Can Conversational XAI Improve User Performance? An Experimental Study

TL;DR

This study evaluates whether conversational XAI can improve user performance by comparing it to Q&A assistance, finding users can outperform models but with no difference between assistance types.

Contribution

It introduces an experimental design to assess conversational XAI's impact on user performance and compares it to traditional Q&A-based explanations.

Findings

Participants significantly outperformed the model in prediction accuracy.

No significant performance difference between conversational and Q&A assistance.

Modest engagement observed among participants.

Abstract

Explainable AI (XAI) techniques aim to provide insights into predictive models and enhance user performance, yet they often fall short of these expectations. Conversational XAI assistants promise to overcome such limitations, but empirical evidence on their impact on objective performance measures remains limited. We propose an experimental design for evaluating explanation assistance through prediction accuracy, model understanding, and error identification. Using an explainable-by-design prediction model, we create conditions where users can outperform the model by identifying and compensating for systematic errors. We compare conversational assistance against Q&A-based assistance to assess which better supports users in working with model explanations. Preliminary results from testing our experimental design show that participants (N=42) in both treatments significantly outperformed the model but reveal no performance differences between assistance types and modest engagement overall. These findings inform refinements for our planned full study, including enhanced engagement interventions and investigation of the mechanisms driving improved predictions.