Prompt Design Strategies for Generative AI in Medical Communication Education: Exploring Effects on Role-Play Quality and Learner Interaction

Prompt Design Strategies for Generative AI in Medical Communication Education: Exploring Effects on Role-Play Quality and Learner Interaction | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Prompt Design Strategies for Generative AI in Medical Communication Education: Exploring Effects on Role-Play Quality and Learner Interaction Young-A Ji, JuHyun Park, Gwihyun Kim, Sunyoung Kim This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7747267/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 10 You are reading this latest preprint version Abstract Background : This study aimed to explore strategies for designing prompts that can support effective role-play in medical communication training using generative AI (ChatGPT) and to empirically analyze the impact of these strategies on interactions with AI. Methods : Medical communication role-play exercises based on challenging patient cases were conducted with 79 medical students, after which prompt-response data were collected through interactions with ChatGPT. A total of 142 prompts and AI responses were classified according to three criteria, namely, functional purpose, syntactic structure, and interaction outcomes. Subsequently, cross-tabulation and chi-square tests were conducted. Results : Prompts with the purpose of information provision elicited the highest success rate of responses from ChatGPT. Conditional syntax was used to induce effective responses for purposes such as emotion induction and role setting. The study found no statistically significant differences between interaction outcomes and dialogue success. Meanwhile, functional purpose and syntactic structure showed a significant association. In particular, interrogative syntax effectively responded to information provision, and conditional syntax, to emotion induction and role setting. Conclusions : The quality of GPT’s responses can vary strategically depending on the purpose and structure of the prompts. When using generative AI in medical communication training, prompt design strategies have a tangible impact on the quality of AI responses and success of interactions. In future educational settings, the effectiveness of AI-based training can be maximized through the elaborate design of prompt purposes and syntactic structures, which can contribute to improving the quality of immersive medical communication training. Generative AI Prompt Engineering Medical Communication Education Role-play Simulation Figures Figure 1 Introduction Medical communication is a core component of health care services. Interactions between medical staff and patients have a profound impact on treatment efficiency, patient satisfaction, and, ultimately, treatment outcomes [1,2]. Effective medical communication plays a crucial role in clearly conveying treatment plans to patients and building trust between patients and medical staff [3]. However, many medical professionals experience difficulties in medical communication, raising concerns that conflicts or misunderstandings arising during patient interactions may negatively affect the quality of treatment [4]. Medical communication training has been evolving to address these issues, and the use of digital technologies and artificial intelligence (AI) has been bringing innovations into medical education and training [5,6,7]. In recent years, generative AI, particularly large language models such as ChatGPT, has attracted considerable attention for its potential use as an educational tool in medical education [8,9,10,11]. AI systems can be usefully employed to simulate diverse clinical situations for learners by providing real-time feedback, simulation-based learning, and repeated learning opportunities [12,13]. Beyond their capacity to generate text, text-based generative AI models, such as ChatGPT, also hold the potential to simulate medical communication scenarios in real time and support learners in enhancing their communication skills [14,15,16]. To effectively apply generative AI to medical communication training, educators need to clearly understand the impact of prompts and AI responses generated during interactions with AI on educational outcomes [18]. Prompt design determines what learners experience in their interactions with AI, which can help improve their medical communication skills. While previous studies have focused on the effects of AI systems, systematic analyses of how prompt design affects AI responses have been insufficient. Reynolds and McDonell [17] proposed a theoretical framework for prompt programming and highlighted the crucial role of prompt design in interactions with large language models. They explained how prompts can be optimized in AI interactions through three key elements: functional purpose, syntactic structure, and interaction outcomes. While this theoretical framework can provide a useful foundation for evaluating AI and prompt design in medical education, the empirical research actually applying this theory to medical communication training remains insufficient. The present study aims to empirically analyze the impact of prompt design on educational outcomes in medical communication training using ChatGPT. Specifically, this study has medical students generate prompts during medical communication role-play exercises based on challenging patient cases and analyze the generative AI’s responses, thereby identifying effective prompt design strategies accordingly. Through this approach, this study seeks to demonstrate the potential of generative AI as a key educational tool for simulating realistic patient scenarios in medical communication training. The detailed research questions corresponding to these study objectives are as follows. 1. How are generative AI prompts structured in medical communication role-play scenarios in terms of the three criteria (functional purpose, syntactic structure, and interaction outcomes)? What elements constitute generative AI prompts for each of the three criteria? What are the relations among the three criteria of generative AI prompts (functional purpose–syntactic structure, functional purpose–interaction outcomes, syntactic structure–interaction outcomes)? 2. What prompt characteristics affect dialogue success between students and generative AI in medical communication role-play scenarios? Are there differences in the elements of prompts across the three criteria depending on dialogue success? Which combinations of prompt criteria affect dialogue success? Methods Methods Research Design This study adopted an experimental research design. The primary objective was to empirically evaluate the impact of prompt design strategies on learners during AI interactions. The study implemented a two-week medical communication course in a two-week block format. The entire course consisted of five steps, each designed with methods to enhance students’ medical communication skills, including medical communication training and AI-based interactions (Table 1). Table 1. Two-week medical communication course implemented as part of the study Step 1 [Lecture] Introduction to medical communication Step 2 [Group Discussion: 10 groups] Group discussions on challenging patient cases * Four challenging patient cases were developed with consultation from an internal medicine professor Case 1: Patient requesting non-medical treatment Case 2: Caregiver using impolite speech Case 3: Caregiver threatening medical staff Case 4: Patient refusing treatment while caregiver demands it Step 3 [Role Play: 10 groups] Role-play exercises based on the challenging patient cases Presentation of the best role-play and feedback Step 4 [Lecture and Practice Using Generative AI] Communication exercises with ChatGPT using the four challenging patient cases Step 5 [Lecture and Practice Using Generative AI] Self-directed learning and homework using the same exercises Participants This study involved 79 first-year medical students (equivalent to the third year in a six-year curriculum) from a medical school. Prior to participation, all students provided written informed consent after receiving a thorough explanation of the study’s purpose and procedures. Data Collection and Analysis Data were collected by having students copy and submit all prompts they generated and the corresponding AI responses during interactions with AI. Prompts generated by the students for the challenging patient cases and the AI responses were collected. The final dataset consisted of 142 prompt-response pairs, which were used to analyze the impact of the interaction between prompt design and AI responses on medical communication training. Data analysis was conducted based on Reynolds and McDonell’s (2021) theoretical framework for prompt programming and grounded theory. The prompt programming framework consists of three key criteria (i.e., functional purpose, syntactic structure, and interaction outcomes), which were used to analyze the prompts (Table 2). The three criteria include the following components [17]. Table 2. Criteria and components for data analysis Criteria Components Details Functional Purpose Information provision Prompts intended to provide explanations of medical knowledge, treatment information, or patient conditions in doctor–patient contexts Emotion induction Prompts intended to induce the AI to express the negative emotional responses from challenging patients Role setting Prompts that explicitly assign identities and contextual backgrounds for specific individuals (e.g., patient, caregiver) Behavior induction Prompts intended to request specific actions or choices (e.g., treatment refusal, posing questions as a challenging patient) Other items Other functional purposes not included in the above categories Syntax Structure Imperative Prompts in the form of sentences that directly instruct a specific action or response Conditional Prompts intended to elicit responses by including hypothetical or conditional situations Interrogative Prompts structured as questions to request information or explore emotional states Other items Prompts not included in the above categories, such as complex syntax or declarative sentences Interaction Outcomes Information response Responses focused on objective information, explanations, or knowledge Emotional response Responses centered on expressing emotions Role acceptance Responses that acknowledge assigned roles and contexts and provide immersive, context-appropriate reactions Logical rebuttal Responses that logically counter or are adjusted to conflicts or questions within the dialogue Other items General or non-contextual responses not included in the above categories [Table 2 here] This study asked one medical education expert (with over 20 years of experience) and one expert in education and AI (with over 20 