In this work, we study the collective behavior of fish shoals in annular domains. Shoal mates are modeled as self-propelled particles moving on a discrete lattice. Collective decision making is determined by information exchange among neighbors. Topological distances are used to specify neighborhoods within the fish shoal. We employ random variables to model fish self-propulsion and obedience to group decisions. Global observables along with topological features are used to measure the schooling phenomenon. The one-dimensional model is verified by simulations and experiments on a shoal of zebrafish in an annular tank. Variations in the shoal size and the obedience parameter are numerically explored.