We evaluated the effects of training novel and existing mands during functional communication training (FCT) to decrease problem behavior for 2 children. A functional analysis (Phase 1) identified mands for FCT. Phase 2 used distinct stimulus conditions to train novel and existing mands. Phase 3 evaluated allocation of responding within a concurrent-schedules design. When reinforcement for either mand was concurrently available, the children used existing mands more than novel mands, but higher levels of problem behavior occurred with existing mands.

An analogue functional analysis revealed that the problem behavior of a young child with developmental delays was maintained by positive reinforcement. A concurrent-schedule procedure was then used to vary the amount of effort required to emit mands. Results suggested that response effort can be an important variable when developing effective functional communication training programs.

We replicated a method for clarifying inconclusive functional analysis outcomes via an extinction analysis of separate topographies of problem behavior with 2 participants. Results suggested that both mild and severe problem behaviors belonged to the same response class. An analysis of response latency was consistent with a response class hierarchy hypothesis, indicating that mild problem behavior nearly always occurred prior to severe topographies of problem behavior.

The effects of noncontingently available alternative stimuli on functional analysis outcomes were evaluated for 2 individuals. Results suggested that noncontingent access to preferred items resulted in reduced levels of attention-maintained behavior. The results are interpreted in terms of establishing operations.DESCRIPTORS: functional analysis, establishing operation, developmental disabilities Pretreatment functional analysis of problem behavior has allowed behavior analysts to tie results of assessment directly to treatment (Iwata et al., 1994). Recently, investigators have begun to evaluate the effects of certain antecedent variables on functional analysis outcomes (e.g., Mueller, SterlingTurner, & Scattone, 2001). One such antecedent, called an establishing operation (EO), is any event that alters the reinforcing value of another stimulus (Michael, 2000). For example, Berg et al. (2000) found that presession exposure to attention resulted in lower rates of problem behavior during contingent attention sessions of a functional analysis, thus affecting the clarity of assessment outcomes. Other factors, such as the noncontingent availability of alternative stimuli during a given functional analysis session, also might decrease the value of the tested reinforcer (Hanley, Piazza, & Fisher, 1997). For example, preferred leisure items often are available during tests for sensitivity to adultWe thank Andrew Gardner, Judy Redmond, and Nathan Call for their assistance with data collection and David Wacker and Wendy Berg for their input on previous versions of this manuscript.Address correspondence to Joel E. Ringdahl, University Hospital School, 100 Hawkins Drive Room 349, Iowa City, Iowa 52242-1011 (e-mail: joel-ringdahl@ uiowa.edu).attention (e.g., Iwata et al., 1994). This practice may result in false-negative findings.The purpose of this study was to evaluate the effects of noncontingently available alternative stimuli on responding during an assessment designed to evaluate the reinforcing effect of attention. These effects were evaluated while 1 individual's problem behavior and a 2nd individual's communicative behavior were assessed. A communication response was included to determine if the effects would be unique to problem behavior. When evaluating potential reinforcers, the effects of alternative stimuli may be important, regardless of whether the behavior is appropriate (and thus targeted for maintenance) or inappropriate (and thus targeted for reduction). METHOD Participants and SettingTwo individuals participated in the evaluation. Tony was a 5-year-old boy who had been referred for assessment and treatment of aggression. Tony functioned in the mild to moderate range of mental retardation, had good adaptive skills, and could communicate vocally. Roland was a 41-year-old man who had been referred for assessment and development of communication skills.

The treatment of behavior maintained by automatic reinforcement (e.g. stereotypy) can present a unique challenge if practitioners cannot control delivery of the maintaining reinforcer. Further, some individuals might engage in stereotypy only when care providers are absent. In the current evaluation, an adolescent boy's hand¯apping was demonstrated to occur in the absence of social contingencies and care providers. To reduce the behavior, two treatment strategies were assessed. In the ®rst approach, verbal reminders to refrain from hand¯apping were delivered on time-based schedules. In the second approach (differential reinforcement of other behavior, DRO), we provided access to a preferred item contingent on prespeci®ed time lengths with no hand¯apping when the participant was alone in a room. Results of the investigation indicated that the verbal reminders were unsuccessful, whereas the DRO program resulted in near-zero levels of stereotypy. Copyright # 2002 John Wiley & Sons, Ltd. EVALUATION AND TREATMENT OF COVERT HAND FLAPPINGAlthough the treatment of many forms of severe problem behavior has bene®ted from functional analysis methodology (cf. Iwata, Dorsey, Slifer, Bauman, & Richman, 1982/1994, the treatment of stereotypic and other behaviors that are maintained by automatic reinforcement continues to be problematic. As Vollmer (1994) described, because the automatic reinforcers maintaining problem behavior are not under the therapist's control, successful treatment of behavior is dif®cult. In many instances, treatment is more complicated because the behavior is most likely to occur when care providers are not present (i.e. the behavior is covert).To date, few studies have described the treatment of covert problem behavior. Grace, Thompson, and Fisher (1996) applied a differential reinforcement schedule to reduce the covert self-injurious behavior (SIB) of one participant. Reinforcers were delivered contingent on absence of permanent products (e.g. tissue damage) of the

We evaluated a differential-reinforcement-based treatment package for the reduction of problem behavior during instructional situations. Differential reinforcement of alternative behavior (DRA; compliance) was implemented across two conditions. During one condition, instructions were presented approximately once every other minute. This condition was considered the terminal goal for treatment. During the second condition, the rate of instructions was gradually increased (beginning at zero and ending when instruction rate was similar to the first condition). Results indicated that DRA with instructional fading resulted in less problem behavior than DRA without instructional fading. These results are similar to previous studies regarding the utility of instructional fading.

No abstract