Analysis of trust in automation survey instruments using semantic network analysis

Abstract

This study analyzed existing survey instruments to provide an integrated list of keywords/constructs to measure the various perceptions of trust building in automation. While the trust between users and automated functions or systems has been an area of substantial research interest to understand the interactions between human and automation, research efforts to measure the trust to date have led to inconclusive and mixed outcomes. Of the existing scales for measuring trust in automation, inadequate development of constructs and the lack of reliability and validity have been identified as major causes for such outcomes. To develop a scale in a more objective and systematic approach, 86 keywords from existing 9 survey instruments were identified. The keyword network was developed based on the semantic textural similarity, and the network centrality analysis provided total 14 keywords with high centrality and degree matrics. The results can suggest some potential solutions about the lack of consistency and the wide array of constructs without adequate analytic justification in prior survey instruments. The outcomes will be utilized to develop a new integrated scale that can be generally applicable to a wide variety of automation adoption or, with slight modifications, in most trust in automation applications.