Chatbot as an emergency exist: Mediated empathy for resilience via human-AI interaction during the COVID-19 pandemic

The promises and challenges of AI-based chatbots in language education through the lens of learner emotions

The integration of AI-based chatbots in language education has garnered significant attention, yet the interplay between chatbots and positive psychology remains underexplored. Filling this gap through a critical analysis of existing theories, measurement scales, and empirical evidence, this paper evaluates the potential benefits and drawbacks of incorporating AI chatbots in language learning environments and how AI chatbots may positively or negatively impact emotional dimensions of language acquisition. The findings unravel that the primary advantages of the AI chatbots are personalized instruction with rapid feedback, a decrease in anxiety levels and a surge in motivation, greater learner independence and self-directed learning, and the fostering of metacognitive abilities. Conversely, the identified obstacles encompass restricted emotional awareness, a deficiency in genuine human interaction, ethical dilemmas and privacy issues, as well as the potential reinforcement of biases and stereotypes. By highlighting the importance of learner emotions in the language learning process, this conceptual analysis review underscores the need for a nuanced understanding of how AI chatbots can support or hinder emotional engagement and motivation. The paper discusses the impacting factors of AI-based chatbots in language education, and strategies for addressing challenges and optimizing chatbot-learner interactions, such as incorporating affective computing techniques and designing culturally-sensitive chatbots. Finally, the article outlines future research directions, emphasizing the need for validated emotion scales in chatbot assisted language learning contexts, longitudinal studies, mixed-methods research, comparative analyses, and investigations into the role of chatbots in fostering emotional intelligence and intercultural competence.

A trial of a chat service for patients and their family members in an emergency department

BACKGROUND

Effective communication between patients and healthcare providers in the emergency department (ED) is challenging due to the dynamic nature of the ED environment. This study aimed to trial a chat service enabling patients in the ED and their family members to ask questions freely, exploring the service's feasibility and user experience.

OBJECTIVES

To identify the types of needs and inquiries from patients and family members in the ED that could be addressed through the chat service and to assess the user experience of the service.

METHODS

We enrolled patients and family members aged over 19 years in the ED, providing the chat service for up to 4 h per ED visit. Trained research nurses followed specific guidelines to respond to messages from the participants. After participation, participants were required to complete a survey. Those who agreed also participated in interviews to provide insights on their experiences with the ED chat service.

RESULTS

A total of 40 participants (20 patients and 20 family members) sent 305 messages (72 by patients and 233 by family members), with patients sending an average of 3.6 messages and family members 11.7. Research nurses resolved 41.4% of patient inquiries and 70.9% of family member inquiries without further healthcare provider involvement. High usability was reported, with positive feedback on communication with healthcare workers, information accessibility, and emotional support.

CONCLUSIONS

The ED chat service was found to be feasible and led to positive user experiences for both patients and their family members.

Empathic Conversational Agent Platform Designs and Their Evaluation in the Context of Mental Health: Systematic Review

BACKGROUND

The demand for mental health (MH) services in the community continues to exceed supply. At the same time, technological developments make the use of artificial intelligence-empowered conversational agents (CAs) a real possibility to help fill this gap.

OBJECTIVE

The objective of this review was to identify existing empathic CA design architectures within the MH care sector and to assess their technical performance in detecting and responding to user emotions in terms of classification accuracy. In addition, the approaches used to evaluate empathic CAs within the MH care sector in terms of their acceptability to users were considered. Finally, this review aimed to identify limitations and future directions for empathic CAs in MH care.

METHODS

A systematic literature search was conducted across 6 academic databases to identify journal articles and conference proceedings using search terms covering 3 topics: "conversational agents," "mental health," and "empathy." Only studies discussing CA interventions for the MH care domain were eligible for this review, with both textual and vocal characteristics considered as possible data inputs. Quality was assessed using appropriate risk of bias and quality tools.

RESULTS

A total of 19 articles met all inclusion criteria. Most (12/19, 63%) of these empathic CA designs in MH care were machine learning (ML) based, with 26% (5/19) hybrid engines and 11% (2/19) rule-based systems. Among the ML-based CAs, 47% (9/19) used neural networks, with transformer-based architectures being well represented (7/19, 37%). The remaining 16% (3/19) of the ML models were unspecified. Technical assessments of these CAs focused on response accuracies and their ability to recognize, predict, and classify user emotions. While single-engine CAs demonstrated good accuracy, the hybrid engines achieved higher accuracy and provided more nuanced responses. Of the 19 studies, human evaluations were conducted in 16 (84%), with only 5 (26%) focusing directly on the CA's empathic features. All these papers used self-reports for measuring empathy, including single or multiple (scale) ratings or qualitative feedback from in-depth interviews. Only 1 (5%) paper included evaluations by both CA users and experts, adding more value to the process.

