Training Structures and the Formation of Equivalence Classes: A Replication

Follow-Up Study, Nodal Number, Response Time, Retention Test, Stimulus Equivalence, Training Structures

Committee members: Henry D Schlinger, Michele Wallace, Erik Arntzen, Andrea Zetlin

Thesis (M.S.) California State University, Los Angeles, 2012

The simultaneous protocol, in which all baseline relations are trained first before testing for emergent relations, is the most frequent used protocol when training and testing for the formation of equivalence classes. When training participants in six conditional discriminations and testing for the emergence of three 3-member or 4-member classes, the differences between OTM and MTO training structures are minimal, while the linear series (LS) training structure is less effective (e.g., Arntzen et al., 2010). However, one of the predictions from the discrimination analysis by Saunders and Green (1999) is that the differences in stimulus equivalence outcome between training structures should increase with an increased number of class members. Hence, the purpose of the present study was to test for differential stimulus equivalence outcome in three 3-member classes vs. three 6-member classes. We also wanted to replicate earlier findings on reaction time, which previously have shown an increase from baseline to testing as well as a higher increase in reaction time on equivalence than symmetry tests. In addition, we explored the nodal number effects in the LS training structure as well as investigated the differential effects of training structures on stimulus equivalence outcome in a follow-up test two to four weeks after initial training and testing.