Predictors of Advice Compliance Intention in AI chatbot-based Mobile Mental Wellness Apps: The Role of Customer Engagement and Attachment

<p><strong>Over the past decade, advancements in artificial intelligence (AI) technology have given rise to AI chatbot-based mobile mental wellness apps (MMWA). These AI chatbots act as mental wellness advisors, offering mental wellness advice to improve the mental wellness of MMWA customers. Although research to date suggests AI chatbot-based MMWAs can be effective for an individual's self-management of their mental wellness, less is known about customer engagement with these AI chatbots, customers' compliance intention towards the advice they receive from the AI chatbots, the advisor characteristics of the MMWA AI chatbots, and customers' attachment to these AI chatbots. It is crucial to investigate customer engagement and customer advice compliance in the context of MMWA AI chatbots. Customer engagement is important to study because it results in positive outcomes for both service marketers (e.g., brand equity, customer loyalty, self-brand connection, and customer commitment) and customers (e.g., personalised experience, feelings of being valued, quick problem resolution, and efficient access to information). Furthermore, marketers are increasingly incorporating tools in their mobile apps that foster customer engagement due to the portability and functionality of these mobile apps. Similarly, it is important to study customer advice compliance in the context of MMWA AI chatbots since in high-contact service settings such as AI chatbot-based MMWA services, customer compliance with service provider advice/instructions has a significant influence on successful service delivery and the effectiveness of service outcomes. Examination of the AI chatbot advisor characteristics is vital because the MMWA AI chatbots are mental wellness advisors. Finally, attachment to these AI chatbots is important to investigate, as attachment is a strong determinant of customer engagement. Thus, the purpose of this research is to gain an understanding of AI chatbot advisor characteristics that predict customers' advice compliance intention towards MMWA AI chatbots and, importantly, the role that customer engagement and attachment play in this process.</strong></p><p>The research purpose is addressed by empirically testing a research model. The research draws upon stimulus-organism-response (S-O-R) theory, advice response theory (ART), attachment theory (AT), and computers-are-social-actors (CASA) theory to develop and test a research model. This study examines relationships between AI chatbot advisor characteristics (perceived likability, perceived trustworthiness, perceived expertise and perceived similarity), customer engagement with the AI chatbot, attachment to the AI chatbot, customers' personal dispositions of attachment (comfort with closeness, comfort depending on others, and attachment anxiety) and customer advice compliance intention.</p><p>This study adopts a quantitative approach to test the proposed model. A single cross-sectional survey is conducted among customers of three AI chatbot-based MMWAs (i.e., Wysa, Youper, and Replika) from the United States and the United Kingdom. The final data set for analysis includes responses from 769 participants. Partial least squares structural equation modelling (PLS-SEM) is used to analyse the measurement model and structural model. The findings reveal that the AI chatbot advisor characteristics (i.e., perceived likability, perceived trustworthiness, perceived expertise, and perceived similarity) are predictors of customer advice compliance intention towards the MMWA AI chatbots. Although not all AI chatbot advisor characteristics (i.e., perceived likability and perceived similarity) influence advice compliance intention directly, all characteristics influence advice compliance intention indirectly, serially through customers' attachment to the AI chatbot and customer engagement with the AI chatbot and/or through customer engagement with the AI chatbot. Study findings highlight the importance of customer engagement with the AI chatbot for fostering customer advice compliance intention. Additionally, customers' dependence on others, their attachment anxiety, and their connection to the AI chatbot play a crucial role in their engagement with the chatbot. Although customers' comfort depending on others and attachment anxiety do not directly influence customer engagement with the AI chatbot, they indirectly influence customer engagement with the AI chatbot through attachment to the AI chatbot. The findings of this study also reveal that customers' comfort with closeness has no influence on customers' attachment to the AI chatbot, nor on customer engagement with the AI chatbot.</p><p>This study makes several contributions. First, this research theorises how specific AI chatbot advisor characteristics directly influence advice compliance intention in the context of AI chatbot-based MMWAs. This study also importantly theorises the role of customer attachment to the AI chatbot and customer engagement with the AI chatbot in inducing higher advice compliance intention. Second, this study extends the scope of application of the advice response theory to marketing. Third, this research advances our understanding of human attachment to AI chatbot advisors. Finally, this research establishes that customers' comfort depending on others and attachment anxiety are important factors influencing customers' attachment to the AI chatbot and customer engagement with the AI chatbot in the studied context. From the practitioner's perspective, the findings of this study can be used by AI chatbot-based MMWA service marketers to program AI chatbots in such a way that the characteristics of AI chatbot advisors can drive higher levels of customer engagement and customer advice compliance intention. In addition, the results of this research can be used by AI chatbot-based MMWA service marketers to formulate strategies to enhance perceived AI chatbot advisor characteristics, customers' attachment to the AI chatbot, customers' comfort depending on others, and customers' attachment anxiety, which will boost customer engagement with the AI chatbot.</p>