Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Feb 19, 2022
Date Accepted: Jun 12, 2022
Language Use in Conversational Agent-based Health Communication: Systematic Review
ABSTRACT
Background:
With the growing application of chatbots and conversational agents (CA) in health communication, a plethora of studies have been undertaken on various technology- and usability-oriented issues, including the design of physical appearance and gender, the accessibility, efficiency and personalization of CA-delivered interventions, the assessment of impersonal closeness, user experience, satisfaction and adherence, and the establishment of trust, and the user self-disclosure or -concealment of information. However, only few investigations focus on the language use and communication style of the chatbot-based health communication, to scrutinize the influence of language communication style on the user perception of CA usability and on the role of CAs in delivering healthcare services.
Objective:
This review aims to present the language designs and communication styles of chatbots and CA in healthcare to identify the achievements made and breakthroughs to be realized, so as to inform users and especially chatbot designers.
Methods:
This review was conducted by following the protocols of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement. We first designed the search strategy according to the research aim and then performed the keyword searches in PubMed and ProQuest databases for retrieving the related publications (n=179). Subsequently, three researchers independently screened and reviewed the publications to select the eligible studies according to the predefined selection criteria. Finally, we synthesized and analyzed the eligible articles (n=11).
Results:
Only 6 of the 11 included publications deal exclusively with the language designs and communication styles of the chatbots or CAs studied, and the remaining 5 are only partly related to this topic. The chatbots’ language communication styles in these studies can be classified into five categories: (i) use of personal pronouns; (ii) responses to health and lifestyle prompts; (iii) respectful language; (iv) empathetic verbal responses; and (v) supportive language with illustrations (symbols and images coupled with phrases). These linguistic styles effectively engaged users in health communication. Meanwhile, there is substantial room for improvement in the following aspects: (i) the same chatbot’s inconsistent responses to the same prompts; (ii) different answers from the same chatbot on different platforms; and (iii) the chatbot’s inability to present large volumes of precoded information on safety-critical health and lifestyle prompts, which were instead primarily answered by web-searched information. Given these deficiencies, natural language input is not able to provide constructive advice on safety-critical health issues currently. Improvement in these respects depends on the enhancement of chatbots’ response generation capabilities based on natural language processing techniques.
Conclusions:
This is the first systematic review of the language style in chatbots-based health communication. The results and limitations identified in the 11 included papers can give fresh insights into the design and development, popularization, and research of CA and chatbot applications, beneficial to the overall social good. This review can provide practical implications for incorporating the positive language styles into the design of health chabots and for improving effective language output into chatbots. In this way, upgraded chatbots will be more capable of handling various health problems, particularly in the context of nationwide and even worldwide public health crises.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.