The contribution of Relational Frame Theory to the development of interventions for impairments of language and cognition.

Abstract

Introduction: Behavior Analysis, Language and Stimulus Equivalence According to behaviour analysis, human language is verbal behaviour, emerging from the social contingencies operating in the verbal community. This perspective on cognitive functioning constitutes a very different approach to human language than that found in mainstream cognitive psychology and in many conceptualisations of language pathology. The conventional approach to the study of language has tended to focus on the symbolic nature of words. Language-able humans are said to be able to manipulate (Clark, 1994), or to use words to refer to objects, events, or relations (Premack, 1976). For the behaviour analyst, however, concepts such as refer and symbol manipulation remain undefined and therefore are not deemed useful in explaining human language. For behaviour analysis the of a word is established through the contingencies that control the emission of the word as an instance of verbal behaviour (Skinner, 1957). For example, the arbitrary vocal dinner has meaning only to the extent that it occasions appropriate behaviour (appropriate being defined by the social contingencies). Originally, behavioural researchers assumed that the meaning of a word was established in much the same way, as are control processes in nonhuman species. In essence, this conceptualisation of symbolic meaning defined symbols as discriminative stimuli. However, both non-behavioural and behavioural psychologists have questioned whether it is possible to accommodate the sheer flexibility of symbolic relations using the principle of discriminative control (cf. Hayes & Hayes, 1989). For instance, it has often been asked how an explanatory framework, which focuses on a history of previous reinforcement, can account for the fact that language-able humans respond appropriately to sentences that they have never previously heard (Chomsky, 1959). This is precisely the crucial question that a growing number of behavioural researchers have focused on in recent years. Specifically, the investigation of the phenomenon termed stimulus has allowed behaviour analysts to explore the behavioural processes whereby stimuli come to control behaviour in new ways in the absence of explicit reinforcement (see Sidman, 1994, for an extensive review). Stimulus equivalence has normally been studied using the matching-to-sample (MTS) procedure. When a subject is taught a series of related conditional discriminations, using the MTS format, the stimuli that enter into these discriminations become related to each other in new ways, without explicit teaching and/or reinforcement. For example, a subject might be taught, given a novel visual sample A1, to select the form B1, from an array containing B1, B2 and B3 (these visual forms are often abstract shapes or nonsense syllables that bear no formal relationship to each other). At the same time, the subject may also be trained, given the sample A1, to select another visual form C1, from an array containing C1, C2 and C3. Without additional training or explicit reinforcement, the subject may derive other relations among the stimuli; therefore the stimuli are said to be members of an equivalence class, and they show the properties of reflexivity, symmetry and transitivity. Reflexivity is inferred from generalised identity matching (e.g. without any training, given A1 as a sample, the subject picks A1 and not A2 or A3 from an array). Symmetry: If the subject is trained, given A1 to pick B1, then the functional reversibility of this trained relation emerges without any direct reinforcement (given B1, pick A1). Transitivity: If a subject is trained 'given A1 pick B1' and 'given B1 pick C1', the relation 'given A1, pick C1' emerges without any training or reinforcement. The phenomenon of equivalence appears to be readily observed in the human species, whereas great difficulty has been encountered in attempting to demonstrate the effect in non-humans (e. …