DIGITAL LIBRARY
METALOGUE’S VIRTUAL AGENT FOR NEGOTIATION: ITS’ EFFECTS ON LEARNING EXPERIENCE, METACOGNITIVE AND INDIVIDUAL-AND-COMMUNITY-LEVEL ATTITUDES PRE-AND-POST INTERACTION
1 Organisational Psychologist, Academic Researcher (GREECE)
2 Department of Strategic Planning, Hellenic Parliament (GREECE)
3 University of Peloponnese (GREECE)
4 European Programs Implementation Service, Hellenic Parliament (GREECE)
About this paper:
Appears in: EDULEARN19 Proceedings
Publication year: 2019
Pages: 1542-1551
ISBN: 978-84-09-12031-4
ISSN: 2340-1117
doi: 10.21125/edulearn.2019.0460
Conference name: 11th International Conference on Education and New Learning Technologies
Dates: 1-3 July, 2019
Location: Palma, Spain
Abstract:
The current paper explores learners’ experience (system’s usability)(macro-features) when the METALOGUE (Multiperspective Multimodal Dialogue: dialogue system with metacognitive abilities) system (virtual agent) was used to teach metacognitive- and individual-and-community level attitudes and skills (micro-factors) in multi-issue negotiation. The virtual agent is component of the METALOGUE pilot study that was run at the Hellenic Parliament in Greece and developed as part of the EU METALOGUE research project (http://metalogue.eu). METALOGUE project has designed and evaluated its’ multimodal dialogue system including several models (i.e. cognitive, learning, interaction and dialogue management) as exemplified in many modalities such as spoken natural language, facial expressions, body posture and biosensor data. The system played the role of one of the dialogue participants and acted as a tutor guiding multiple users, with its strategic impact expanded upon intelligent virtual environments (IVEs), cognitive modelling, multimodal, multi-party human computer interaction (HCI) and technology-enhanced learning (TEL) alike.

The METALOGUE pilot study employed the pre-and-post assessment approach (experimental design) to explore the effects of the METALOGUE system on the learning experience of 41 mature learners of different demographic and academic background. First, 5 participants interacted in English in pairs with the system to investigate consecutive and overlapping visual signals and evaluate the load of information presented to achieve completeness and informativeness in the real time system context. Then, 41 participants fluent in English took part in the system evaluation session followed. They were introduced to the system’s functionalities, witnessed 2 demo interactions from the facilitators, were given time to ask for clarifications and signed the consent forms. Then, each learner negotiated 3 diverse multi-issue scenarios with the system for 15 minutes for all 3 sessions. Finally, they were asked to answer pre-and-post learning experience questionnaire in using the system (macro-factors) and metacognitive-and individual-and-community-level attitudes and skills (self-efficacy, self-regulation, interpersonal and problem-solving skills, civic action, individual readiness to change and mastery goal orientation) (micro-factors) through multiple choice related questions adopted for the needs of the present research.

Findings indicated significant and favourable associations between learners’ system usability questions post-interaction and between self-efficacy, self-regulation, individual readiness to change, mastery goal orientation, interpersonal and problem-solving skills and civic action before-and-after system’s learning experience. Implications, limitations and avenues for further research are discussed in view of multimodal dialogue learning environment.

Keywords:
Multimodal dialogue system, interactive learning, metacognition, individual-and-community attitudes and skills.