A Grounded Interaction Protocol for Explainable Artificial Intelligence

TL;DR

This paper proposes a structured interaction protocol for explainable AI systems, grounded in analysis of real explanation dialogues, to improve meaningful human-AI communication.

Contribution

It introduces a novel grounded interaction protocol formalized within the agent dialogue framework, based on analysis of 398 explanation dialogues.

Findings

The model closely follows human explanation dialogues.

Grounded theory effectively identifies key components of explanation interactions.

Evaluation with 101 dialogues shows the protocol's applicability.

Abstract

Explainable Artificial Intelligence (XAI) systems need to include an explanation model to communicate the internal decisions, behaviours and actions to the interacting humans. Successful explanation involves both cognitive and social processes. In this paper we focus on the challenge of meaningful interaction between an explainer and an explainee and investigate the structural aspects of an interactive explanation to propose an interaction protocol. We follow a bottom-up approach to derive the model by analysing transcripts of different explanation dialogue types with 398 explanation dialogues. We use grounded theory to code and identify key components of an explanation dialogue. We formalize the model using the agent dialogue framework (ADF) as a new dialogue type and then evaluate it in a human-agent interaction study with 101 dialogues from 14 participants. Our results show that the proposed model can closely follow the explanation dialogues of human-agent conversations.