讲座简介: | The pull-to-center bias is found to prevail robustly in newsvendor orders in laboratory experiments. To explain this bias, various behavioral mechanisms have been proposed, among which regret aversion is a widely mentioned mechanism. We test this mechanism with an EEG-based neural measure on experienced regret that is exactly triggered by comparisons with counterfactual payoffs. The strength of the regret signal effectively predicts the magnitude of the pull-to-center bias out-of-sample, suggesting the avoidance of experienced regret as a fundamental driver of this behavioral bias. Algorithmic recommendation pushes subjects closer to optimal behavior, and its role is more significant for subjects who are more prone to experienced regret. Our study identifies a behavioral channel through which algorithm supports decision makers, and justifies the general relevance of neuro activities to operations management. |