Translational evaluation of training structures in equivalence-based instruction.

Area of Science:

• Behavioral Psychology
• Learning Sciences
• Instructional Design

Background:

• Equivalence-based instruction (EBI) is proposed as an efficient teaching method by minimizing the need to explicitly teach all emergent relations.
• Research is needed to empirically validate the efficiency claims of EBI across different instructional structures.

Purpose of the Study:

• To evaluate the efficiency of EBI compared to complete instruction (CI).
• To investigate whether EBI's efficiency is due to teaching fewer relations.
• To compare different EBI training structures: linear series (EBI-LS) and one-to-many (EBI-OTM).

Main Methods:

• Three groups of college students were trained on establishing stimulus classes.
• Two groups received EBI with either EBI-LS or EBI-OTM structures.
• A control group (CI) received training on all possible relations.

Main Results:

• The EBI-OTM group required fewer instructional trials than the CI group.
• EBI-LS did not show a significant reduction in trials compared to CI.
• Posttest performance was comparable between EBI-OTM and CI, but poorer for EBI-LS.
• No differences in performance were observed on a function transfer test across groups.

Conclusions:

• EBI, particularly with a one-to-many structure, is an efficient instructional approach compared to CI.
• The efficiency of EBI is not attributable to teaching fewer relations, as both EBI groups were trained on the same number of relations.