Investigating the use of the Relational Evaluation Procedure (REP) in children

The core aim of this thesis was to explore the use of the Relational Evaluation Procedure (REP) in children, both typically developing, as well as developmentally delayed. The REP (e.g., Stewart, Barnes-Holmes & Roche, 2004) is a methodology based on Relational Frame Theory (RFT; Hayes, Barnes-Holmes & Roche, 2001), a contextual behaviour analytic approach to language and cognition that conceptualises these phenomena in terms of derived relational responding (DRR). In order to analyse and experimentally manipulate DRR efficiently, the REP was developed. This protocol induces participants to evaluate the stimulus relation(s) being presented on a given trial using confirmation or disconfirmation (i.e., “Yes” and “No”, or analogous terms). Recently, the REP has been used to rapidly train up relational responding using stimuli possessing already established cue functions (e.g., the English words “Yes” and “No”). It has evolved into several forms of protocol (e.g., SMART, IRAP and RCP), useful for researching various dimensions of relational responding. Most work using the REP has been conducted using adults as participants; however given the advantages of this methodology and its potential as an educational methodology, it seems useful to examine its use in children. The purpose of the current work was to investigate the use of the REP in a range of ages of children, both typically developing and developmentally delayed, over the course of several experiments. Study 1 (n = 26) explored the use of an REP protocol, namely the NSD-REP (Non-arbitrary Same and Different Relational Evaluation Procedure), focused on affirmation and disconfirmation of non-arbitrary same/different relations as an assessment protocol in young typically developing children (i.e., 2.5-4 years). This study involved examining quantitative correlations between the REP protocol, the Preschool Language Scale-4th Edition (PLS4; Zimmerman, Steiner & Pond, 2002) and the Stanford Binet Intelligence Scales-5th Edition (SB5; Roid, 2003). The results showed significant positive correlations between the REP, the PLS4 and the SB5. The second (n = 4), third (n = 8) and fourth (n = 3) studies, involving young typically developing children, explored the use of the NSD-REP as a training protocol. Using a single subject design, REP training was shown to be effective in establishing different levels of affirmation/disconfirmation of non-arbitrary same/different relations in these young participants. Study 5 involved an analysis of the data from Studies 2-4 as a means of providing a preliminary test of the effects of NSD-REP training on the intellectual potential of the participants involved in those studies. The results indicated that participants showed improvement on measures of cognitive and linguistic ability. Study 6 continued the exploration of the NSD-REP with children on the autism spectrum, aged 5-12 years. Initally 9 children were assessed on the PLS4. Subsequently, three participants were trained on Level 2 of the NSD-REP, before being re-assessed on the PLS4. The findings illustrated a significant correlation between the NSD-REP and the PLS4. Additionally, all children met criterion within a number of training sessions and this was accompanied by improvements in PLS4 age-equivalent scores in the case of all three participants. The results of Studies 1- 6 provided detailed insights into the NSD-REP and the protocol evolved based on findings from this regime of testing. Using the more refined version of this methodology, Studies 7-9 were conducted with a slightly older pool of typically developing children. Study 7 (n = 23) provided further substantiation of the NSD-REP as an assessment tool, this time with neurotypical children aged 6-7 years. This study explored correlations between the NSD-REP, the SB5, the Relational Abilities Index (RAI) and standardised school assessments (i.e., Math and Reading). The results showed significant correlations between the NSD-REP, SB5 and the math assessment. Studies 8 (n = 3) and 9 (n = 3) advanced this work further by using the NSD-REP to teach children from this older age group and a younger group (i.e., 4 years) to respond to negation controlled questions in the context of non-arbritary same and different relations. Using a group design, Study 10 (n = 28) further advanced the use of the REP by using an already established REP computer protocol (i.e., SMART- Strengthening Mental Abilities Using Relational Training) and comparing it with a computer coding programme (i.e., Scratch) on IQ and academically relevant tasks. Significant improvements were seen in measures of overall intellectual performance and in academic attainment, in the SMART group alone, and there was significantly more improvement on these measures in the SMART group than in the Scratch group. These results extend previous work using this REP format to train young children in relational framing. In summary, the work presented in this thesis is promising for the future development of the REP. It provides information concerning the age at which children could begin to be taught to use the REP and information concerning the development of key skills useful for the REP, including a repertoire of yes/ no responding. The research also indicates the potential of the protocol as a comprehensive tool for assessment and training of relations with both typically developing children and children with ASD. This would constitute a key addition to existing tools for behavioural intervention.2018-03-2