The purpose of the present investigation was to evaluate the relative effectiveness of three item presentation methods during noncontingent reinforcement (NCR). Four individuals with developmental disabilities and problem behavior maintained by automatic reinforcement were recruited for the study. Single-item stimulus preference assessments were conducted to identify a variety of preferred items associated with low levels of problem behavior. During the treatment analysis, a multielement design embedded within a reversal design was utilized, and the following NCR conditions were compared: (a) presenting preferred items one at a time, (b) presenting multiple items simultaneously, and (c) rotating single items after a predetermined time period. Results indicate that all three NCR presentations methods were effective in decreasing problem behavior, but the single-item condition resulted in the lowest levels of problem behavior for three of the four participants. (Contains 4 figures.)