The Psychological Structure and Influence of Interactive Naturalness
Cao Jianqin,Zhang Jingyu,Zhang Liang,Wang Xiaoyu
DOI: https://doi.org/10.3724/sp.j.1041.2023.00055
2023-01-01
Acta Psychologica Sinica
Abstract:This paper investigated the structure and function of interactive naturalness through three studies. After command-line interaction and graphic interaction, natural interactions have increased and become mainstream. In measuring interactive experience, the traditional usability standard is not enough to measure the usability and naturality of product interaction. The possible reason is that natural interaction is no longer limited to the product's features but is more concerned with the human experience ( the perceived naturalness) or natural experience. A direct, reliable measurement tool is urgently needed to evaluate the natural experience. Most studies used traditional usability dimensions such as visibility and ease of use to simply measure naturalness. But if naturalness simply equals usability, then many traits beyond product usability cannot be measured. Some studies directly measured naturalness with a single item or synonyms of naturalness, but the reliability of these measurements is not enough. This paper aimed to develop and verify a conceptually solid and quantitatively validated scale to measure the naturalness of interaction. In Study 1, a comprehensive item set related to the concept of natural interaction was established by using qualitative methods including dictionary retrieval, literature review, and expert interviews. In Study 2, participants (n = 353) were recruited to evaluate the experience of different intelligent connected vehicles. In addition to the interactive naturalness scale, several key consumer behaviors and traditional usability criteria, including drive intention, satisfaction, and usability, were also collected. In Study 3, new samples (n = 349) and more criterion-related variables ( two key consumption behaviors were added, namely perceived loyalty and recommendation intention) were used to further verify the validity and reliability of the developed measurement tool. We used SPSS 25.0 and Jamovi 1.2.27 to analyze the data. According to Study 1, the naturalness of the interactive experience scale of 9 items was developed based on the qualitative research. In Study 2, the exploratory factor analysis found that the two-dimensional model (six items for joyful fluency and three items for universal awareness) best suited the data. Confirmatory factor analysis verified the stability of the two- factor model. Correlation analysis and hierarchical regression analysis suggested that these two components had good criterion-related validity and joyful fluency played a crucial role in predicting satisfaction and drive intention. In Study 3, a new sample (n = 349) was used to validate the validity of the scale further. And the validity of the scale was further verified in Study 3. The criterion-related variables used in Study 2 also obtained the same correlation and regression results in Study 3. Two factors of interactive naturalness had a significant predictive effect on the newly included variables, namely recommendation intention and perceived loyalty. In addition, joyful fluency was more related to basic vehicle functions, while universal awareness was more related to advanced intelligent interaction functions. This study explored the structure and function of interactive naturalness. A psychometrically sound tool was obtained to measure the interactive naturalness experience of intelligent products in two dimensions: joyful fluency and universal awareness. We found that interactive naturalness is strongly linked to key experiential and post-purchase dimensions and has an additional contribution to predicting these variables that traditional usability dimensions cannot include. This finding was also supported by the difference of correlation between two factors in different vehicle functions. These results indicate that the scale developed in this study can measure the natural interaction experience of intelligent products reliably and effectively. This tool can be used in future human-computer interaction research and guidance for interface and product design.