Background: Both industries and academia are paying deep attention to the concept of trust in automation since it is significant that creates people’s desire to use and buy autonomous cars. Objectives: Regarding the previously introduced trust questionnaire by our research team, this study was conducted to present more objective insights of this questionnaire to show its convergent validity based on the physiological data. Methods: Using a dynamic driving simulator, the 8-step experiment was performed on 29 min with 21 participants. The steps provided building, improvement, destruction, and re-building of trust. The experiment contained both manual and autonomous driving. Subjects completed stress and trust scales regarding the experience they performed. Finally, they answered to the trust questionnaire; their heart rate was acquired throughout the experiment. Results: The Mean ± SD of participants’ trust score was 139 ± 5.76. Self-driving car’s unsafe behaviors were accompanied by a rise in the individuals’ heart rates (objective metrics of stress). Throughout the two phases of fastest and riskiest parts of the scenario, people felt the highest stress (6 out of 10) and the least trust (4.33 out of 10), that were obtained from the mentioned one question scale. Conclusion: Regarding the results, the mean score of people’s trust from our multi-dimensional questionnaire is far from the desired score, which is 235. Objective (heart rate) and other subjective data (stress and trust) confirmed that the participants’ trust in an autonomous vehicle was low. So, it can be claimed that there is a consistency between subjective and objective data with the results of our comprehensive trust questionnaire.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Beat Trustfully: The Correlation Between Heart Rate and a Multi-dimensional Trust Questionnaire

  • Saeedeh Mosaferchi,
  • Rosaria Califano,
  • Giuseppe Pica,
  • Luca Orlando,
  • Francesco Villecco,
  • Alessandro Naddeo

摘要

Background: Both industries and academia are paying deep attention to the concept of trust in automation since it is significant that creates people’s desire to use and buy autonomous cars. Objectives: Regarding the previously introduced trust questionnaire by our research team, this study was conducted to present more objective insights of this questionnaire to show its convergent validity based on the physiological data. Methods: Using a dynamic driving simulator, the 8-step experiment was performed on 29 min with 21 participants. The steps provided building, improvement, destruction, and re-building of trust. The experiment contained both manual and autonomous driving. Subjects completed stress and trust scales regarding the experience they performed. Finally, they answered to the trust questionnaire; their heart rate was acquired throughout the experiment. Results: The Mean ± SD of participants’ trust score was 139 ± 5.76. Self-driving car’s unsafe behaviors were accompanied by a rise in the individuals’ heart rates (objective metrics of stress). Throughout the two phases of fastest and riskiest parts of the scenario, people felt the highest stress (6 out of 10) and the least trust (4.33 out of 10), that were obtained from the mentioned one question scale. Conclusion: Regarding the results, the mean score of people’s trust from our multi-dimensional questionnaire is far from the desired score, which is 235. Objective (heart rate) and other subjective data (stress and trust) confirmed that the participants’ trust in an autonomous vehicle was low. So, it can be claimed that there is a consistency between subjective and objective data with the results of our comprehensive trust questionnaire.