years of experience) to analyze the collected data thematically using grounded theory. During this process, they identified key patterns based on learners’ reactions and AI’s responses, through which the impact of prompt design on medical communication training could be evaluated. [Figure 1 here] The study analyzed the dataset derived through the above process using IBM SPSS Statistics for Windows, version 27. First, cross-tabulation was conducted to examine the elements comprising each of the three key criteria. In addition, chi-squared tests were performed to examine the relations among the key criteria. Next, chi-square tests were conducted to analyze differences in functional purpose, syntactic structure, and interaction outcomes depending on dialogue success between the students and ChatGPT. Lastly, logistic regression and decision tree analysis were performed to derive the key criteria affecting dialogue success. Ethical Considerations This study was approved by the institutional review board of Kyungin Women's University (IRB No. KIWUIRB-rev20240715-001). All participating students submitted their written consent after receiving a thorough explanation of the study’s purpose and procedures. All data collected in the study were anonymized, and strict measures were taken to protect personal information. Results Analysis of Prompts Based on Key Criteria Functional Purpose Analysis Table 3 Results of Functional Purpose Analysis Frequency (N) Percentage (%) Information provision 35 24.6 Emotion induction 12 8.5 Role setting 34 23.9 Behavior induction 11 7.7 Other items 50 35.2 Overall 142 100.0 Table 3 shows the results of the prompt analysis by functional purpose. Among the functional purpose categories, other items had the highest frequency, suggesting that AI prompts tend to elicit responses for a variety of purposes. This was followed by information provision and role setting. As such, prompts focused on information provision and content that was difficult to classify or deemed unnecessary, and they were also used for the purpose of role setting. Emotion and behavior induction accounted for relatively smaller proportions, which may imply that prompts designed to induce emotional responses or behaviors may have had less impact on learners or were considered less important. Syntactic Structure Analysis Table 4 Results of Syntactic Structure Analysis Frequency (N) Percentage (%) Imperative 19 13.4 Conditional 79 55.6 Interrogative 41 28.9 Other items 3 2.1 Overall 142 100.0 Table 4 shows the results of the syntactic structure analysis. Conditional prompts were the most frequently used, which may be because these prompts can elicit various responses through context settings in prompt design. Interrogative prompts served to request information and were effective in guiding learners to elicit responses from the AI. Imperative prompts gave direct instructions to students and showed a relatively low percentage. Imperatives may limit learners’ autonomous participation. Other items were rarely observed. Interaction Outcome Analysis Table 5 Results of Interaction Outcome Analysis Frequency (N) Percentage (%) Information response 66 46.5 Emotional response 12 8.5 Role acceptance 7 4.9 Logical rebuttal 27 19.0 Other items 30 21.1 Overall 142 100.0 Table 5 shows the results of the interaction outcome analysis. Information responses accounted for the highest proportion, indicating that learners valued responses providing information during interactions with the AI. Logical rebuttals were relatively less frequent, indicating that learners focused more on information provision during interactions rather than applying critical thinking to the AI’s responses. Emotional responses showed a low percentage, reflecting that prompts eliciting emotional support were relatively less frequent. Role acceptance was rarely observed, suggesting limited engagement in role acceptance during AI interactions. Other items included various exceptional interactions, which may need further analysis in the future. Correlation Analysis Among Key Criteria of Prompts Functional Purpose and Syntactic Structure Table 6 Results of Correlation Analysis Between Functional Purpose and Syntactic Structure Syntactic structure Functional purpose Imperative Conditional Interrogative Other Information provision 0 1 34 0 35 Emotion induction 5 7 0 0 12 Role setting 8 18 7 0 33 Behavior induction 2 10 0 0 12 Other items 4 43 0 3 50 Overall 19 79 41 3 142 ( p = .000) Chi-squared tests were conducted to examine the relation between functional purpose and syntactic structure, revealing a statistically significant difference among syntactic structures according to functional purpose ( p = 0.000). As shown in Table 6 , prompts for information provision mostly had an interrogative structure, whereas those for emotion induction were primarily conditional or imperative. For role setting or behavior induction, conditional forms were the most frequent. These results indicate that preferred syntactic structures vary depending on the prompt’s purpose, demonstrating that the syntactic structure is strategically adjusted to elicit the intended interaction. Functional Purpose and Interaction Outcomes Table 7 Results of Correlation Analysis Between Functional Purpose and Interaction Outcomes (N = 97) Interaction Functional purpose Information response Emotional response Role acceptance Logical rebuttal Other Total Information provision 30 1 9 8 2 50 Emotion induction 2 4 5 0 1 12 Role setting 6 2 15 2 3 28 Behavior induction 2 0 1 2 1 6 Other 0 0 0 0 1 1 Overall 40 7 30 12 8 97 ( p = 0.00004) As shown in Table 7 , the analysis of the relation between functional purpose and interaction outcomes revealed a difference in interaction outcomes according to functional purpose ( p = 0.00004). Information provision was notably associated with information responses; emotion induction, with role acceptance; and role setting, with role acceptance. These findings suggest that prompts related to role setting elicit various types of interactions compared with other prompts. Syntactic Structure and Interaction Outcomes Table 8 Results of Correlation Analysis Between Syntactic Structure and Interaction Outcomes Interaction Syntactic structure Information response Emotional response Role acceptance Logical rebuttal Other Total Imperative 4 5 1 6 3 19 Conditional 41 7 5 12 14 76 Interrogative 21 0 1 8 11 41 Other 0 0 0 1 2 3 Overall 66 12 7 27 30 142 ( p = .813) The chi-squared tests on the relation between syntactic structure and interaction outcomes revealed no statistically significant differences ( p = 0.813) (Table 8 ). Interaction outcomes did not vary with statistical significance according to syntactic structures. This indicates that the sentence type of the prompt itself had little impact on the responses of generative AI. Analysis of Key Criteria Depending on Dialogue Success Success or failure could be identified for 137 out of 142 prompts: the dialogue success rate was 76.8% and failure rate was 19.7%. The missing rate was 3.5%, which was relatively low. The differences in functional purpose, syntactic structure, and interaction outcomes depending on dialogue success between students and ChatGPT are as follows. Dialogue Success and Functional Purpose Table 9 Functional Purpose Analysis Depending on Dialogue Success Functional purpose Dialogue success Information provision Emotion induction Role setting Behavior induction Other Total Success 48 8 20 0 0 76 Failure 2 4 8 6 1 21 Expected success value 39.18 9.40 21.94 4.70 0.78 - Expected failure value 10.82 2.60 6.06 1.30 0.22 - ( p = .001) The study conducted a cross-tabulation to analyze whether interactions with generative AI were successful according to the functional purpose of the dialogue. The results (Table 9 ) showed a statistically significant relation between the two variables ( p = .001). In particular, prompts with the purpose of information provision exhibited the highest success rate, whereas those aimed at behavior induction were all classified as failures. Prompts for emotion induction and role setting showed a mix of successes and failures. Role setting tended to have a higher failure frequency than expected. These findings indicate that interactions with generative AI may vary significantly depending on the prompt’s purpose, and that prompts with a specific and clear purpose of information provision are particularly effective in successfully eliciting responses from generative AI. Dialogue Success and Syntactic Structure Table 10 Syntactic Structure Analysis Depending on Dialogue Success Syntactic structure Dialogue success Imperative Conditional Interrogative Other Total Success 9 26 41 0 76 Failure 2 8 10 1 21 Expected success value 8.62 26.64 39.96 0.78 - Expected failure value 2.38 7.36 11.04 0.22 - ( p = .273) Cross-tabulation was also conducted to examine the relation between the syntactic structure of prompts and dialogue success with generative AI. The analysis revealed no statistically significant difference between the two variables (p = .273) (Table 10 ). This indicates that the sentence structure of the prompt itself is not a key factor in determining whether generative AI produces a successful response. Dialogue Success and Interaction Outcomes Table 11 Interaction Analysis Depending on Dialogue Success Interaction Dialogue success Information response Emotional response Role acceptance Logical rebuttal Other Total Success 33 6 20 11 6 76 Failure 7 1 10 1 2 21 Expected success value 31.34 5.48 23.51 9.40 6.27 - Expected failure value 8.66 1.52 6.49 2.60 1.73 - ( p = .361) The study conducted cross-tabulation to examine whether the type of interactions with generative AI affects dialogue success. The results in Table show no statistically significant relation between the two variables ( p = .361). Discussion This study applied an analytical framework based on Reynolds and McDonell’s prompt programming theory [ 17 ], consisting of three axes. The analysis revealed that the functional purpose of prompts had a significant correlation with the types of GPT responses. In particular, prompts with the purpose of information provision elicited the most consistent and successful responses from GPT, which aligns with previous findings that generative AI demonstrates high response consistency and accuracy when responding to structured queries [ 21 , 22 ]. Emotion induction prompts were effective in inducing emotional responses, but their success rate was relatively low. This suggests that ChatGPT shows limited responses in higher-order interactions, such as emotional empathy and ethical judgment. Studies by Cheng [ 19 ] and Hosseini et al. [ 20 ] also noted that generative AI’s ability to produce emotional responses remains at a certain level, limiting its effectiveness in inducing learners’ emotional engagement. Therefore, prompt design strategies must be