Abstract: Ant colonies vividly exemplify decentralized organization in biological systems. Thousands of individual ants collectively make decisions and build complex structures without any well-informed leaders or centralized control. Instead, group behavior emerges from interactions among individuals that attend only to limited local information and act on it with appropriate decision rules. I will describe empirical and modeling studies that analyze how this process works in one well-studied example: nest-site selection by emigrating colonies of Temnothorax ants. The colony?s challenge is to choose the best among several potential new homes, even when few of the scouts who organize the move assess more than one of the options. The ants? solution is a strategy of graded commitment to potential new homes, governed by each scout?s decision on whether and how to recruit nestmates to a site she has found. Positive feedback generated by this recruitment plays a central role, as does non-linearity introduced by a quorum rule that scouts use to decide on the highest level of commitment to a candidate site. These rules allow a colony to choose the best available site with high probability. Furthermore, ants can quantitatively tune the parameters governing their individual behavior to emphasize either speed or accuracy of decision-making, depending on the urgency of the colony?s need to move. Ultimately, the goal of this detailed, algorithmic description of collective decision-making is a fuller understanding of the mechanisms by which group behavior can emerge from individual behavior, both in natural and artificial systems.