Exploring Shared Monitoring in Large Language Model (LLM)‐Supported Online Collaborative Problem Solving for High‐Cohesion and Low‐Cohesion Groups.

Background: Recent research increasingly highlights the central role of interventions in enhancing shared monitoring during collaborative problem‐solving. However, traditional intervention approaches suffer from limitations in timeliness and adaptability. Large language model (LLM), equipped with de...

Descripción completa

Detalles Bibliográficos
Publicado en:Journal of Computer Assisted Learning Vol. 42; no. 3; pp. 1 - 22
Autores principales: Liu, Xiaoyun, Du, Xu, Hung, Jui‐Long, Li, Hao, Yang, Shuoqiu, Xie, Yiqian
Formato: equations & formulas pictorial research tables/charts Journal Article
Publicado: Wiley-Blackwell Jun2026
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=194050893&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 194050893
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        02664909
        6M1
      jtl: Journal of Computer Assisted Learning
      issn: 02664909
      maglogo: Y
    pubinfo:
      dt: Jun2026
      vid: 42
      iid: 3
      pid: 480
      pub: Wiley-Blackwell
      place: Malden, Massachusetts
    artinfo:
      ui:
        194050893
        194050893
        194050893
        10.1002/jcal.70249
        194050893
      ppf: 1
      ppct: 21
      formats:
      tig:
        atl: Exploring Shared Monitoring in Large Language Model (LLM)‐Supported Online Collaborative Problem Solving for High‐Cohesion and Low‐Cohesion Groups.
      aug:
        au:
          Liu, Xiaoyun
          Du, Xu
          Hung, Jui‐Long
          Li, Hao
          Yang, Shuoqiu
          Xie, Yiqian
        affil: Faculty of Artificial Intelligence Education, Central China Normal University, Wuhan, People's Republic of China
      sug:
        subj:
          Artificial Intelligence, Generative
          Collaboration
          Problem Solving
          Monitoring, Physiologic Evaluation
          Group Dynamics
          Task Performance and Analysis
          Goals and Objectives
          Human
          Funding Source
          Male
          Female
          Adult
          Random Assignment
          Feedback
          Simulations
          Behavior Evaluation
          Wilcoxon Signed Rank Test
          Emotions Evaluation
          Cognition Evaluation
          Descriptive Statistics
          Chi Square Test
          Models, Statistical
          Correlational Studies
          Social Cohesion
          Student Attitudes
          Adult: 19-44 years
          Male
          Female
      ab: Background: Recent research increasingly highlights the central role of interventions in enhancing shared monitoring during collaborative problem‐solving. However, traditional intervention approaches suffer from limitations in timeliness and adaptability. Large language model (LLM), equipped with deep semantic parsing and contextual perception, can dynamically detect latent challenges and provide targeted, timely, context‐sensitive feedback. Objectives: This study examines how LLM‐supported interventions affect group shared monitoring during dynamic CPS processes. Methods: This study designed a collaborative problem‐solving platform integrated with LLM, and 28 students from a university in China participated in CPS activities. Chi‐square tests, conditional random fields, linear mixed models and correlation analyses were adopted to examine the changes in both monitoring behaviour and equality of monitoring participation in high‐cohesion (HCGs) and low‐cohesion groups (LCGs) after LLM‐supported group metacognitive scaffolding (LLM‐GMS) intervention, as well as their effects on collaborative performance. Results and Conclusions: The results show that (1) LLM‐GMS activated more socio‐cognitive and behavioural monitoring in HCGs, whereas LCGs mainly exhibited heightened behavioural monitoring. (2) Descriptive analyses revealed divergent trends in monitoring participation equality across group types, with HCGs showing increased equality and LCGs exhibiting a decline. (3) In HCGs, socio‐emotional monitoring was positively associated with collaborative performance, whereas participation equality and behavioural monitoring exhibited negative associations with collaborative performance. In contrast, among LCGs, behavioural monitoring was positively related to performance, whereas socio‐cognitive monitoring was unexpectedly negatively associated with performance. Implications: These findings highlight that LLM‐GMS can be a valuable tool for supporting collaborative learning, but its effectiveness depends on group characteristics and its implementation approach. Summary: What is already known about this topic ○Shared monitoring is recognized as a crucial prerequisite for achieving high‐quality collaborative problem solving, as it facilitates the exchange of task understanding, clarification of goal progress, and coordination of collaborative strategies among group members.○However, students often lack the metacognitive skills necessary to monitor both individual and collective learning processes in complex learning environments. Relying solely on the spontaneous efforts of team members rarely yields effective monitoring behaviours thereby highlighting the need for external support.○Traditional intervention approaches, nevertheless, remain limited in terms of immediacy, adaptability, effectiveness and personalization.○Recently, large language model (LLM), with their advanced capabilities in natural language understanding and generation, has demonstrated significant potential in addressing these challenges.What this paper adds ○Developed an LLM‐supported collaborative platform to scaffold shared monitoring in online collaborative problem solving.○Examined differential effects of LLM scaffolding in HCGs and LCGs.○LLM scaffolding activated socio‐cognitive monitoring and participation equality in HCGs.○Behavioural monitoring showed divergent associations with collaborative performance, being positively related to performance in LCGs but negatively related in HCGs.Implications for practice and/or policy ○This study offers practical implications for educators, particularly in the design of scaffolding strategies and the balance between guidance and learner autonomy.○Educators may consider providing more socio‐cognitive prompts to HCGs and more behavioural guidance to LCGs.○Adopt a 'gradually fade scaffolds' approach to reduce prompts earlier to build independence.Key Points ○LLM‐GMS demonstrates differential activating effects on shared monitoring behaviours in online collaborative problem‐solving. More socio‐cognitive and behavioural monitoring are observed in HCGs, and only enhanced behavioural monitoring appears in LCGs.○Distinct trends in monitoring participation equality are observed across group types, with increased equality in HCGs and decreased equality in LCGs, potentially reflecting differences in groups' scaffold transformation capacity.○The associations between shared monitoring and collaborative performance vary by group cohesion: socio‐emotional monitoring is positively associated with performance in HCGs, whereas behavioural monitoring shows a positive association in LCGs.○The effectiveness of LLM‐GMS depends on the alignment between scaffold design and group cohesion, requiring adaptive and differentiated intervention strategies for distinct group types.○Over‐scaffolding in HCGs and over‐reliance on AI in LCGs are potential disruptive factors, highlighting the necessity of a 'gradually fade scaffolds' approach to balance guidance and learner autonomy.
      pubtype: Academic Journal
      doctype:
        equations & formulas
        pictorial
        research
        tables/charts
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N