Edited by James L. McClelland, Stanford University, Stanford, CA, and approved October 3, 2020 (received for review August 10, 2020)
“What I cannot efficiently break, I cannot understand.” Understanding the vulnerabilities of human choice processes allows us to detect and potentially avoid adversarial attacks. We develop a general framework for creating adversaries for human decision-making. The framework is based on recent developments in deep reinforcement learning models and recurrent neural networks and can in principle be applied to any decision-making task and adversarial objective. We show the performance of the framework in three tasks involving choice, response inhibition, and social decision-making. In all of the cases the framework was successful in its adversarial attack. Furthermore, we show various ways to interpret the models to provide insights into the exploitability of human choice.
Adversarial examples are carefully crafted input patterns that are surprisingly poorly classified by artificial and/or natural neural networks. Here we examine adversarial vulnerabilities in the processes responsible for learning and choice in humans. Building upon recent recurrent neural network models of choice processes, we propose a general framework for generating adversarial opponents that can shape the choices of individuals in particular decision-making tasks toward the behavioral patterns desired by the adversary. We show the efficacy of the framework through three experiments involving action selection, response inhibition, and social decision-making. We further investigate the strategy used by the adversary in order to gain insights into the vulnerabilities of human choice. The framework may find applications across behavioral sciences in helping detect and avoid flawed choice.
Advertisers, confidence tricksters, politicians, and rogues of all varieties have long sought to manipulate our decision-making in their favor, against our own best interests. Doing this efficiently requires a characterization of the processes of human choice that makes good predictions across a wide range of potentially unusual inputs. It therefore constitutes an excellent test of our models of choice (1). We have recently shown that recurrent neural network (RNN) models provide accurate, flexible, and informative treatments of human decision-making (2⇓–4).