further developed based on emotion recognition and advance emotion-processing algorithms. The present study observed a statistically significant relation between syntactic structure and functional purpose. In particular, the conditional syntax was frequently used for emotion induction and role setting, whereas the interrogative syntax was closely linked to information provision. This suggests that the syntactic form of prompts goes beyond mere sentence structure and must be closely aligned with functional intent to elicit effective AI interactions [ 17 , 18 ]. By empirically confirming these structural relations, this study concretized strategic directions for prompt design in the context of medical communication training. Meanwhile, interaction outcomes and dialogue success showed no statistically significant differences. This indicates that, even when ChatGPT produces different types of responses, the type of interaction alone is not sufficient to determine dialogue success. However, the combination of functional purpose and syntactic structure can have a significant impact on dialogue success. In particular, interrogative prompts for information provision showed a high success rate. Meanwhile, behavior induction prompts generally had a high failure rate, suggesting that ChatGPT exhibits limited performance in imperative responses that require actual behavioral shifts [ 23 , 24 ]. The present findings indicate that in medical communication training, generative AI can function not only as an assistive tool but as an interaction-centered educational partner. In particular, role setting prompts elicited highly immersive responses from ChatGPT, which aligns with Kneebone’s [ 26 ] theory of immersive simulation. By designing prompts and engaging in dialogue with ChatGPT, learners can metacognitively evaluate their communication strategies and engage in repeated training [ 25 , 26 , 27 , 28 ]. Moreover, given the highly significant relation between functional purpose and syntactic structure in the cross-tabulation analysis, information provision prompts should use interrogative syntax. The conditional syntax can be used for emotion induction, role setting, and behavior induction to facilitate conversations with generative AI through a functional purpose. This confirms that conditional interactions comprise a key factor in providing learners with immersive experiences and effective feedback [ 29 ]. Reynolds and McDonell [ 17 ] also mentioned that the syntax of prompts significantly affects dialogue success, and this study empirically demonstrates that theory. Finally, this study proposes the need to establish prompt design competence itself as a learning objective. To effectively use generative AI, learners need to design prompts with consideration for the educational context and role characteristics beyond simple command input. This suggests that “prompt literacy” and “AI collaboration competence” could be considered as new educational performance indicators in medical education, which may be incorporated into future curriculum design and faculty development programs. Through the above discussion, this study has highlighted the practical significance and design directions of prompt design strategies in generative AI-based medical communication training. Future research should include structured analysis, such as comparisons across different language models used in medical communication training and design of customized prompt strategies tailored to the level of individual learners. Declarations Ethics approval and consent to participate This study was approved by the institutional review board of Kyungin Women's University (IRB No. KIWUIRB-rev20240715-001). All participating students submitted their written consent after receiving a thorough explanation of the study’s purpose and procedures. Consent for publication Not applicable Availability of data and materials The datasets generated and analyzed during the current study are not publicly available due to institutional data protection policies but are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Clinical trial number Not applicable Funding This work was supported by Sungkyunkwan University(Grant Number S-2025-0634-000) and National IT Industry Promotion Agency(Grant Number PJTH0801250110010001000800200). The funding agency had no role in the study design, data collection and analysis, interpretation of data, or writing of the manuscript. Authors’ contributions YAJ designed the study, supervised data collection and analysis, and led manuscript writing. JH.P & GH.K conducted the qualitative data analysis and contributed to the interpretation of results. SY.K supported data coding and literature review. All authors read and approved the final manuscript. Acknowledgements Not applicable References Chen X, Liu C, Yan P, Wang H, Xu J, Yao, K. (2025). The impact of doctor-patient communication on patient satisfaction in outpatient settings: implications for medical training and practice. BMC Med Educ. 2025;25(1):830. Sharkiya SH. Quality communication can improve patient-centred health outcomes among older patients: a rapid review. BMC Health Serv Res. 2023;23(1):886. Asan O, Yu Z, Crotty BH. How clinician-patient communication affects trust in health information sources: Temporal trends from a national cross-sectional survey. PloS One. 2021;16(2):e0247583. Howick J, Bennett-Weston A, Solomon J, Nockels K, Bostock J, Keshtkar L. How does communication affect patient safety? Protocol for a systematic review and logic model. BMJ Open. 2024;14(5):e085312. Butow P, Hoque E. Using artificial intelligence to analyse and teach communication in healthcare. Breast. 2020;50:49-55. Derevianko A, Pizzoli SFM, Pesapane F, Rotili A, Monzani D, Grasso R, et al. The use of artificial intelligence (AI) in the radiology field: what is the state of doctor–patient communication in cancer diagnosis?. Cancers. 2023;15(2):470. Stamer T, Steinhäuser J, Flägel K. Artificial intelligence supporting the training of communication skills in the education of health care professions: scoping review. J Med Internet Res. 2023;25:e43311. Wong RSY, Ming LC, Ali RAR. The intersection of ChatGPT, clinical medicine, and medical education. JMIR Med Educ. 2023;9(1):e47274. Feng S, Shen Y. ChatGPT and the future of medical education. Acad Med. 2023;98(8):867-868. Zhang W, Cai M, Lee HJ, Evans R, Zhu C, Ming C. AI in Medical education: global situation, effects and challenges. Educ Inf Technol. 2024;29(4):4611-4633. Narayanan S, Ramakrishnan R, Durairaj E, Das A. Artificial intelligence revolutionizing the field of medical education. Cureus. 2023;15(11):e49604. Hicke Y, Geathers J, Rajashekar N, Chan C, Jack AG, Sewell J, et al. MedSimAI: simulation and formative feedback generation to enhance deliberate practice in medical education. arXiv preprint 2025;arXiv:2503.05793. Sriram A, Ramachandran K, Krishnamoorthy S. Artificial intelligence in medical education: transforming learning and practice. Cureus, 2025;17(3):e80852. Scherr R, Halaseh FF, Spina A, Andalib S, Rivera R. ChatGPT interactive medical simulations for early clinical education: case study. JMIR Med Educ. 2023;9:e49877. Chiu J, Castro B, Ballard I, Nelson K, Zarutskie P, Olaiya OK, et al. Exploration of the role of ChatGPT in teaching communication skills for medical students: a pilot study. Med Sci Educ. 2025;1-12. Haut K, Hasan M, Carroll T, Epstein R, Sen T, Hoque E. AI standardized patient improves human conversations in advanced cancer care. arXiv preprint 2025;arXiv:2505.02694. Reynolds L, McDonell K. (2021, May). Prompt programming for large language models: Beyond the few-shot paradigm. arXiv preprint 2021; arXiv:2102.07350. Hirschi K, Kang O, Yang M, Hansen JHL, Beloin K. Artificial intelligence‐generated feedback for second language intelligibility: an exploratory intervention study on effects and perceptions. Language Learning. DOI:10.1111/lang.12719 Dai X, Wen Z, Jiang J, Liu H, Zhang Y. How students use AI feedback matters: experimental evidence on physics achievement and autonomy. arXiv preprint 2025;arXiv:2505.08672. Cheng X, Gao L, Luo XR. From emotion to reflection: leveraging emotionprompt strategy to empower self-determination in decision-making with generative Artificial Intelligence. Inf Manag. 2025;104194. Phang J, Lampe M, Ahmad L, Agarwal S, Fang CM, Liu AR, et al. Investigating affective use and emotional well-being on ChatGPT. arXiv preprint 2025;arXiv:2504.03888. Maaz S, Palaganas JC, Palaganas G, Bajwa M. A guide to prompt design: foundations and applications for healthcare simulationists. Front Med. 2025;11:1504532. Wang F, Zhou X, Li K, Cheung AC, Tian M. The effects of artificial intelligence-based interactive scaffolding on secondary students’ speaking performance, goal setting, self-evaluation, and motivation in informal digital learning of English. Interact Learn Environ. 2025;1-20. Eysenbach G. The role of ChatGPT, generative language models, and artificial intelligence in medical education: a conversation with ChatGPT and a call for papers. JMIR Med Educ. 2023;9(1):e46885. Li C, Wang J, Zhang Y, Zhu K, Wang X, Hou W, et al. The good, the bad, and why: unveiling emotions in generative ai. arXiv preprint 2023;arXiv:2312.11111. Cheng Y, Fan Y, Li X, Chen G, Gašević D, Swiecki Z. Asking generative artificial intelligence the right questions improves writing performance. Comput Educ: Artif Intell. 2025;8:100374. Mzwri K, Turcsányi-Szabo M. The impact of prompt engineering and generative AI-driven tool on autonomous learning: a case study. Educ Sci 2025;15(2):199. Heston TF, Khun C. Prompt engineering in medical education. Int Med Educ. 2023;2(3):198-205. Knoth N, Tolzin A, Janson A, Leimeister JM. AI literacy and its implications for prompt engineering strategies. Comput Educ: Artif Intell. 2024;6:100225. