DiMaCA: a dialogue management model combining communication strategies and affective backchannels to foster WTC-friendly agent conversations

Design · All designs
Evidence · no claims cited

Description

DiMaCA is a dialogue management model specifying rules for when an embodied conversational agent deploys communication strategies (CS) and affective backchannels (AB) during L2 conversations. Its rationale is 'increasing L2 learners' confidence via CSs and reducing their level of anxiety towards communication via ABs', in line with MacIntyre's WTC model. The model is implemented as a strategies management module that detects six conversation states (e.g., learner silent, learner NUNA, agent NUNA) and triggers adapted CS and AB accordingly, with CS chosen in a heuristically predefined order and AB chosen stochastically.

Design Implications

Context

Requirements

• A dialogue script defining the conversation domain, intent detection via external speech recognition services, and databases of CS and AB options per conversation state

Constraints

• The Guessing strategy is only an approximation: the agent does not perform dynamic reasoning about learner intention from a wide range of possibilities but relies on a narrow range of expected answers specified in advance in the dialogue script

Target Learners

• Japanese undergraduate and graduate EFL learners

Learning Goals

• Willingness to communicate in English as a second language

Claims

Related Designs

• Nine communication strategies and four categories of affective backchannels implemented in DiMaCA
• Embodied conversational agent Jack for restaurant-context English conversation practice

Examples

-

Key Sources

• Emmanuel Ayedoun, Yuki Hayashi, Kazuhisa Seta. (2018). Adding Communicative and Affective Strategies to an Embodied Conversational Agent to Enhance Second Language Learners' Willingness to Communicate. Int J Artif Intell Educ. https://doi.org/10.1007/s40593-018-0171-6

Page metadata

Field	Value
type	design
id	dimaca-dialogue-management-model
title	DiMaCA: a dialogue management model combining communication strategies and affective backchannels to foster WTC-friendly agent conversations
description	DiMaCA is a dialogue management model specifying rules for when an embodied conversational agent deploys communication strategies (CS) and affective backchannels (AB) during L2 conversations.
status	review
generated	by process:settle-candidates on 2026-10-28
trust	unverified — no one has confirmed this page's content
sources	1 — emmanuel-ayedoun-2018

Frontmatter is the page's machine-readable record — see the Schema & Guide.