Research On The Influence Of User Personality And Robot Reliability On User Trust

With the development of cutting-edge technologies such as artificial intelligence,big data,cloud computing and the Internet of Things,robots and their industrial supporting facilities have gradually improved.As an important aspect of human-robot interaction design and evaluation,user trust plays an increasingly important role in the industry and academia.At present,the research on human-robot interaction trust mainly focuses on the trust mechanism and subjective evaluation of trust in the process of interaction,but there are few researches on the physiological measurement of user interaction trust and the influence of user personality and robot reliability on interaction trust.This study uses multi-modal data collection methods to screen the effective eye movement indicators and functional near-infrared spectroscopy(fNIRS)indicators for trust measurement in the process of human-robot interaction,and combine the subjective evaluation of trust to evaluate the process of human-robot interaction The impact of user personality and robot reliability on user trust is explored.The main research contents are as follows:(1)Through extensive reading of relevant literatures of this study,the design characteristics and functional characteristics of Alpha robot were analyzed and summarized,and the robot interactive action prototype and experimental interactive operation scheme were designed.Subjective personality questionnaire was used to investigate the personality types of users,and the target users were screened and grouped.(2)An experiment was conducted on the influence of user personality and robot reliability on user trust.The user completes the experimental task according to the experimental process,collects the eye movement data,fNIRS data and subjective trust evaluation data of the user’s interaction with the robot,initially selects the physiological measurement indicators and preprocesses the data.The eye movement indicators pupil diameter maximum,average pupil diameter,fixation count,revisits and scanpath length per second.The brain area covered by fNIRS is the medial prefrontal cortex and the right temporal lobe layer.(3)Correlation analysis between the results of different trust measurement methods.Using statistical analysis,the relationship between the physiological measurement data collected in the interactive experiment and the user’s subjective trust evaluation data is analyzed to determine the eye movement trust index and the fNIRS trust index.The determined eye movement trust indicators are pupil diameter maximum,fixation count,revisits and scanpath length per second.The trust brain areas of fNIRS are the medial prefrontal cortex and the right temporal lobe.(4)Subjective and physiological measurement trust data processing and statistical testing.The experimental data obtained by multi-modal measurement is analyzed according to the dimensions of main effects and interaction effects,and combined with simple effect analysis,corresponding interpretations are given to the data results,and the influence of user personality and robot reliability on user trust is explored.Under different user personality levels,the trust of open users is higher than that of closed users.The more closed the personality,the larger the pupil diameter of the user,the more fixation points and the number of backtracking,and the higher the degree of brain activation.Under different robot reliability levels,users who interact with highreliability robots have higher trust than users who interact with low-reliability robots.Users interacting with low-reliability robots have a larger average pupil diameter,a higher saccade rate,a correspondingly higher number of fixation points and backtracking times,and a higher degree of brain activation.There is an interactive effect between the user’s personality and the reliability of the robot,which together act on the user’s interactive trust.Under the interaction,users showed different visual behaviors and activation levels of the medial prefrontal lobe and the right temporal lobe.This paper provides a theoretical basis for the adjustment of user trust level by robot reliability and user personality differences,and fills the theoretical gap in user trust research and application in the process of human-robot interaction.At the same time,it is confirmed that eye tracking and functional near-infrared can be used as effective measurement tools for human-robot interaction trust,providing technical support for future trust calibration research.