To cooperatively transport a large load, it is important that
carriers conform in their efforts and align their forces. A downside of
conformism is that it may decrease the group’s responsiveness to
external information. Combining experiment and theory, we show how
ants optimize collective transport. On the single-ant scale, optimization
stems form decision rules that balance individuality and compliance.
Macroscopically, these rules poise the system at the transition between
random walk and ballistic motion where the collective response to the
divergence of the susceptibility at a phase transition. Our theoretical
models predict that the ant-load system can be transitioned through
the critical point of this mesoscopic system by varying its size; we
present experiments supporting this prediction. Our findings show that
the efficient group-level processes can arise form transient
amplification of individual based knowledge.