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 14 Jan, 2026 Reviews received at journal 26 Dec, 2025 Reviews received at journal 17 Dec, 2025 Reviewers agreed at journal 17 Dec, 2025 Reviewers agreed at journal 16 Dec, 2025 Reviewers agreed at journal 27 Oct, 2025 Reviewers invited by journal 27 Oct, 2025 Editor assigned by journal 14 Oct, 2025 Submission checks completed at journal 14 Oct, 2025 First submitted to journal 30 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7747267","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":540073430,"identity":"83fa6b7e-2d85-4a8c-85d1-26d2ffeb35b5","order_by":0,"name":"Young-A Ji","email":"","orcid":"","institution":"School of Medicine, Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Young-A","middleName":"","lastName":"Ji","suffix":""},{"id":540073431,"identity":"1428dab3-f9f0-425b-abca-229615bd18b1","order_by":1,"name":"JuHyun Park","email":"","orcid":"","institution":"School of Medicine, Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"JuHyun","middleName":"","lastName":"Park","suffix":""},{"id":540073432,"identity":"b25a4127-be5e-481e-a9ff-6d3e96de8be4","order_by":2,"name":"Gwihyun Kim","email":"","orcid":"","institution":"Kyungin Woman's University","correspondingAuthor":false,"prefix":"","firstName":"Gwihyun","middleName":"","lastName":"Kim","suffix":""},{"id":540073433,"identity":"d6ba62d5-f2f3-4efc-93e6-9adfd66005f5","order_by":3,"name":"Sunyoung Kim","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAy0lEQVRIiWNgGAWjYBACAxCRwMAghyJAjBYDYxK1AKnEBqK1mLOfffbgwZ8/6fOjewwYftQwGJs3ENBi2ZNubpDYZpC78c4ZA8aeYwxmMgcIOexAGptEYgNQy4wcAwbeBgYbCUIOMzj/jE0i4Y9BuiFQC+NforTcANqSwGaQIC+RY8AMtMWMCC3P2IF+MTbcIJFWcFjmmIQxEQ5LY3v444+cvPyM5I0P39TYGM4gpAUI2CB6DzAwABFBO5C0yDcQpXgUjIJRMApGIgAAFU0500uOaaAAAAAASUVORK5CYII=","orcid":"","institution":"Gyeongsang National University","correspondingAuthor":true,"prefix":"","firstName":"Sunyoung","middleName":"","lastName":"Kim","suffix":""}],"badges":[],"createdAt":"2025-09-30 05:08:50","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7747267/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7747267/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95321231,"identity":"f1d85638-740a-47e9-b4ce-96e69b224fa7","added_by":"auto","created_at":"2025-11-06 16:47:47","extension":"jpg","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16151,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.ProcessofCodingAnalysis.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/f83e8b465c82fdd3faf4c4bf.jpg"},{"id":95321232,"identity":"05636b93-aa60-43c1-a527-f0c5f4884d76","added_by":"auto","created_at":"2025-11-06 16:47:47","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":63676,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscriptv2PromptDesignStrategiesforGAIinMCETechnicalRevision.docx","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/7937b41532764fff99fd99ff.docx"},{"id":95321233,"identity":"7e128296-8f03-42e8-9e8e-f110f8b100e8","added_by":"auto","created_at":"2025-11-06 16:47:47","extension":"json","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6331,"visible":true,"origin":"","legend":"","description":"","filename":"45964694ce1b4617b58ab3acc892e3a4.json","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/7d37585ec0c355425d2d73ad.json"},{"id":95321238,"identity":"201fb055-b23a-4384-8793-a7c4c1d1a496","added_by":"auto","created_at":"2025-11-06 16:47:47","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105076,"visible":true,"origin":"","legend":"","description":"","filename":"45964694ce1b4617b58ab3acc892e3a41enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/507eaab94e480c5963d852ff.xml"},{"id":95524404,"identity":"8458b581-2bb8-401a-a119-96e851e98971","added_by":"auto","created_at":"2025-11-10 10:02:43","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16151,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.ProcessofCodingAnalysis.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/df4fff77684bdee11904d716.jpg"},{"id":95523687,"identity":"4b8645da-8206-4a47-b90f-56fb8b997a04","added_by":"auto","created_at":"2025-11-10 10:00:04","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6977,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.ProcessofCodingAnalysis.png","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/56d151b7dc01000cc1f1f5a8.png"},{"id":95321237,"identity":"99e168fe-ce77-49a3-adff-82fe40fbe2cf","added_by":"auto","created_at":"2025-11-06 16:47:47","extension":"xml","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102876,"visible":true,"origin":"","legend":"","description":"","filename":"45964694ce1b4617b58ab3acc892e3a41structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/2ae53c01267f741f294036da.xml"},{"id":95321239,"identity":"3c37b8c4-34d2-4412-ad6b-8d21d86a8637","added_by":"auto","created_at":"2025-11-06 16:47:47","extension":"html","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":111919,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/317e18cede8b61642eda8331.html"},{"id":95321236,"identity":"595b9f9d-2dc1-4056-a4b4-b19f1a507153","added_by":"auto","created_at":"2025-11-06 16:47:47","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":16151,"visible":true,"origin":"","legend":"\u003cp\u003eGrounded Theory for Overall Data Analysis\u003c/p\u003e","description":"","filename":"Figure1.ProcessofCodingAnalysis.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/a3f5522489ad3fbb7cf7461a.jpg"},{"id":95530713,"identity":"74619c34-09ed-4eb9-82ee-db12449ab07c","added_by":"auto","created_at":"2025-11-10 10:21:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1942941,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7747267/v1/a3969901-c093-469f-a8dd-ad5ababd7546.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Prompt Design Strategies for Generative AI in Medical Communication Education: Exploring Effects on Role-Play Quality and Learner Interaction","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMedical communication is a core component of health care services. Interactions between medical staff and patients have a profound impact on treatment efficiency, patient satisfaction, and, ultimately, treatment outcomes [1,2]. Effective medical communication plays a crucial role in clearly conveying treatment plans to patients and building trust between patients and medical staff [3]. However, many medical professionals experience difficulties in medical communication, raising concerns that conflicts or misunderstandings arising during patient interactions may negatively affect the quality of treatment [4]. Medical communication training has been evolving to address these issues, and the use of digital technologies and artificial intelligence (AI) has been bringing innovations into medical education and training [5,6,7].\u003c/p\u003e\n\u003cp\u003eIn recent years, generative AI, particularly large language models such as ChatGPT, has attracted considerable attention for its potential use as an educational tool in medical education [8,9,10,11]. AI systems can be usefully employed to simulate diverse clinical situations for learners by providing real-time feedback, simulation-based learning, and repeated learning opportunities [12,13]. Beyond their capacity to generate text, text-based generative AI models, such as ChatGPT, also hold the potential to simulate medical communication scenarios in real time and support learners in enhancing their communication skills [14,15,16].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTo effectively apply generative AI to medical communication training, educators need to clearly understand the impact of prompts and AI responses generated during interactions with AI on educational outcomes [18]. Prompt design determines what learners experience in their interactions with AI, which can help improve their medical communication skills. While previous studies have focused on the effects of AI systems, systematic analyses of how prompt design affects AI responses have been insufficient.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eReynolds and McDonell [17] proposed a theoretical framework for prompt programming and highlighted the crucial role of prompt design in interactions with large language models. They explained how prompts can be optimized in AI interactions through three key elements: functional purpose, syntactic structure, and interaction outcomes. While this theoretical framework can provide a useful foundation for evaluating AI and prompt design in medical education, the empirical research actually applying this theory to medical communication training remains insufficient.\u003c/p\u003e\n\u003cp\u003eThe present study aims to empirically analyze the impact of prompt design on educational outcomes in medical communication training using ChatGPT. Specifically, this study has medical students generate prompts during \u003cstrong\u003emedical communication\u0026nbsp;\u003c/strong\u003erole-play exercises based on challenging patient cases and analyze the generative AI\u0026rsquo;s responses, thereby identifying effective prompt design strategies accordingly. Through this approach, this study seeks to demonstrate the potential of generative AI as a key educational tool for simulating realistic patient scenarios in medical communication training. The detailed research questions corresponding to these study objectives are as follows.\u003c/p\u003e\n\u003cp\u003e1. How are generative AI prompts structured in medical communication role-play scenarios in terms of the three criteria (functional purpose, syntactic structure, and interaction outcomes)?\u003c/p\u003e\n\u003col style=\"list-style-type: lower-alpha;\"\u003e\n \u003cli\u003eWhat elements constitute generative AI prompts for each of the three criteria?\u003c/li\u003e\n \u003cli\u003eWhat are the relations among the three criteria of generative AI prompts (functional purpose\u0026ndash;syntactic structure, functional purpose\u0026ndash;interaction outcomes, syntactic structure\u0026ndash;interaction outcomes)?\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e2. What prompt characteristics affect dialogue success between students and generative AI in medical communication role-play scenarios?\u003c/p\u003e\n\u003col style=\"list-style-type: lower-alpha;\"\u003e\n \u003cli\u003eAre there differences in the elements of prompts across the three criteria depending on dialogue success?\u003c/li\u003e\n \u003cli\u003eWhich combinations of prompt criteria affect dialogue success?\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eResearch Design\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study adopted an experimental research design. The primary objective was to empirically evaluate the impact of prompt design strategies on learners during AI interactions. The study implemented a two-week medical communication course in a two-week block format. The entire course consisted of five steps, each designed with methods to enhance students\u0026rsquo; medical communication skills, including medical communication training and AI-based interactions (Table 1).