We examined the effects of three different presession conditions on tangibly maintained problem behavior for 2 students with autism, using individual-participant multielement designs. First, an analogue functional analysis demonstrated that problem behavior was maintained by access to tangible items. Next, topographies of item rejection were identified. Finally, students were exposed to (a) brief access, (b) no access, and (c) satiation to the tangible items prior to tangible sessions. The results demonstrated high levels of problem behavior following the brief-access and no-access presession conditions and low levels of problem behavior following the satiation condition. The findings are discussed in the context of how satiation might best be defined for these sorts of evaluations. (Contains 3 figures.)