Characterizing Human Explanation Strategies to Inform the Design of Explainable AI for Building Damage Assessment

TL;DR

This paper investigates how humans explain their damage assessments in satellite imagery to inform the design of more effective explainable AI systems for high-stakes visual detection tasks.

Contribution

It identifies six major human explanation strategies in damage assessment, providing insights to guide the development of task-specific XAI methods.

Findings

Six major human explanation strategies identified

Insights for designing human-centered XAI methods

Implications for improving AI explanations in damage detection

Abstract

Explainable AI (XAI) is a promising means of supporting human-AI collaborations for high-stakes visual detection tasks, such as damage detection tasks from satellite imageries, as fully-automated approaches are unlikely to be perfectly safe and reliable. However, most existing XAI techniques are not informed by the understandings of task-specific needs of humans for explanations. Thus, we took a first step toward understanding what forms of XAI humans require in damage detection tasks. We conducted an online crowdsourced study to understand how people explain their own assessments, when evaluating the severity of building damage based on satellite imagery. Through the study with 60 crowdworkers, we surfaced six major strategies that humans utilize to explain their visual damage assessments. We present implications of our findings for the design of XAI methods for such visual detection contexts, and discuss opportunities for future research.