Explanations Can Reduce Overreliance on AI Systems During Decision-Making

TL;DR

This paper demonstrates that AI explanations can reduce overreliance in decision-making when users strategically choose to engage with explanations, influenced by task difficulty, explanation complexity, and perceived benefits, challenging prior assumptions of inevitability.

Contribution

The paper introduces a formal cost-benefit framework showing that explanations can reduce overreliance when users weigh engagement costs and benefits, supported by five empirical studies.

Findings

Overreliance is affected by task difficulty and explanation complexity.

Monetary incentives influence the likelihood of relying on AI explanations.

Quantifying explanation utility supports the strategic engagement model.

Abstract

Prior work has identified a resilient phenomenon that threatens the performance of human-AI decision-making teams: overreliance, when people agree with an AI, even when it is incorrect. Surprisingly, overreliance does not reduce when the AI produces explanations for its predictions, compared to only providing predictions. Some have argued that overreliance results from cognitive biases or uncalibrated trust, attributing overreliance to an inevitability of human cognition. By contrast, our paper argues that people strategically choose whether or not to engage with an AI explanation, demonstrating empirically that there are scenarios where AI explanations reduce overreliance. To achieve this, we formalize this strategic choice in a cost-benefit framework, where the costs and benefits of engaging with the task are weighed against the costs and benefits of relying on the AI. We manipulate the costs and benefits in a maze task, where participants collaborate with a simulated AI to find the exit of a maze. Through 5 studies (N = 731), we find that costs such as task difficulty (Study 1), explanation difficulty (Study 2, 3), and benefits such as monetary compensation (Study 4) affect overreliance. Finally, Study 5 adapts the Cognitive Effort Discounting paradigm to quantify the utility of different explanations, providing further support for our framework. Our results suggest that some of the null effects found in literature could be due in part to the explanation not sufficiently reducing the costs of verifying the AI's prediction.