A failure to find effects of relational operant training on relational abilities or scholastic aptitude of school children: A randomised controlled trial

SMART (Strengthening Mental Abilities with Relational Training) is a popular relational training intervention based on Relational Frame Theory which has shown substantial promise in having a positive impact on the intellectual performance and scholastic abilities of school children. Studies investigating SMART to date have, however, been constrained in their inferences by several methodological limitations, in particular issues of small sample sizes, weak control conditions, and a lack of preregistration of study plans and outcomes. In this randomized controlled trial, we aimed to overcome these existing issues by comparing an elaborated version of SMART training to an active comparator (namely, N-Back working memory training) in a larger, preregistered sample of school children. Children (N = 107) completed six weeks (18 sessions) of either SMART or N-Back training, completing measures of relational abilities (the Relational Abilities Index), fluid intelligence (Raven's Matrices), and scholastic abilities (the Drumcondra Reading and Maths Tests) before and after training. Contrary to previous studies, we found no significant difference between the effect of SMART compared to N-Back on any of the outcome measures. We discuss potential interpretations of these results (from the perspective of both proponents and critics of relational training claims): in particular, the lack of change in relational abilities rendered any lack of effects in the further outcomes measures unsurprising. We note that there has been a good deal of variability in how SMART has been implemented in the literature, and call for greater specificity in how SMART should be most effectively implemented. We also call for more concentrated efforts to provide high quality tests of the claims of SMART and relational training generally.