Reed Orchinik
Assistant Professor, Department of Social and Decision Sciences, Carnegie Mellon University
About
Hi! I'm an Assistant Professor in the Department of Social and Decision Sciences at Carnegie Mellon University. I completed my PhD in Management Science at MIT Sloan, where I was advised by Professors Rahul Bhui and David Rand.
My research focuses on how information environments shape cognition, including the ways we process information like news, persuasive messages, and marketing communications. Using computational and formal models, experiments, and large-scale data, I link cognitive processes with the environments in which they operate to better understand the foundations for important phenomena like polarization, resistance to pro-social behavior, and consumption.
I grew up in Phoenix, AZ, and graduated from Swarthmore College in 2019 with high honors in Economics and Political Science. I previously worked as a senior research analyst at the Federal Reserve Bank of New York with economists in the Money and Payment Studies group.
Featured Research
Orchinik, R., Martel, C., Rand, D., Bhui, R.
Management Science, forthcoming
Belief in misinformation has been attributed to digital media environments that promote intuitive thinking, which is thought to foster uncritical acceptance of content. We propose that this intuitive "truth-bias" may be an ecological adaptation to environments where information is typically accurate. Across a large-scale pre-registered survey experiment and an incentivized replication, we test whether intuitions indeed adapt to the base rate of true versus false content. Participants viewed news feeds composed primarily of either true or false headlines. We find that individuals make more — and faster — errors when encountering the less frequent headline type, and fewer errors with the more common type. Computational modeling of the deliberative process reveals these effects are driven by intuitive responses that function like Bayesian priors about content accuracy, which exhibit some persistence. Our findings suggest that susceptibility to misinformation may not merely reflect a cognitive failure, but rather a byproduct of learning from statistical regularities in digital environments.
Understanding and Protecting Against AI Persuasion of Consumers
Orchinik, R., Nam, J., Rand, D. G.
Large language models (LLMs) are increasingly being deployed in persuasive situations. Here, in the context of consumer choice, we identify the extent to which AI can actually change consumers' behavior, how the AI persuades consumers, and how people resist AI persuasion. We find that conversations with LLMs shift consumers' attitudes and incentivized product choices, even in domains where consumers hold strong pre-existing preferences (smartphones, meat consumption, and online retailers) with persuasive effects of 18–33% among initial skeptics. Comparing theory-based persuasion strategies, we find that non-factual strategies (scarcity and urgency, social proof, emotional appeals, and cognitive-bias framing) raised perceived manipulation and skepticism, and these inferences strongly predicted weaker persuasion. Factual and expertise-based appeals, conversely, triggered fewer such inferences and were the most effective strategies. This points to a weakness in psychological resistance to AI persuasion: when persuasion occurs through factual argumentation, people struggle to identify, and thus resist, AI manipulation. Building on this observation, in two final studies we show that alerting participants to the possibility of AI manipulation through a light-touch AI literacy intervention reduces AI persuasion by 48% (without reducing trust in generative AI more generally). Together, these studies illuminate why information-based AI persuasion is particularly effective, and how to protect against it.
