The Drawback of Insight: Detailed Explanations Can Reduce Agreement with XAI

TL;DR

The paper tackles the assumption that explanations in eXplainable AI are universally beneficial, proposing that user traits such as neuroticism and technological comfort shape agreement with AI recommendations. Through two online studies (norming and agreeability) involving three explanation types—no explanation, placebic, and meaningful—the authors show that individuals with higher neuroticism or lower tech comfort lean toward minimal or no explanations. The analysis using a linear mixed-effects model reveals a significant interaction between explanation type and these traits, while conscientiousness shows no reliable effect. The work argues for trait-aware, adaptive XAI design to improve user collaboration and accessibility, highlighting the need to personalize explanations rather than assuming one-size-fits-all transparency.

Abstract

With the emergence of Artificial Intelligence (AI)-based decision-making, explanations help increase new technology adoption through enhanced trust and reliability. However, our experimental study challenges the notion that every user universally values explanations. We argue that the agreement with AI suggestions, whether accompanied by explanations or not, is influenced by individual differences in personality traits and the users' comfort with technology. We found that people with higher neuroticism and lower technological comfort showed more agreement with the recommendations without explanations. As more users become exposed to eXplainable AI (XAI) and AI-based systems, we argue that the XAI design should not provide explanations for users with high neuroticism and low technology comfort. Prioritizing user personalities in XAI systems will help users become better collaborators of AI systems.

Figures (2)

• Figure 1: Relation of neuroticism and agreement to AI classifier
• Figure 2: Relation of tech comfort and agreement to AI classifier