Innovative problem solving in nonhuman animals: the effects of group size revisited

Sociality is associated with a variety of costs and benefits, one of which can be to increase the likelihood of individuals solving novel problems. Several hypotheses explaining why groups show higher innovative problem-solving efficiencies than individuals alone have been proposed including the sharing of antipredator vigilance and the pool-of-competence effect, whereby larger groups containing a more diverse range of individuals are more likely to contain individuals with the skills necessary to solve the particular problem at hand. Interference between group members may cause groups to have lower problem solving abilities, however. Using a simulation approach, we model the shape of the relationship between group-level problem-solving probability and group size across a range of facilitation and inhibition scenarios, various population distributions of problem solving, and a task requiring 1 action or 2 actions to be solved. Simulations showed that both sharing of antipredator vigilance and the addition of competent individuals to an existing group lead to positive relationships between group-level problem solving and group size that reach 100% solving probability, whereas interference effects generate group-solving probabilities that rise to a maximum and decrease again, generating a group size for which problem solving is maximized. In contrast, both inhibition and facilitation scenarios generate identical patterns of individual efficiencies. Our results have important implications for our ability to understand the mechanisms that underpin group-size effects on problem solving in nonhumans.