The Impact of Explanations on Fairness in Human-AI Decision-Making: Protected vs Proxy Features
This addresses fairness issues in human-AI teams for applications like hiring or lending, but it is incremental as it builds on existing explanation research.
The study investigated how explanations affect human-AI decision-making fairness, finding that explanations help detect direct biases but not indirect ones, and tend to increase agreement with model biases, while disclosures can mitigate this for indirect biases.
AI systems have been known to amplify biases in real-world data. Explanations may help human-AI teams address these biases for fairer decision-making. Typically, explanations focus on salient input features. If a model is biased against some protected group, explanations may include features that demonstrate this bias, but when biases are realized through proxy features, the relationship between this proxy feature and the protected one may be less clear to a human. In this work, we study the effect of the presence of protected and proxy features on participants' perception of model fairness and their ability to improve demographic parity over an AI alone. Further, we examine how different treatments -- explanations, model bias disclosure and proxy correlation disclosure -- affect fairness perception and parity. We find that explanations help people detect direct but not indirect biases. Additionally, regardless of bias type, explanations tend to increase agreement with model biases. Disclosures can help mitigate this effect for indirect biases, improving both unfairness recognition and decision-making fairness. We hope that our findings can help guide further research into advancing explanations in support of fair human-AI decision-making.