The effect of experience on collective decision-making

Social groups repeatedly solving a complex task can improve their collective performance. To study the mechanisms of collective improvement, we tested the effect of experience on collective decision-making using acorn ants (Temnothorax ambiguus). During a six-emigration training phase, colonies in the choice treatment gained experience choosing to move into one of two nests varying in quality, while colonies in the no-choice treatment had only a single available nest. Both treatments were tested in a subsequent test with two nests of varying quality. We found that experience improved decision-making speed, regardless of treatment. We also found that colonies of the choice treatment were more proficient by carrying a larger proportion of individuals directly into the better-quality nest. However, there was no steady improvement in proficiency throughout their training. Using social network analysis, we quantified changes in group performance over successive emigrations. We found that network density, our measure for social connectedness, and the coefficient of variation of out-strength distribution, our measure for workload distribution, did not differ between treatments and remained stable over successive emigrations. We conclude that collective experience with decision-making may improve subsequent group performance, but the mechanisms of improvement remain unclear. Further research on decision-making in house-hunting ants will advance our understanding of the mechanisms underpinning collective improvement.