@inproceedings{74b47f1c1c5c4a82a4a8696155295000,
title = {A constructionist approach to student modeling: tracing a student’s constructions through an agent-based tutoring architecture},
author = {Katrien Beuls},
editor = {Linda Bradley and Sylvie Thou{"e}sny},
doi = {10.14705/rpnet.2013.000137},
isbn = {978-1-908416-12-4},
year = {2013},
date = {2013-01-01},
booktitle = {EUROCALL Conference proceedings: Learning from the Past, Looking to the Future.},
pages = {45--50},
abstract = {Construction Grammar is a well-established linguistic theory that takes the notion of a construction as the basic unit of language. A construction is a symbolic unit that builds up relational form-meaning mappings through a range of language-dependent categorizations. Although the Construction Grammar framework has developed a powerful descriptive and processing model of language, its current practitioners use it mainly to describe specific constructions in a language (from a synchronic or a diachronic perspective) or to theorize about its basic principles. Yet, the potential of Construction Grammar for language teaching or SLA has largely remained ignored, except for a few rare investigations that have confirmed the potential of learning and teaching constructions. Therefore, this paper demonstrates the benefits of adopting the Construction Grammar approach for modelling a student's linguistic knowledge and skills in a language tutoring application. The two major computational implementations of Construction Grammar, Embodied Construction Grammar (ECG) and Fluid Construction Grammar (FCG), have already experimented with simulations of constructional acquisition in first language learning. However, in this paper we propose a tutoring architecture for (adult) second language learning that relies on a student model that consists of the constructions that the student is thought (by the tutor) to have acquired. This student model is embodied in a fully operational student agent, which has a construction inventory, a grammar engine (to process constructions) and learning strategies (to update constructions after learning). The tutoring architecture also contains a tutor agent, which models a competent language user and has the same three components as the student agent. Additionally, the tutor agent has direct access to the student agent's states and can dispose of a range of tutoring strategies. Through linguistic interactions between the real student and the tutor agent, the student agent models the behaviour of the real student and tries to predict his input. The student construction inventory is aligned to the real student's input after every interaction. This innovative architecture, implemented in Fluid Construction Grammar, is demonstrated here for the use case of Spanish past tense expressions, which remains a complex task even for the most advanced learners of Spanish. Through the use of carefully designed diagnostics and repairs we show that the student construction inventory can be updated to maximally approach the real student's linguistic knowledge of the target domain.},
keywords = {computer-assisted language learning, construction grammar, Fluid Construction Grammar, intelligent tutoring systems, student modelling},
pubstate = {published},
tppubtype = {inproceedings}
}