Collective versus individual cognition in decision-making by ant colonies

Decision-making by social insects is marked by a tension between group and individual cognition. Choices can emerge from the interactions of many poorly informed individuals, none of which directly compares all options and chooses the best. At the same time, colony members have much in common with solitary insects that regularly make decisions on their own. This creates ambiguity about whether a collective decision is truly an emergent property, or instead results from the choices of one or more well-informed individuals. Using the model system of nest site selection in Temnothorax ants, I show that individual workers are fully capable of making the decisions normally carried out by the colony as a whole, provided they have timely access to all options. However, individuals typically lack this full information, and so are more likely to contribute a small part to a highly distributed decision. Furthermore, fully informed individuals are more susceptible than colonies to certain systematic errors that lessen their ability to consistently make the best choice. These results indicate that sophisticated collective decision-making algorithms do not necessarily yield wholesale new cognitive abilities lacking in individual ants. Instead, they expand the scale of problem that can be solved, and refine the efficacy of the solution.