This has some undesired consequences. The items with highest exposure rates might become known to the candidates, which implies a test security problem, and it might compromise the testing results. Besides this risk, there is also the loss of investments. Considerable efforts and money have been put in writing and pre-testing items that are not selected for CAT. To deal with both the problems of overexposure and underexposure of items in the bank, exposure control methods have been proposed. The most famous exposure control method has been proposed by Sympson e Hetter (1985). In their method, they conduct a probability experiment after an item has been selected.