Information, entropy and the paradox of choice: A model for understanding human choice behavior

Abstract

Choice overload occurs when individuals feel overwhelmed by excessive alternatives during decision making. Although larger choice sets are often assumed to be more satisfying, behavioral evidence reveals an inverted U-shaped relationship between satisfaction and choice set size. However, quantitative frameworks linking information processing to choice satisfaction remain underdeveloped. Here, we develop a simple framework based on relative entropy and effective information to explain this behavior. We propose that satisfaction depends on the probability of finding an ideal option within a choice set and is determined by the informational structure of preferential choice probabilities relative to a baseline state of indifference. Small to moderately sized sets allow efficient comparison and identification of preferred options, thereby maximizing both effective information and satisfaction. As the number of alternatives increases, cognitive limitations increase uncertainty, leading to reduced effective information and satisfaction. This mechanism naturally produces the experimentally observed inverted U-shaped dependence of satisfaction on choice set size. Behavioral experiments across varying choice set sizes closely matched model predictions, suggesting that effective information provides a robust metric for choice satisfaction. These findings offer a principled theoretical account of the paradox of choice and carry broader implications for consumer psychology and human choice behavior.