Association for Behavior Analysis International

The body of empirical evidence in support of the Relational Frame Theory account of verbal behavior has grown steadily in the 30 years since the theory was developed. However, the theory itself remained somewhat unchanged since the first book-length treatment published in 2001. This has limited the theory’s ability to fully explain numerous complex phenomena, including rule-governed behavior and perspective-taking. Recent conceptual developments in the theory, however, look set to change this shortcoming, particularly the development of: the multi-dimensional multi-level framework (MDML); the Differential Arbitrarily Applicable Relational Responding Effects model (DAARRE); and the Relating, Orienting, and Evoking (ROE) model. This symposium explores each of the ways these developments in RFT, while building upon existing concepts. The symposium also examines the various ways in which these developments extend RFT’s account of both rule-governed behavior and perspective-taking, thereby offering greater conceptual and experimental precision.

The concept of rule-governed behavior or instructional control has been widely recognized for many decades within the behavior-analytic literature. It has also been argued that the human capacity to formulate and follow increasingly complex rules may undermine sensitivity to direct contingencies of reinforcement, and that excessive reliance upon rules may be an important variable in human psychological suffering. While the concept of rules would appear to have been relatively useful within behavior analysis, it seems wise from time to time to reflect upon the utility of even well-established concepts within a scientific discipline. Doing so may be particularly important if it begins to emerge that the existing concept does not readily orient researchers toward potentially important variables associated with that very concept. The primary purpose of the current paper is to engage in this reflection. Specifically, we will focus on the link that has been made between rule-governed behavior and derived relational responding, and consider the extent to which it might be useful to supplement talk of rules or instructions with terms that refer to the dynamics of derived relational responding.

Perspective-taking appears to be a key process in the development of “self”. Working under the rubric of Relational Frame Theory (RFT), researchers have investigated the role of perspective-taking as deictic relational responding in the analysis of self in relation to others, place, and time. This research has primarily been conducted through the use of an extended developmental protocol, which is now almost 20 years old. These relational performances have specifically been referred to as responding in accordance with I versus YOU, HERE versus THERE, and NOW versus THEN. The current paper reviews these relational concepts of self and perspective-taking in the context of the original protocol designed to study these performances in young children, populations with language and other developmental difficulties, and even clinical samples. The paper summarizes the various studies and the developmental outcomes that have been observed. The paper also identifies limitations in the use of the original protocol and in the concepts themselves.

In spite of initial wide-spread use of the original RFT-based protocol on perspective-taking as deictic relational responding, conceptual and empirical work on this topic has slowed considerably in the last 10 years. This appears to result mainly from methodological limitations of the original protocol and in difficulties in directly targeting deictic relations with other procedures. However, recent research has attempted to use the well-established Implicit Relational Assessment Procedure (IRAP) to study deictic relations, as a means of assessing relational perspective-taking. In addition, this work has been more recently extended with developments in RFT itself, especially what has been referred to as the multi-dimensional multi-level framework (MDML). The current paper reviews this framework and its many implications for understanding perspective-taking as derived relational responding. The paper also presents some empirical evidence that suggests that the IRAP may have brought the field closer to a functional-analytic approach to perspective-taking and the concept of self than was previously available.

The paper will begin with a brief overview of relational frame theory (RFT; Hayes, Barnes-Holmes, & Roche, 2001), identifying the basic units of analysis proposed by RFT, as a behavior-analytic account of human language and cognition. The impact these analytic units have had, and still have, on RFT research will also be reviewed. A relatively new RFT concept, known as the multi-dimensional multi-level (MDML) framework will then be presented. A recent model of specific properties of relational framing, the differential arbitrarily applicable relational responding effects (DAARRE) model, will also be considered. A case will then be made to integrate the MDML framework and the DAARRE model into a hyper-dimensional, multi-level (HDML) framework. This integration yields a new conceptual unit of analysis within RFT, which involves relating, orienting, and evoking, or ROEing (pronounced “rowing”). The ROE seemingly requires that all of the behavioral processes (but not operations) in the analysis of human language and cognition need to be re-worked, thus rendering RFT a genuinely “scary” prospect for the entire field.