ABSTRACT

When problem behavior is dangerous, there is precedent in the literature for inferring its function by conducting a functional analysis (FA) of less severe response topographies that reliably precede it (e.g., FAs of “precursor” behavior). In such cases, it could be advantageous to target socially desirable responses (e.g., mands) rather than less severe problem behavior. In our study, we compared the results of FAs of problem behavior with two FAs (one with prompts and one without) of mands and found poor correspondence in both instances. Although we conclude that FAs of mands are not viable replacements for FAs of problem behavior, these analyses may possess some therapeutic utility. Potential alternative uses for this methodology are discussed.