The ability of a testing system to adjust the difficulty of questions based on the test-taker's performance.