Support from conversational agent on daily stress for user satisfaction improvement

Abstract

Daily stress is a universally experienced phenomenon in our modern lives [1]. Studies have shown that cumulative stress in daily life can result in cardiovascular disease [2] and psychological or behavioral disorders, such as depression and anxiety [3]. To regulate stress, social sharing is often used as a method. By talking to others, sharers can gain support that help them cope with their stress events [4]. However, people are not always comfortable sharing their stress events with others, such as the stigmatised issues [5]. In addition, people are not always able to find the right person to talk to. Their friends may be tired of hearing about the same issues [6]. Besides, professional help may be too expensive and require long waits [7].With the widespread use of conversational agent (CA), the CA should provide support for the users who encounter with daily stress to help them get socio-emotional benefits and therefore improve the user satisfaction [8]. At the same time, the gender of the CA should be considered as a factor which influence the user satisfaction of daily stress support.MethodsIn this study, we developed a 2 (CA genders)×3 (types of support) between-subject experiment on user satisfaction of CA with daily stress support function. The CA had two types of gender, including male and female. Three support types included control (without stress support function), stress (with stress detection and support function), and stressor (with stress and stressor detection, and provide support targeted at stressor) groups. So, there were six types of CA in this experiment.Participants (N = 76) were asked to share one personal daily stress experience with a CA, which was assigned from the six CAs randomly and provided support via the Wizard-of-Oz method (a human-operated CA).Results & DiscussionResults showed that participants felt better after sharing with the CA in stress group and stressor group, as evidenced by reduced stress intensity. The interaction experience of participants showed that the likeability, enjoyment, and user satisfaction of stress group were significantly higher than that of control group. While, the user satisfaction of stressor group had no significant difference with that of stress group. For the gender the CA, the female CA achieved higher likeability in control group and stress group. Meanwhile, the female CA achieved higher enjoyment, engagement, and user satisfaction in stressor group.According to the interview after the experiment, the participants in the stressor group proposed more requirements for the targeted advice from CA. They explained that they felt the high intelligent level of the CA. So, they were looking forward for the help from the CA for the solution of their daily stress events.ConclusionAccording to the above results, the support for daily stress from CA can improve the user satisfaction. Between different genders of CA, the female CA is more preferred in this study. What’s more, the CA should provide targeted advice for users with different stressors.Reference[1] Piazza, J. R., Charles, S. T., Sliwinski, M. J., Mogle, J., & Almeida, D. M. (2013). Affective reactivity to daily stressors and long-term risk of reporting a chronic physical health condition. Annals of Behavioral Medicine, 45(1), 110-120.[2] Schubert, C., Lambertz, M., Nelesen, R. A., Bardwell, W., Choi, J. B., & Dimsdale, J. E. (2009). Effects of stress on heart rate complexity—a comparison between short-term and chronic stress. Biological psychology, 80(3), 325-332.[3] Cohen, S., Kessler, R. C., & Gordon, L. U. (Eds.). (1997). Measuring stress: A guide for health and social scientists. Oxford University Press on Demand.[4] Rim ́e, B. (2009). Emotion elicits the social sharing of emotion: Theory and empirical review. Emotion Review, 1(1), 60–85.[5] Smart, L., & Wegner, D. M. (2000). The hidden costs of hidden stigma. In T. F. Heatherton, R. E. Kleck, M. R. Hebl, & J. G. Hull (Eds.), The social psychology of stigma (pp. 220–242). New York: Guildford Press.[6] Forest, A. L., Kille, D. R., Wood, J. V., & Holmes, J. G. (2014). Discount and disengage: How chronic negative expressivity undermines partner responsiveness to negative disclosures. Journal of Personality and Social Psychology, 107(6), 1013–1032.[7] Hunt, J., & Eisenberg, D. (2010). Mental health problems and help-seeking behavior among college students. Journal of Adolescent Health, 46(1), 3–10.[8] Pauw, L. S., Sauter, D. A., van Kleef, G. A., Lucas, G. M., Gratch, J., & Fischer, A. H. (2022). The avatar will see you now: Support from a virtual human provides socio-emotional benefits. Computers in Human Behavior, 136, 107368.