Use this URL to cite or link to this record in EThOS:
Title: Beyond e-learning : an intelligent pedagogical agent to guide students in problem-based learning
Author: Beaumont, Christopher
ISNI:       0000 0004 2740 9551
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Problem-based learning (PBL) has a strong focus on skills and requires students to use self-regulated learning (SRL). However, SRL puts considerable demands on learners and often leads to students making frequent requests for guidance. This PhD thesis set out to explore the extent to which computer-based technology can provide effective guidance to promote the development of students' SRL skills in a PBL context Firstly a qualitative study involving 19 students over 12 weeks established the types of guidance needed by students in a PBL scenario. Subsequently available systems for authoring intelligent tutoring systems were evaluated, identifying those capable of conducting dialogue with multiple users via a web interface and able to adapt responses based on a persistently stored individual student model. A prototype was constructed using the Lingubot® chatbot and evaluated for pedagogical effectiveness and usability using an iterative approach culminating -in an in-depth and holistic qualitative study of eight students, using a think-aloud protocol, interview, questionnaire and analysis of the PBLcoach conversation logs. Results provide evidence that the PBLcoach successfully conducted extended dialogue (30-70 minutes) which enabled all participants to analyse the scenario and identify a well-formed learning goal- one of the foundational skills needed in a PBL task. This thesis provides an original contribution by identifying and formalising five areas for which students require guidance and which comprised the following key requirements for the PBLlSRL tutoring system: Task-guidance (clarification, planning, analysing, deadlines, deliverables, monitoring, evaluating fitness for purpose of research and assessment); Scenario guidance (clarification and requests for further information); Learning resources (how and where to fmd information); Group-work (roles, dealing with issues) and Subject-guidance (understanding, elaboration, application). Important considerations for supportive dialogue in a PBLcoach are also identified: sufficiency and clarity of responses in the student's language to avoid misleading guidance; reinforcement of acceptable responses to build confidence; student control of the level of help available and more levels of hint to attend to student variation. An extensive knowledge base of possible responses is needed, together with an error recovery strategy to enable students to recover after off-task or inappropriate input. Other important findings were to minimise cognitive load through providing guidance at the most appropriate time and level of detail and to pay special attention to responses that engendered positive affective response, particularly the, sense that progress towards a goal was being made.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available