Decoding AI's Nudge: A Unified Framework to Predict Human Behavior in AI-assisted Decision Making

TL;DR

This work addresses how AI-assisted decision making can be understood and predicted through a unified, interpretable framework that treats AI guidance as nudges to human decision strategies. It first learns a population-level independent decision model via Bayesian variational inference, then models AI assistance as structured updates to this model across three forms: Immediate, Delayed, and Explanation, with explicit delta vectors controlling feature relevance and attention. The approach demonstrates superior predictive performance over baselines on a diabetes-prediction task and reveals how cognitive styles modulate susceptibility to different AI nudges. Practically, it offers a principled way to tailor AI assistance to individual users and task forms, improving decision support while preserving interpretability. Limitations include a modest feature set and lack of sequential feedback, suggesting directions for extending the framework to richer domains and dynamic interactions.

Abstract

With the rapid development of AI-based decision aids, different forms of AI assistance have been increasingly integrated into the human decision making processes. To best support humans in decision making, it is essential to quantitatively understand how diverse forms of AI assistance influence humans' decision making behavior. To this end, much of the current research focuses on the end-to-end prediction of human behavior using ``black-box'' models, often lacking interpretations of the nuanced ways in which AI assistance impacts the human decision making process. Meanwhile, methods that prioritize the interpretability of human behavior predictions are often tailored for one specific form of AI assistance, making adaptations to other forms of assistance difficult. In this paper, we propose a computational framework that can provide an interpretable characterization of the influence of different forms of AI assistance on decision makers in AI-assisted decision making. By conceptualizing AI assistance as the ``{\em nudge}'' in human decision making processes, our approach centers around modelling how different forms of AI assistance modify humans' strategy in weighing different information in making their decisions. Evaluations on behavior data collected from real human decision makers show that the proposed framework outperforms various baselines in accurately predicting human behavior in AI-assisted decision making. Based on the proposed framework, we further provide insights into how individuals with different cognitive styles are nudged by AI assistance differently.

Figures (5)

• Figure 1: The illustration of how immediate assistance nudges human decision makers.
• Figure 2: The illustration of how delayed recommendation nudges human decision makers.
• Figure 3: The illustration of how explanation only nudges human decision makers.
• Figure 4: Comparing the performance of our method with logistic regression models when changing the size of training data under three forms of AI assistance.
• Figure 5: Comparing the nudging effects of AI assistance on decision makers with different cognitive styles across three forms of AI assistance.