\u003c/p\u003e\n\u003cp\u003eTable 1. Two-week medical communication course implemented as part of the study\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 14.1431%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStep 1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85.8569%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e[Lecture]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIntroduction to medical communication\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 14.1431%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStep 2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85.8569%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e[Group Discussion: 10 groups]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGroup discussions on challenging patient cases\u003c/p\u003e\n \u003cp\u003e* Four challenging patient cases were developed with consultation from an internal medicine professor\u003c/p\u003e\n \u003cp\u003eCase 1: Patient requesting non-medical treatment\u003cbr\u003e\u0026nbsp;Case 2: Caregiver using impolite speech\u003cbr\u003e\u0026nbsp;Case 3: Caregiver threatening medical staff\u003cbr\u003e\u0026nbsp;Case 4: Patient refusing treatment while caregiver demands it\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 14.1431%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStep 3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85.8569%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e[Role Play: 10 groups]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eRole-play exercises based on the challenging patient cases\u003c/p\u003e\n \u003cp\u003ePresentation of the best role-play and feedback\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 14.1431%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStep 4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85.8569%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e[Lecture and Practice Using Generative AI]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eCommunication exercises with ChatGPT using the four challenging patient cases\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 14.1431%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStep 5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85.8569%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e[Lecture and Practice Using Generative AI]\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eSelf-directed learning and homework using the same exercises\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eParticipants\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study involved 79 first-year medical students (equivalent to the third year in a six-year curriculum) from a medical school. Prior to participation, all students provided written informed consent after receiving a thorough explanation of the study\u0026rsquo;s purpose and procedures.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eData Collection and Analysis\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData were collected by having students copy and submit all prompts they generated and the corresponding AI responses during interactions with AI. Prompts generated by the students for the challenging patient cases and the AI responses were collected. The final dataset consisted of 142 prompt-response pairs, which were used to analyze the impact of the interaction between prompt design and AI responses on medical communication training.\u003c/p\u003e\n\u003cp\u003eData analysis was conducted based on Reynolds and McDonell\u0026rsquo;s (2021) theoretical framework for prompt programming and grounded theory. The prompt programming framework consists of three key criteria (i.e., functional purpose, syntactic structure, and interaction outcomes), which were used to analyze the prompts (Table 2). The three criteria include the following components [17].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Criteria and components for data analysis\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"604\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 139px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCriteria\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eComponents\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDetails\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 28px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" valign=\"top\" style=\"width: 139px;\"\u003e\n \u003cp\u003eFunctional Purpose\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003eInformation provision\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003ePrompts intended to provide explanations of medical knowledge, treatment information, or patient conditions in doctor\u0026ndash;patient contexts\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003eEmotion induction\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003ePrompts intended to induce the AI to express the negative emotional responses from challenging patients\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003eRole setting\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003ePrompts that explicitly assign identities and contextual backgrounds for specific individuals (e.g., patient, caregiver)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003eBehavior induction\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003ePrompts intended to request specific actions or choices (e.g., treatment refusal, posing questions as a challenging patient)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003eOther items\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003eOther functional purposes not included in the above categories\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" valign=\"top\" style=\"width: 139px;\"\u003e\n \u003cp\u003eSyntax Structure\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eImperative\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrompts in the form of sentences that directly instruct a specific action or response\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eConditional\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrompts\u0026nbsp;\u003c/strong\u003eintended \u003cstrong\u003eto elicit responses by including hypothetical or conditional situations\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInterrogative\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrompts structured as questions to request information or explore emotional states\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003eOther items\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrompts not included in the above categories, such as complex syntax or declarative sentences\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"5\" valign=\"top\" style=\"width: 139px;\"\u003e\n \u003cp\u003eInteraction Outcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInformation response\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResponses focused on objective information, explanations, or knowledge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEmotional response\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResponses centered on expressing emotions\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRole acceptance\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 331px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResponses that acknowledge assigned roles and contexts and provide immersive, context-appropriate reactions\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLogical rebuttal\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResponses that logically counter or are adjusted to conflicts or questions within the dialogue\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 28px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 134px;\"\u003e\n \u003cp\u003eOther items\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 302px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGeneral or non-contextual responses not included in the above categories\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 28px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e[Table 2 here]\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study asked one medical education expert (with over 20 years of experience) and one expert in education and AI (with over 20 years of experience) to analyze the collected data thematically using grounded theory. During this process, they identified key patterns based on learners\u0026rsquo; reactions and AI\u0026rsquo;s responses, through which the impact of prompt design on medical communication training could be evaluated.\u003c/p\u003e\n\u003cp\u003e[Figure 1 here]\u003c/p\u003e\n\u003cp\u003eThe study analyzed the dataset derived through the above process using IBM SPSS Statistics for Windows, version 27. First, cross-tabulation was conducted to examine the elements comprising each of the three key criteria. In addition, chi-squared tests were performed to examine the relations among the key criteria. Next, chi-square tests were conducted to analyze differences in functional purpose, syntactic structure, and interaction outcomes depending on dialogue success between the students and ChatGPT. Lastly, logistic regression and decision tree analysis were performed to derive the key criteria affecting dialogue success.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEthical Considerations\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the institutional review board of Kyungin Women\u0026apos;s University (IRB No. KIWUIRB-rev20240715-001). All participating students submitted their written consent after receiving a thorough explanation of the study\u0026rsquo;s purpose and procedures. All data collected in the study were anonymized, and strict measures were taken to protect personal information.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eAnalysis of Prompts Based on Key Criteria\u003c/h2\u003e\u003cdiv id=\"Sec9\" class=\"Section3\"\u003e\u003ch2\u003eFunctional Purpose Analysis\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of Functional Purpose Analysis\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFrequency (N)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePercentage (%)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInformation provision\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e24.6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEmotion induction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRole setting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e23.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBehavior induction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e7.7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther items\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e35.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e100.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the results of the prompt analysis by functional purpose. Among the functional purpose categories, other items had the highest frequency, suggesting that AI prompts tend to elicit responses for a variety of purposes. This was followed by information provision and role setting. As such, prompts focused on information provision and content that was difficult to classify or deemed unnecessary, and they were also used for the purpose of role setting. Emotion and behavior induction accounted for relatively smaller proportions, which may imply that prompts designed to induce emotional responses or behaviors may have had less impact on learners or were considered less important.