Working Paper: CEPR ID: DP6535
Authors: Isabelle Brocas; Juan D. Carrillo
Abstract: Building on evidence from neurobiology and neuroscience, we model the physiological limitations faced by individuals in the process of decision-making that starts with sensory perception and ends in action selection. The brain sets a neuronal threshold, observes whether the neuronal cell firing activity reaches the threshold or not, and takes the optimal action conditional on that (limited) information. We show that the optimal threshold is set in a way that existing beliefs are most likely to be confirmed and least likely to be refuted. The conclusion holds in static and dynamic settings, and with linear and quadratic loss functions. We then relate our result to the somatic marker theory, and argue that it provides support for the hypothesis that emotions help decision-making. Last, we discuss the implications for choices in concrete vs. abstract situations, for interactions in cooperative vs. competitive activities, for reactions to expected vs. unexpected events, and for the choice of cognitive vs. affective encoding channels.
Keywords: Bayesian learning; emotion; information processing; neurobiology; neuroeconomics
JEL Codes: Z0
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| level of confidence in a belief (D80) | optimal neuronal threshold (C45) |
| optimal neuronal threshold (C45) | action taken based on that belief (D91) |
| emotions (D91) | quality of decision-making (D91) |
| presence or absence of emotional states (D91) | neuronal thresholds (D87) |
| neuronal thresholds (D87) | suboptimal choices (D91) |
| initial beliefs (D83) | optimal neuronal thresholds (C45) |