CONCLUSIONS

The integration of CA design and its evaluation is crucial to produce empathic CAs. Future studies should focus on using a clear definition of empathy and standardized scales for empathy measurement, ideally including expert assessment. In addition, the diversity in measures used for technical assessment and evaluation poses a challenge for comparing CA performances, which future research should also address. However, CAs with good technical and empathic performance are already available to users of MH care services, showing promise for new applications, such as helpline services.

The acceptability and effectiveness of artificial intelligence-based chatbot for hypertensive patients in community: protocol for a mixed-methods study

BACKGROUND

Chatbots can provide immediate assistance tailored to patients' needs, making them suitable for sustained accompanying interventions. Nevertheless, there is currently no evidence regarding their acceptability by hypertensive patients and the factors influencing the acceptability in the real-world. Existing evaluation scales often focus solely on the technology itself, overlooking the patients' perspective. Utilizing mixed methods can offer a more comprehensive exploration of influencing factors, laying the groundwork for the future integration of artificial intelligence in chronic disease management practices.

METHODS

The mixed methods will provide a holistic view to understand the effectiveness and acceptability of the intervention. Participants will either receive the standard primary health care or obtain a chatbot speaker. The speaker can provide timely reminders, on-demand consultations, personalized data recording, knowledge broadcasts, as well as entertainment features such as telling jokes. The quantitative part will be conducted as a quasi-randomized controlled trial in community in Beijing. And the convergent design will be adopted. When patients use the speaker for 1 month, scales will be used to measure patients' intention to use the speaker. At the same time, semi-structured interviews will be conducted to explore patients' feelings and influencing factors of using speakers. Data on socio-demography, physical examination, blood pressure, acceptability and self-management behavior will be collected at baseline, as well as 1,3,6, and 12 months later. Furthermore, the cloud database will continuously collect patients' interactions with the speaker. The primary outcome is the efficacy of the chatbot on blood pressure control. The secondary outcome includes the acceptability of the chatbot speaker and the changes of self-management behavior.

DISCUSSION

Artificial intelligence-based chatbot speaker not only caters to patients' self-management needs at home but also effectively organizes intricate and detailed knowledge system for patients with hypertension through a knowledge graph. Patients can promptly access information that aligns with their specific requirements, promoting proactive self-management and playing a crucial role in disease management. This study will serve as a foundation for the application of artificial intelligence technology in chronic disease management, paving the way for further exploration on enhancing the communicative impact of artificial intelligence technology.

TRIAL REGISTRATION

Biomedical Ethics Committee of Peking University: IRB00001052-21106, 2021/10/14; Clinical Trials: ChiCTR2100050578, 2021/08/29.

Digital clinical empathy in a live chat: multiple findings from a formative qualitative study and usability tests

BACKGROUND

Clinical empathy is considered a crucial element in patient-centered care. The advent of digital technology in healthcare has introduced new dynamics to empathy which needs to be explored in the context of the technology, particularly within the context of written live chats. Given the growing prevalence of written live chats, this study aimed to explore and evaluate techniques of digital clinical empathy within a familial cancer-focused live chat, focusing on how health professionals can (a) understand, (b) communicate, and (c) act upon users' perspectives and emotional states.

METHODS

The study utilized a qualitative approach in two research phases. It examined the expected and implemented techniques and effectiveness of digital clinical empathy in a live chat service, involving semi-structured interviews with health professionals (n = 9), focus group discussions with potential users (n = 42), and two rounds of usability tests between health professionals (n = 9) and users (n = 18). Data were examined using qualitative content analysis.

RESULTS

Expected techniques of digital clinical empathy, as articulated by both users and health professionals, involve reciprocal engagement, timely responses, genuine authenticity, and a balance between professionalism and informality, all while going beyond immediate queries to facilitate informed decision-making. Usability tests confirm these complexities and introduce new challenges, such as balancing timely, authentic responses with effective, personalized information management and carefully framed referrals.

CONCLUSIONS

The study reveals that the digital realm adds layers of complexity to the practice of clinical empathy. It underscores the importance of ongoing adaptation and suggests that future developments could benefit from a hybrid model that integrates the strengths of both AI and human health professionals to meet evolving user needs and maintain high-quality, empathetic healthcare interactions.

An Integrative Survey on Mental Health Conversational Agents to Bridge Computer Science and Medical Perspectives

Mental health conversational agents (a.k.a. chatbots) are widely studied for their potential to offer accessible support to those experiencing mental health challenges. Previous surveys on the topic primarily consider papers published in either computer science or medicine, leading to a divide in understanding and hindering the sharing of beneficial knowledge between both domains. To bridge this gap, we conduct a comprehensive literature review using the PRISMA framework, reviewing 534 papers published in both computer science and medicine. Our systematic review reveals 136 key papers on building mental health-related conversational agents with diverse characteristics of modeling and experimental design techniques. We find that computer science papers focus on LLM techniques and evaluating response quality using automated metrics with little attention to the application while medical papers use rule-based conversational agents and outcome metrics to measure the health outcomes of participants. Based on our findings on transparency, ethics, and cultural heterogeneity in this review, we provide a few recommendations to help bridge the disciplinary divide and enable the cross-disciplinary development of mental health conversational agents.

