Uncertain reasoning about agents’ beliefs and reasoning