Infectious　respiratory　diseases　are　known　to　have　high　levels　of　airborne　transmissibility.　However,　traditional　ventilation　methods　based　on　perfect　mixing　often　lead　to　the　diffusion　of　airborne　pathogens.　A　ventilation　method　that　can　reduce　air　mixing　is　necessary.　The　ventilation　system　should　also　be　able　to　adapt　to　changes　in　the　use　scenario　of　a　given　room.　For　this　reason,　a　new　type　of　ventilation　method,　referred　to　as　multi-vent　module-based　adaptive　ventilation　(MAV),　is　proposed,　and　its　performance　in　contaminant　diffusion　control　is　evaluated　in　this　study.　Computational　fluid　dynamics　(CFD)　is　applied　to　investigate　the　contaminant　distribution　in　a　two-desk　office　with　MAV.　Tracer　gas　(CO2)　is　used　to　simulate　coughed　contaminants　from　an　infected　person.　For　three　different　MAV　air　distributions　(vertical,　parallel,　and　cross　mode)　and　traditional　mixing　ventilation　(MV),　the　contaminant　concentration　distributions　and　contaminant　variations　in　the　oronasal　areas　of　the　occupants　are　compared.　The　results　show　that　MAV　results　in　a　smaller　contaminant　diffusion　area　and　lower　contaminant　diffusion　speed.　Furthermore,　MAV　can　reduce　the　peak　contaminant　concentrations　at　the　oronasal　areas　of　the　occupants　to　12%　of　that　of　MV.　The　levels　of　performance　of　the　three　MAV　modes　are　different　because　of　the　different　airflow　patterns　that　they　create.　For　the　tested　location　of　the　infected　person,　the　vertical　mode　has　the　best　performance.　The　average　cumulative　inhalation　of　contaminants　in　the　vertical　mode　is　23.7%　lower　than　that　in　the　cross　mode　and　36.5%　lower　than　that　in　the　parallel　mode.