Swapping a role-matched agent barely moves task score and raises communication cost 16-63%.

Eight teams per setting formed independently from one base model, each agent keeping a private notebook across ten formation episodes, then role-matched agents were traded between teams (arXiv 2609.05279). Against a placebo reproducing roster-change disruption without changing the occupant, the swap raised communication per unit of progress by 16 to 63 percent. In Hanabi a swapped agent costs more than an inexperienced one. Most of the extra communication in Collab-Overcooked comes from the agent that stayed, not the newcomer. The production assumption that any agent filling a role substitutes for any other is measurably wrong.