Relational operants: processes and implications: a response to Palmer's review of Relational Frame Theory.

Palmer has recently criticized Relational Frame Theory (RFT) on the grounds that it has developed data in search of a principle. In this reply, we show that he has done so by attacking fundamental concepts within behavior analysis itself, including the functional nature of an operant and contingencies of reinforcement as a behavioral process. His claim that RFT appeals to new behavioral principles to explain the development of relational operants is shown to be incorrect: As with any operant, RFT appeals to a history of contacted consistencies in contingencies across multiple ex-emplars to explain them. New principles only emerge later as a logically necessary extension of such operants if they exist--a view that Palmer failed to address or appreciate. Palmer's desire to see the use of methods other than matching-to-sample is proper but already largely satisfied in the empirical literature on RFT. We show Palmer's defense of Skinner's definition of verbal behavior to be illogical and unresponsive to the empirical challenge behavior analysis faces. Palmer's alternative common sense mediational associationistic account is another in more than a century of such accounts, all of which have failed empirically. At its root, Palmer's criticism is based on a mechanistic philosophy that is hostile to a traditional functional behavior analytic approach.