Integrating human factors into the design of large-scale distributed applications requires capturing broad patterns of decision-making over time. The proposed theoretical framework introduces a dynamic model that resembles empirical dispersal patterns between the quality of an option and the number of individuals choosing that option. We use the notion of the Ideal Free Distribution (IFD) to estimate the resulting population-dependent equilibrium point and reduce uncertainty about how groups of individuals choose between available options. Our contribution is twofold. First, we identify conditions that lead to the IFD under constrained choice. Second, we illustrate how biases in decision-making can lead to systemic deviations from the IFD.