[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82110-en":3,"doc-seo-82110-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},82110,1099514067415,"Rowan","https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502",8,"Research & Report","Micro-level AI Feedback Features and Student Responses in Consecutive LLM Tutoring Interactions","AI-assisted feedback research links micro-level features—such as concrete elaboration, affective language, and response length—to learning outcomes. This study examines how feedback in one user–AI interaction relates to student confusion and understanding in the next interaction within a naturalistic LLM tutoring setting. Using 16,851 StudyChat conversations and chi-square and GEE models, results show concrete elaboration increases understanding and reduces re-confusion, affective language shows no significant effect, and longer responses correlate with lower understanding.","Micro-level AI Feedback Features and Student Responses in Consecutive LLM Tutoring Interactions  \nShayla Sharmin, Mohammad Fahim Abrar, Roghayeh Leila Barmaki  \nUniversity of Delaware, USA  \nAI-assisted feedback research has shown that micro-level feedback features, such as concrete elaboration, affective language, and response length, are associated with learning outcomes. Existing studies have primarily examined these features using session-or task-level measures. We examine how feedback provided in one user-AI interaction is associated with student confusion and understanding in the immediately following interaction in a naturalistic tutoring setting. We focus on three micro-level features of AI feedback: concrete elaboration (analogies, comparison-based explanations, or worked examples), affective language (encouragement, empathy, or apology), and response length. We analysed 16,851 conversational user-AI interactions from the StudyChat dataset, a naturalistic record of student interactions with an LLM tutor in an undergraduate AI course, and identified 1,718 cases in which students expressed confusion and continued to a subsequent interaction. Using chi-square tests and Generalized Estimating Equations, we found that concrete elaboration was associated with higher understanding and lower re-confusion in the student’s next interaction. Empathetic language showed no significant association with either outcome, while longer responses were independently associated with lower understanding. These findings highlight the value of examining feedback across consecutive user-AI interactions and suggest that concrete elaboration may play an important role in supporting immediate student understanding.  \nImplications for practice or policy:  \n• AI tutoring designers should incorporate concrete elaborations, such as analogies, comparisons, and worked examples, when student confusion is detected to improve immediate understanding.  \n• Instructional designers should configure AI tutors to provide concise explanations during confused turns, as lengthy responses may reduce student understanding.  \n• Educators using AI tutoring systems should prioritize explanatory feedback over affective language when addressing student confusion in real time.  \n• Developers of AI learning tools may use confusion detection mechanisms to trigger targeted explanatory support rather than generic encouragement.  \nKeywords: AI-assisted feedback; Micro-level feedback, large language models; concrete elaboration; affective feedback; response length, StudyChat  \nIntroduction  \nAs AI-powered learning tools become increasingly common, AI-assisted feedback (AIFB) systems have emerged as a scalable way to help students recognise misunderstandings and improve their thinking (Ba, Yang, et al. , 2025) . Along with providing correct information, how it is delivered is also important. Micro-level features, such as concrete elaboration, affective feedback, and response length, represent different ways in which AI feedback may support learning. Concrete elaboration helps students connect new concepts to existing knowledge by grounding the explanation in something tangible rather than simply correcting an error. Concrete elaboration includes both comparison-based explanations, sometimes called analogies (e.g., relating a concept to a familiar idea), and worked examples (e.g., illustrating a concept with a small, runnable code snippet that shows the concept in action) . Affective language, emotionally toned language including encouragement and self-correction, serves a different purpose. It focuses on the learner’s emotional  \nexperience by recognizing frustration, uncertainty, or effort. Response length may also matter longer explanations can provide more information but can also make feedback harder to process (Shute, 2008) . Most AIFB studies evaluate outcomes at the task or session level. They show whether students learned after completing an activity. In this study, we f","cbCaii4TiWDhj8dZ","https://ap.wps.com/l/cbCaii4TiWDhj8dZ","pdf",926075,1,14,"English","en",105,"# Introduction\n## Micro-level AI feedback features\n## Consecutive user–AI interaction perspective\n# Method and data\n## StudyChat dataset\n## Confusion window concept\n# Research questions and analysis approach\n## RQ1–RQ3","[{\"question\":\"What micro-level features of AI feedback are analyzed in this study?\",\"answer\":\"The study focuses on concrete elaboration (analogies, comparisons, worked examples), affective language (encouragement, empathy, apology), and response length.\"},{\"question\":\"How is the relationship between one AI response and the next student interaction measured?\",\"answer\":\"The analysis uses “confusion windows” consisting of two consecutive user–AI interactions, coding predictors from the AI response in the first turn and outcomes from the student prompt in the immediately following turn.\"},{\"question\":\"What are the main findings about concrete elaboration, affective language, and response length?\",\"answer\":\"Concrete elaboration is associated with higher understanding and lower re-confusion in the next interaction. Empathetic/affective language shows no significant association, while longer responses are independently associated with lower understanding.\"}]",1784178259,35,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"micro-level-ai-feedback-features-and-student-responses-in-consecutive-llm-tutoring-interactions","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/micro-level-ai-feedback-features-and-student-responses-in-consecutive-llm-tutoring-interactions/82110/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What micro-level features of AI feedback are analyzed in this study?","Question",{"text":75,"@type":76},"The study focuses on concrete elaboration (analogies, comparisons, worked examples), affective language (encouragement, empathy, apology), and response length.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the relationship between one AI response and the next student interaction measured?",{"text":80,"@type":76},"The analysis uses “confusion windows” consisting of two consecutive user–AI interactions, coding predictors from the AI response in the first turn and outcomes from the student prompt in the immediately following turn.",{"name":82,"@type":73,"acceptedAnswer":83},"What are the main findings about concrete elaboration, affective language, and response length?",{"text":84,"@type":76},"Concrete elaboration is associated with higher understanding and lower re-confusion in the next interaction. Empathetic/affective language shows no significant association, while longer responses are independently associated with lower understanding.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]