Evaluating Actionability in Explainable AI

TL;DR

This paper explores how explanations in XAI influence user actions by creating a detailed catalog linking information types to specific actions, aiding AI creators in designing more actionable explanations.

Contribution

It introduces a catalog mapping information categories to user actions, enabling better evaluation and design of explanations in XAI systems.

Findings

Identified 12 information categories used in decision-making

Mapped 60 user actions to these information categories

Provided a framework for testing explanation effectiveness

Abstract

A core assumption of Explainable AI (XAI) is that explanations are useful to users -- that is, users will do something with the explanations. Prior work, however, does not clearly connect the information provided in explanations to user actions to evaluate effectiveness. In this paper, we articulate this connection. We conducted a formative study through 14 interviews with end users in education and medicine. We contribute a catalog of information and associated actions. Our catalog maps 12 categories of information that participants described relying on to take 60 different actions. We show how AI Creators can use the catalog's specificity and breadth to articulate how they expect information in their explanations to lead to user actions and test their assumptions. We use an exemplar XAI system to illustrate this approach. We conclude by discussing how our catalog expands the design space for XAI systems to support actionability.