Cognitive Model Priors for Predicting Human Decisions

TL;DR

This paper introduces a novel approach combining cognitive model priors with neural networks and provides a large-scale dataset, significantly improving the prediction of human decision-making under uncertainty.

Contribution

It develops cognitive model priors through pretraining neural networks on synthetic data and releases a large human decision dataset, enhancing predictive accuracy and benchmarking.

Findings

Pretraining neural networks with cognitive model data improves predictions.

Fine-tuning on real data yields state-of-the-art results.

The dataset enables analysis of when cognitive priors are most effective.

Abstract

Human decision-making underlies all economic behavior. For the past four decades, human decision-making under uncertainty has continued to be explained by theoretical models based on prospect theory, a framework that was awarded the Nobel Prize in Economic Sciences. However, theoretical models of this kind have developed slowly, and robust, high-precision predictive models of human decisions remain a challenge. While machine learning is a natural candidate for solving these problems, it is currently unclear to what extent it can improve predictions obtained by current theories. We argue that this is mainly due to data scarcity, since noisy human behavior requires massive sample sizes to be accurately captured by off-the-shelf machine learning methods. To solve this problem, what is needed are machine learning models with appropriate inductive biases for capturing human behavior, and larger datasets. We offer two contributions towards this end: first, we construct "cognitive model priors" by pretraining neural networks with synthetic data generated by cognitive models (i.e., theoretical models developed by cognitive psychologists). We find that fine-tuning these networks on small datasets of real human decisions results in unprecedented state-of-the-art improvements on two benchmark datasets. Second, we present the first large-scale dataset for human decision-making, containing over 240,000 human judgments across over 13,000 decision problems. This dataset reveals the circumstances where cognitive model priors are useful, and provides a new standard for benchmarking prediction of human decisions under uncertainty.