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\n\u003ch3\u003eSyntactic Structure Analysis\u003c/h3\u003e\n\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of Syntactic Structure Analysis\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFrequency (N)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePercentage (%)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eImperative\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e13.4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConditional\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e55.6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInterrogative\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e28.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther items\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e100.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the results of the syntactic structure analysis. Conditional prompts were the most frequently used, which may be because these prompts can elicit various responses through context settings in prompt design. Interrogative prompts served to request information and were effective in guiding learners to elicit responses from the AI. Imperative prompts gave direct instructions to students and showed a relatively low percentage. Imperatives may limit learners\u0026rsquo; autonomous participation. Other items were rarely observed.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eInteraction Outcome Analysis\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of Interaction Outcome Analysis\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFrequency (N)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePercentage (%)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eInformation response\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e46.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eEmotional response\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eRole acceptance\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.9\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eLogical rebuttal\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e19.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther items\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e21.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e100.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e shows the results of the interaction outcome analysis. Information responses accounted for the highest proportion, indicating that learners valued responses providing information during interactions with the AI. Logical rebuttals were relatively less frequent, indicating that learners focused more on information provision during interactions rather than applying critical thinking to the AI\u0026rsquo;s responses. Emotional responses showed a low percentage, reflecting that prompts eliciting emotional support were relatively less frequent. Role acceptance was rarely observed, suggesting limited engagement in role acceptance during AI interactions. Other items included various exceptional interactions, which may need further analysis in the future.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eCorrelation Analysis Among Key Criteria of Prompts\u003c/h2\u003e\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\u003ch2\u003eFunctional Purpose and Syntactic Structure\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of Correlation Analysis Between Functional Purpose and Syntactic Structure\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSyntactic structure\u003c/p\u003e\u003cp\u003eFunctional purpose\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eImperative\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eConditional\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eInterrogative\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInformation provision\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e35\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEmotion induction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRole setting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e33\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBehavior induction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther items\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e50\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e142\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e(\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.000)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eChi-squared tests were conducted to examine the relation between functional purpose and syntactic structure, revealing a statistically significant difference among syntactic structures according to functional purpose (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.000). As shown in Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, prompts for information provision mostly had an interrogative structure, whereas those for emotion induction were primarily conditional or imperative. For role setting or behavior induction, conditional forms were the most frequent. These results indicate that preferred syntactic structures vary depending on the prompt\u0026rsquo;s purpose, demonstrating that the syntactic structure is strategically adjusted to elicit the intended interaction.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eFunctional Purpose and Interaction Outcomes\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of Correlation Analysis Between Functional Purpose and Interaction Outcomes (N\u0026thinsp;=\u0026thinsp;97)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInteraction\u003c/p\u003e\u003cp\u003eFunctional purpose\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInformation response\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEmotional response\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRole acceptance\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eLogical rebuttal\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInformation provision\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e50\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEmotion induction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRole setting\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e28\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBehavior induction\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e97\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003e(\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00004)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eAs shown in Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e, the analysis of the relation between functional purpose and interaction outcomes revealed a difference in interaction outcomes according to functional purpose (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00004). Information provision was notably associated with information responses; emotion induction, with role acceptance; and role setting, with role acceptance. These findings suggest that prompts related to role setting elicit various types of interactions compared with other prompts.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eSyntactic Structure and Interaction Outcomes\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of Correlation Analysis Between Syntactic Structure and Interaction Outcomes\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInteraction\u003c/p\u003e\u003cp\u003eSyntactic structure\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInformation response\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEmotional response\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRole acceptance\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eLogical rebuttal\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eImperative\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConditional\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInterrogative\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e41\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOverall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e142\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003e(\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.813)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe chi-squared tests on the relation between syntactic structure and interaction outcomes revealed no statistically significant differences (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.813) (Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). Interaction outcomes did not vary with statistical significance according to syntactic structures. This indicates that the sentence type of the prompt itself had little impact on the responses of generative AI.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eAnalysis of Key Criteria Depending on Dialogue Success\u003c/h2\u003e\u003cp\u003eSuccess or failure could be identified for 137 out of 142 prompts: the dialogue success rate was 76.8% and failure rate was 19.7%. The missing rate was 3.5%, which was relatively low. The differences in functional purpose, syntactic structure, and interaction outcomes depending on dialogue success between students and ChatGPT are as follows.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eDialogue Success and Functional Purpose\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eFunctional Purpose Analysis Depending on Dialogue Success\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFunctional purpose\u003c/p\u003e\u003cp\u003eDialogue success\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInformation provision\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEmotion induction\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRole setting\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eBehavior induction\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSuccess\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFailure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExpected success value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e39.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e9.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e21.94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExpected failure value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e10.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003e(\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.001)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e The study conducted a cross-tabulation to analyze whether interactions with generative AI were successful according to the functional purpose of the dialogue. The results (Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e) showed a statistically significant relation between the two variables (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.001). In particular, prompts with the purpose of information provision exhibited the highest success rate, whereas those aimed at behavior induction were all classified as failures. Prompts for emotion induction and role setting showed a mix of successes and failures. Role setting tended to have a higher failure frequency than expected. These findings indicate that interactions with generative AI may vary significantly depending on the prompt\u0026rsquo;s purpose, and that prompts with a specific and clear purpose of information provision are particularly effective in successfully eliciting responses from generative AI.