Chatbot for communicating with university students in emergency situation

Chatbots have arrived in higher education, and professors are trying to make the most of them. Typically, chatbots are used to help students learn academic subjects. In times of crisis, such as the COVID-19 pandemic, students who were not living with their families during the course, especially international students, were isolated and in critical situations. The student services offices were in constant contact with these students to solve problems, advise them and support them during their stay, within the constraints of confinement and the guidelines dictated by the country at the time. The student services offices were overwhelmed trying to help these students because, although the students' problems were very recurrent, the government guidelines changed from one day to the next. This article proposes the use of a chatbot to provide initial support to students during crisis situations, and facilitate communication between them and the university. The chatbot was tested by more than 160 students and student services staff. The findings support the use of chatbots as a potential tool to facilitate communication with students in emerging emergency situations, and encourage universities to adopt these types of smart tools to be prepared to respond quickly and efficiently to students in times of crisis.

How AI's Self-Prolongation Influences People's Perceptions of Its Autonomous Mind: The Case of U.S. Residents

The expanding integration of artificial intelligence (AI) in various aspects of society makes the infosphere around us increasingly complex. Humanity already faces many obstacles trying to have a better understanding of our own minds, but now we have to continue finding ways to make sense of the minds of AI. The issue of AI's capability to have independent thinking is of special attention. When dealing with such an unfamiliar concept, people may rely on existing human properties, such as survival desire, to make assessments. Employing information-processing-based Bayesian Mindsponge Framework (BMF) analytics on a dataset of 266 residents in the United States, we found that the more people believe that an AI agent seeks continued functioning, the more they believe in that AI agent's capability of having a mind of its own. Moreover, we also found that the above association becomes stronger if a person is more familiar with personally interacting with AI. This suggests a directional pattern of value reinforcement in perceptions of AI. As the information processing of AI becomes even more sophisticated in the future, it will be much harder to set clear boundaries about what it means to have an autonomous mind.

When Artificial Intelligence Voices Human Concerns: The Paradoxical Effects of AI Voice on Climate Risk Perception and Pro-Environmental Behavioral Intention

Artificial intelligence (AI)-enabled text-to-speech transformation has been widely employed to deliver online information in various fields. However, few studies have investigated the effect of the AI voice in environmental risk communication, especially in the field of climate change, an issue that poses a severe threat to global public health. To address this gap, the current study examines how the AI voice impacts the persuasive outcome of climate-related information and the potential mechanism that underlies this process. Based on the social and affect heuristics of voice, we propose a serial mediation model to test the effect of climate-related information delivered by different voice types (AI voice vs. human voice) in eliciting risk perception and motivating pro-environmental behavioral intention. Through an online auditory experiment (N = 397), we found the following. First, the AI voice was as effective as the human voice in eliciting risk perception and motivating pro-environmental behavioral intention. Second, compared with human voice, the AI voice yielded a listener's lower level of perceived identity oneness with the speaker, which decreased risk perception and subsequently inhibited pro-environmental behavioral intention. Third, compared with human voice, the AI voice produced a higher level of auditory fear, which increased risk perception and thereby led to stronger pro-environmental behavioral intention. The paradoxical role of the AI voice and its wise use in environmental risk communication for promoting global public health are discussed.

How users make judgements about the quality of online health information: a cross-sectional survey study

BACKGROUND

People increasingly use the Internet to seek health information. However, the overall quality of online health information remains low. This situation is exacerbated by the unprecedented "infodemic", which has had negative consequences for patients. Therefore, it is important to understand how users make judgements about health information by applying different judgement criteria.

OBJECTIVE

The objective of this study is to determine how patients apply different criteria in their judgement of the quality of online health information during the pandemic. In particular, we investigate whether there is consistency between the likelihood of using a particular judgement criterion and its perceived importance among different groups of users.

METHODS

A cross-sectional survey was conducted in one of the leading hospitals in a coastal province of China with a population of forty million. Combined-strategy sampling was used to balance the randomness and the practicality of the recruiting process. A total of 1063 patients were recruited for this study. Chi-square and Kruskal-Wallis analyses were used to analyse the survey data.

RESULTS

In general, patients make quality judgement of health information more frequently based on whether it is familiar, aesthetic, and with expertise. In comparison, they put more weights on whether health information is secure, trustworthy, and with expertise when determining its quality. Criteria that were considered more important were not always those with a higher likelihood of being used. Patients may not use particular criteria, such as familiarity, identification, and readability, more frequently than others even if they consider them to be more important than other do and vice versa. Surprisingly, patients with a primary school degree put more weight on whether health information is comprehensive than those with higher degrees do in determining its quality. However, they are less likely to use this guideline in practice.

CONCLUSIONS

To the best of our knowledge, this is the first study to investigate the consistency between the likelihood of using certain quality judgement criteria and their perceived importance among patients grouped by different demographic variables and eHealth literacy levels. The findings highlight how to improve online health information services and provide fine-grained customization of information for users.