Instruments for Measuring Psychological Dimensions in Human-Robot Interaction: A Systematic Review of Psychometric Properties.

Background

Numerous user-related psychological dimensions can significantly influence the dynamics between humans and robots. For developers and researchers, it is crucial to have a comprehensive understanding of the psychometric properties of the available instruments used to assess these dimensions, as they indicate the reliability and validity of the assessment.

Objective

This study aims to provide a systematic review of the instruments available for assessing the psychological aspects between people and social and domestic robots, offering a summary of their psychometric properties and the quality of the evidence.

Methods

A systematic review was conducted following the PRISMA guidelines across different databases including Scopus, PubMed and IEEEXplore. The search strategy encompassed studies meeting the following inclusion criteria: a) the instrument could assess psychological dimensions related to social and domestic robots, including attitudes, beliefs, opinions, feelings, and perceptions; b) the study focused on validating the instrument; c) it evaluated the psychometric properties of the instrument; d) it underwent peer review; e) it was in English. Studies focusing on industrial robots, rescue robots, robotic arms, or those primarily concerned with technology validation or measuring anthropomorphism were excluded. Independent reviewers extracted instruments properties and methodological quality of their evidence following the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) guidelines.

Results

From 3,828 identified records, the research strategy yielded 34 articles that validated and examined the psychometric properties of 27 instruments designed to assess individuals' psychological dimensions in relation to social and domestic robots. These instruments encompass a broad spectrum of psychological dimensions. While most studies predominantly focused on structural validity and internal consistency, consideration of other psychometric properties was frequently inconsistent or absent. Despite their significance in the clinical context, no instrument has evaluated measurement error and responsiveness. Most of the instruments were targeted at both adults and older adults (18 years old and above). There is a limited number of instruments specifically designed for children, older adults, and healthcare contexts.

Conclusions

Given the strong interest in assessing psychological dimensions in the human-robot relationship, there is a need to develop new instruments using more rigorous methodologies and to consider a broader range of psychometric properties. This is essential to ensure the creation of reliable and valid measures for assessing people's psychological dimensions toward social and domestic robots. Among the limitations, the review included instruments applicable to both social robots and domestic robots, while excluding those for some specific types of robots (e.g., industrial robots).

Clinicaltrial

The review was not registered with any relevant database.