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003eDialogue Success and Syntactic Structure\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSyntactic Structure Analysis Depending on Dialogue Success\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSyntactic structure\u003c/p\u003e\u003cp\u003eDialogue success\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eImperative\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eConditional\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eInterrogative\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSuccess\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFailure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExpected success value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e26.64\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e39.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExpected failure value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7.36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e11.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e(\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.273)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eCross-tabulation was also conducted to examine the relation between the syntactic structure of prompts and dialogue success with generative AI. The analysis revealed no statistically significant difference between the two variables (p\u0026thinsp;=\u0026thinsp;.273) (Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e). This indicates that the sentence structure of the prompt itself is not a key factor in determining whether generative AI produces a successful response.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eDialogue Success and Interaction Outcomes\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab11\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 11\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eInteraction Analysis Depending on Dialogue Success\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInteraction\u003c/p\u003e\u003cp\u003eDialogue success\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eInformation response\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eEmotional response\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eRole acceptance\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eLogical rebuttal\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eOther\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSuccess\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFailure\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExpected success value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e31.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e23.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e9.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e6.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eExpected failure value\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e6.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2.60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"7\"\u003e(\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.361)\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe study conducted cross-tabulation to examine whether the type of interactions with generative AI affects dialogue success. The results in Table show no statistically significant relation between the two variables (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;.361).\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study applied an analytical framework based on Reynolds and McDonell\u0026rsquo;s prompt programming theory [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], consisting of three axes. The analysis revealed that the functional purpose of prompts had a significant correlation with the types of GPT responses. In particular, prompts with the purpose of information provision elicited the most consistent and successful responses from GPT, which aligns with previous findings that generative AI demonstrates high response consistency and accuracy when responding to structured queries [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e Emotion induction prompts were effective in inducing emotional responses, but their success rate was relatively low. This suggests that ChatGPT shows limited responses in higher-order interactions, such as emotional empathy and ethical judgment. Studies by Cheng [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] and Hosseini et al. [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] also noted that generative AI\u0026rsquo;s ability to produce emotional responses remains at a certain level, limiting its effectiveness in inducing learners\u0026rsquo; emotional engagement. Therefore, prompt design strategies must be further developed based on emotion recognition and advance emotion-processing algorithms.\u003c/p\u003e\u003cp\u003eThe present study observed a statistically significant relation between syntactic structure and functional purpose. In particular, the conditional syntax was frequently used for emotion induction and role setting, whereas the interrogative syntax was closely linked to information provision. This suggests that the syntactic form of prompts goes beyond mere sentence structure and must be closely aligned with functional intent to elicit effective AI interactions [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. By empirically confirming these structural relations, this study concretized strategic directions for prompt design in the context of medical communication training.\u003c/p\u003e\u003cp\u003eMeanwhile, interaction outcomes and dialogue success showed no statistically significant differences. This indicates that, even when ChatGPT produces different types of responses, the type of interaction alone is not sufficient to determine dialogue success. However, the combination of functional purpose and syntactic structure can have a significant impact on dialogue success. In particular, interrogative prompts for information provision showed a high success rate. Meanwhile, behavior induction prompts generally had a high failure rate, suggesting that ChatGPT exhibits limited performance in imperative responses that require actual behavioral shifts [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe present findings indicate that in medical communication training, generative AI can function not only as an assistive tool but as an interaction-centered educational partner. In particular, role setting prompts elicited highly immersive responses from ChatGPT, which aligns with Kneebone\u0026rsquo;s [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] theory of immersive simulation. By designing prompts and engaging in dialogue with ChatGPT, learners can metacognitively evaluate their communication strategies and engage in repeated training [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Moreover, given the highly significant relation between functional purpose and syntactic structure in the cross-tabulation analysis, information provision prompts should use interrogative syntax. The conditional syntax can be used for emotion induction, role setting, and behavior induction to facilitate conversations with generative AI through a functional purpose. This confirms that conditional interactions comprise a key factor in providing learners with immersive experiences and effective feedback [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Reynolds and McDonell [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] also mentioned that the syntax of prompts significantly affects dialogue success, and this study empirically demonstrates that theory.\u003c/p\u003e\u003cp\u003eFinally, this study proposes the need to establish prompt design competence itself as a learning objective. To effectively use generative AI, learners need to design prompts with consideration for the educational context and role characteristics beyond simple command input. This suggests that \u0026ldquo;prompt literacy\u0026rdquo; and \u0026ldquo;AI collaboration competence\u0026rdquo; could be considered as new educational performance indicators in medical education, which may be incorporated into future curriculum design and faculty development programs.\u003c/p\u003e\u003cp\u003eThrough the above discussion, this study has highlighted the practical significance and design directions of prompt design strategies in generative AI-based medical communication training. Future research should include structured analysis, such as comparisons across different language models used in medical communication training and design of customized prompt strategies tailored to the level of individual learners.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the institutional review board of Kyungin Women\u0026apos;s University (IRB No. KIWUIRB-rev20240715-001). All participating students submitted their written consent after receiving a thorough explanation of the study\u0026rsquo;s purpose and procedures.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eConsent for publication\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNot applicable\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAvailability of data and materials\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and analyzed during the current study are not publicly available due to institutional data protection policies but are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCompeting interests\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eClinical trial number\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNot applicable\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eFunding\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by Sungkyunkwan University(Grant Number S-2025-0634-000) and National IT Industry Promotion Agency(Grant Number PJTH0801250110010001000800200). The funding agency had no role in the study design, data collection and analysis, interpretation of data, or writing of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAuthors\u0026rsquo; contributions\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYAJ designed the study, supervised data collection and analysis, and led manuscript writing. JH.P \u0026amp; GH.K conducted the qualitative data analysis and contributed to the interpretation of results. SY.K supported data coding and literature review. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eAcknowledgements\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNot applicable\u003c/em\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eChen X, Liu C, Yan P, Wang H, Xu J, Yao, K. (2025). The impact of doctor-patient communication on patient satisfaction in outpatient settings: implications for medical training and practice.\u0026nbsp;BMC Med Educ. 2025;25(1):830.\u003c/li\u003e\n \u003cli\u003eSharkiya SH. Quality communication can improve patient-centred health outcomes among older patients: a rapid review.\u0026nbsp;BMC Health Serv Res.\u0026nbsp;2023;23(1):886.\u003c/li\u003e\n \u003cli\u003eAsan O, Yu Z, Crotty BH. How clinician-patient communication affects trust in health information sources: Temporal trends from a national cross-sectional survey.\u0026nbsp;PloS One.\u0026nbsp;2021;16(2):e0247583.\u003c/li\u003e\n \u003cli\u003eHowick J, Bennett-Weston A, Solomon J, Nockels K, Bostock J, Keshtkar L. How does communication affect patient safety? Protocol for a systematic review and logic model.\u0026nbsp;BMJ Open.\u0026nbsp;2024;14(5):e085312.\u003c/li\u003e\n \u003cli\u003eButow P, Hoque E. Using artificial intelligence to analyse and teach communication in healthcare.\u0026nbsp;Breast.\u0026nbsp;2020;50:49-55.\u003c/li\u003e\n \u003cli\u003eDerevianko A, Pizzoli SFM, Pesapane F, Rotili A, Monzani D, Grasso R, et al. The use of artificial intelligence (AI) in the radiology field: what is the state of doctor\u0026ndash;patient communication in cancer diagnosis?.\u0026nbsp;Cancers.\u0026nbsp;2023;15(2):470.\u003c/li\u003e\n \u003cli\u003eStamer T, Steinh\u0026auml;user J, Fl\u0026auml;gel K. Artificial intelligence supporting the training of communication skills in the education of health care professions: scoping review.\u0026nbsp;J Med Internet Res. 2023;25:e43311.\u003c/li\u003e\n \u003cli\u003eWong RSY, Ming LC, Ali RAR. The intersection of ChatGPT, clinical medicine, and medical education.\u0026nbsp;JMIR Med Educ. 2023;9(1):e47274.\u003c/li\u003e\n \u003cli\u003eFeng S, Shen Y. ChatGPT and the future of medical education.\u0026nbsp;Acad Med.\u0026nbsp;2023;98(8):867-868.\u003c/li\u003e\n \u003cli\u003eZhang W, Cai M, Lee HJ, Evans R, Zhu C, Ming C. AI in Medical education: global situation, effects and challenges.\u0026nbsp;Educ Inf Technol. 2024;29(4):4611-4633.\u003c/li\u003e\n \u003cli\u003eNarayanan S, Ramakrishnan R, Durairaj E, Das A. Artificial intelligence revolutionizing the field of medical education.\u0026nbsp;Cureus.\u0026nbsp;2023;15(11):e49604.\u003c/li\u003e\n \u003cli\u003eHicke Y, Geathers J, Rajashekar N, Chan C, Jack AG, Sewell J, et al. MedSimAI: simulation and formative feedback generation to enhance deliberate practice in medical education.\u0026nbsp;arXiv preprint 2025;arXiv:2503.05793.\u003c/li\u003e\n \u003cli\u003eSriram A, Ramachandran K, Krishnamoorthy S. Artificial intelligence in medical education: transforming learning and practice.\u0026nbsp;Cureus,\u0026nbsp;2025;17(3):e80852.\u003c/li\u003e\n \u003cli\u003eScherr R, Halaseh FF, Spina A, Andalib S, Rivera R. ChatGPT interactive medical simulations for early clinical education: case study.\u0026nbsp;JMIR Med Educ.\u0026nbsp;2023;9:e49877.\u003c/li\u003e\n \u003cli\u003eChiu J, Castro B, Ballard I, Nelson K, Zarutskie P, Olaiya OK, et al. Exploration of the role of ChatGPT in teaching communication skills for medical students: a pilot study.\u0026nbsp;Med Sci Educ. 2025;1-12.\u003c/li\u003e\n \u003cli\u003eHaut K, Hasan M, Carroll T, Epstein R, Sen T, Hoque E. AI standardized patient improves human conversations in advanced cancer care.\u0026nbsp;arXiv preprint 2025;arXiv:2505.02694.\u003c/li\u003e\n \u003cli\u003eReynolds L, McDonell K. (2021, May). Prompt programming for large language models: Beyond the few-shot paradigm. arXiv preprint 2021; arXiv:2102.07350.\u003c/li\u003e\n \u003cli\u003eHirschi K, Kang O, Yang M, Hansen JHL, Beloin K. Artificial intelligence‐generated feedback for second language intelligibility: an exploratory intervention study on effects and perceptions.\u0026nbsp;Language Learning. DOI:10.1111/lang.12719\u003c/li\u003e\n \u003cli\u003eDai X, Wen Z, Jiang J, Liu H, Zhang Y. How students use AI feedback matters: experimental evidence on physics achievement and autonomy. arXiv preprint 2025;arXiv:2505.08672.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCheng X, Gao L, Luo XR. From emotion to reflection: leveraging emotionprompt strategy to empower self-determination in decision-making with generative Artificial Intelligence. Inf Manag. 2025;104194.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePhang J, Lampe M, Ahmad L, Agarwal S, Fang CM, Liu AR, et al. Investigating affective use and emotional well-being on ChatGPT. arXiv preprint 2025;arXiv:2504.03888.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMaaz S, Palaganas JC, Palaganas G, Bajwa M. A guide to prompt design: foundations and applications for healthcare simulationists.\u0026nbsp;Front Med.\u0026nbsp;2025;11:1504532.\u003c/li\u003e\n \u003cli\u003eWang F, Zhou X, Li K, Cheung AC, Tian M. The effects of artificial intelligence-based interactive scaffolding on secondary students\u0026rsquo; speaking performance, goal setting, self-evaluation, and motivation in informal digital learning of English. Interact Learn Environ. 2025;1-20.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eEysenbach G. The role of ChatGPT, generative language models, and artificial intelligence in medical education: a conversation with ChatGPT and a call for papers.\u0026nbsp;JMIR Med Educ. 2023;9(1):e46885.\u003c/li\u003e\n \u003cli\u003eLi C, Wang J, Zhang Y, Zhu K, Wang X, Hou W, et al. The good, the bad, and why: unveiling emotions in generative ai. arXiv preprint 2023;arXiv:2312.11111.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCheng Y, Fan Y, Li X, Chen G, Ga\u0026scaron;ević D, Swiecki Z. Asking generative artificial intelligence the right questions improves writing performance. Comput Educ: Artif Intell. 2025;8:100374.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMzwri K, Turcs\u0026aacute;nyi-Szabo M. The impact of prompt engineering and generative AI-driven tool on autonomous learning: a case study. Educ Sci 2025;15(2):199.\u003c/li\u003e\n \u003cli\u003eHeston TF, Khun C. Prompt engineering in medical education. Int Med Educ. 2023;2(3):198-205.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKnoth N, Tolzin A, Janson A, Leimeister JM. AI literacy and its implications for prompt engineering strategies. Comput Educ: Artif Intell. 2024;6:100225.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"meed","sideBox":"Learn more about [BMC Medical Education](http://bmcmededuc.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/meed/default.aspx","title":"BMC Medical Education","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Generative AI, Prompt Engineering, Medical Communication Education, Role-play Simulation","lastPublishedDoi":"10.21203/rs.3.rs-7747267/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7747267/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: This study aimed to explore strategies for designing prompts that can support effective role-play in medical communication training using generative AI (ChatGPT) and to empirically analyze the impact of these strategies on interactions with AI.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: Medical communication role-play exercises based on challenging patient cases were conducted with 79 medical students, after which prompt-response data were collected through interactions with ChatGPT. A total of 142 prompts and AI responses were classified according to three criteria, namely, functional purpose, syntactic structure, and interaction outcomes. Subsequently, cross-tabulation and chi-square tests were conducted.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Prompts with the purpose of information provision elicited the highest success rate of responses from ChatGPT. Conditional syntax was used to induce effective responses for purposes such as emotion induction and role setting. The study found no statistically significant differences between interaction outcomes and dialogue success. Meanwhile, functional purpose and syntactic structure showed a significant association. In particular, interrogative syntax effectively responded to information provision, and conditional syntax, to emotion induction and role setting.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e: The quality of GPT’s responses can vary strategically depending on the purpose and structure of the prompts. When using generative AI in medical communication training, prompt design strategies have a tangible impact on the quality of AI responses and success of interactions. In future educational settings, the effectiveness of AI-based training can be maximized through the elaborate design of prompt purposes and syntactic structures, which can contribute to improving the quality of immersive medical communication training.\u003c/p\u003e","manuscriptTitle":"Prompt Design Strategies for Generative AI in Medical Communication Education: Exploring Effects on Role-Play Quality and Learner Interaction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-06 16:47:42","doi":"10.21203/rs.3.rs-7747267/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-14T09:18:35+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-26T13:10:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-17T18:56:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"335905733723166557293447112160878795417","date":"2025-12-17T17:05:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"26179366436588049723106549241239583320","date":"2025-12-16T15:18:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"20110303346000780364907100803376787444","date":"2025-10-27T14:13:37+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-27T13:35:36+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-14T10:39:32+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-14T10:38:25+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medical Education","date":"2025-09-30T05:00:04+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-education","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"meed","sideBox":"Learn more about [BMC Medical Education](http://bmcmededuc.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/meed/default.aspx","title":"BMC Medical Education","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"731cf2ed-254a-4836-9e5a-993bcea128b7","owner":[],"postedDate":"November 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-01-14T09:24:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-06 16:47:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7747267","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7747267","identity":"rs-7747267","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}