detail a guideline that provides high-level questions and examples to help the reader make confident evaluations of existing scales in HRI. The guideline is then used to evaluate the Godspeed and Robotic Social Attributes Scale (RoSAS). RoSAS is found to be adequately validated, whereas Godspeed warrants further investigation before it should be used in HRI contexts. The article concludes by offering advice on the use of custom scales and provides references for further enhancing expertise in this domain.

There are currently no psychometrically valid tools to measure the perceived danger of robots. To fill this gap, we provided a definition of perceived danger and developed and validated a 12-item bifactor scale through four studies. An exploratory factor analysis revealed four subdimensions of perceived danger: affective states, physical vulnerability, ominousness, and cognitive readiness. A confirmatory factor analysis confirmed the bifactor model. We then compared the perceived danger scale to the Godspeed perceived safety scale and found that the perceived danger scale is a better predictor of empirical data. We also validated the scale in an in-person setting and found that the perceived danger scale is sensitive to robot speed manipulations, consistent with previous empirical findings. Results across experiments suggest that the perceived danger scale is reliable, valid, and an adequate predictor of both perceived safety and perceived danger in human-robot interaction contexts.

The Robotic Social Attributes Scale (RoSAS) is widely used in human-robot interaction research to measure the social perception of robots, including warmth, competence, and discomfort. As previous researchers have found ambiguous support for the RoSAS's three-factor structure, the current study aims to evaluate the proposed structure by conducting a confirmatory factor analysis (CFA) using openly available datasets. The CFA (n = 1107) showed that the three-factor model had a poor model fit. This suggests that the RoSAS's three dimensions might better be used as separate scales instead of measuring a broad concept of social perception. When separating by stimulus type, only stimuli using words and vignettes had an acceptable model fit, indicating that the RoSAS might be more suitable for word/vignette stimuli. We recommend using the RoSAS's individual subscales as separate constructs rather than measuring social attributes in general. This approach also aligns with what most research has already adopted.

Construct measurement scales are commonly employed in HRI research. We provide a half-day tutorial (4 hours) that aims to empower researchers with the tools to find appropriate scales for their research and assess the quality of those scales confidently and efficiently. There are no prerequisites required for attendees. We aim to recruit researchers interested in using scales but who lack confidence in evaluating their development. The first part of the tutorial will teach attendees how to assess the quality of HRI scales. To accomplish this, we will review basic topics in psychometric theory and a guideline (developed by the organizers) that outlines best practices in scale development and validation. In the second part, we will apply this guideline to two frequently used HRI scales: Godspeed and RoSAS. Attendees are also encouraged to bring scales they are interested in reviewing. The third part aims to help attendees find appropriate scales for their research. To accomplish this, we will debut a new HRI scale database we have developed. This database is the first centralized online repository of HRI scales and contains over 40 of the most used and cited HRI scales covering a wide array of topics of interest such as, trust, embodiment, safety, and attitudes towards robots. We will demonstrate how to access and use the information contained within the database. Our goal for this tutorial is to promote active engagement from attendees throughout the session, ultimately striving to improve the quality and replicability of results in HRI studies.

The perception of danger in HRI settings has become increasingly important as interactions between robots and humans become more commonplace. Previously, a perceived danger scale was developed and validated. Here, we shortened this scale to create the Perceived Danger-Short Form (PD-SF) scale. Experiment 1 used pre-existing data and standard procedures to shorten the scale from 12 items to 4. Experiment 2 validated the short form in a new experiment where participants observed images of robots holding kitchen items of varying levels of danger in close proximity to a human. PD-SF was able to capture differences across the kitchen items. Results from both experiments indicate that PD-SF is a reliable and psychometrically valid measure of perceived danger in HRI contexts.

In this paper, we study the problem of gesture recognition as a method for divers to communicate with an underwater robot. Gesture is a common method of communication between divers, and yet autonomous underwater vehicles have very limited capacity to understand gesture given lighting and visibility constraints (e.g., from water turbidity and diver depth). Traditional deep learning methods are limited in this domain because of a lack of sufficient training data. We show that it is not enough to learn a gesture in a laboratory setting, because the appearance changes dramatically underwater. We show how hyperdimensional computing can solve this problem by permitting hypervectors to serve as abstract representations of gestures, yielding rapid adaptation to new environments and new gestures. We experimentally verify this approach using a novel dataset of 6 diving relevant gestures. We show that we can accurately adapt to a gesture learned in a laboratory setting to work with a gesture observed underwater. Our approach compares favorably to a ResNet-18, which performs well in laboratory conditions (91.9% accuracy), but performs poorly underwater (53.9% accuracy). Our proposed approach is capable of rapid adaptation, resulting in an accuracy of 83.8% on underwater gestures with just one additional example from each class added to the support set. Finally, we also show the ability to adapt to new gestures not present in our original training set. We use hypervectors to learn new gestures from the Sign Language MNIST dataset, providing a high level of accuracy with a limited amount of training data.