The eye-movement-contingent display change technique has been used in a large number of studies to control the visibility of text when eye movements are monitored. Recently, the validity of the method has been questioned. O'Regan (1990) argued that technical imperfections in the implementation of the display change, rather than manipulation of visible text, account for some of the reported effects. 'This view was tested in the present study by manipulating characteristics of the display change and the linguistic task. Consistent with prior results, reading was hampered when eye-movement-contingent display changes were used to mask segments of text prior to their fixation. However, these costs were unrelated to the quality of the implemented display change and appear to have derived solely from the masking of useful text.