Matrix training techniques arrange instruction for stimulus relations that facilitate emergent responding to novel stimulus arrangements, which is a phenomenon known as recombinative generalization. The current study compared two common matrix training approaches, an overlapping (OV) design and a non-overlapping (NOV) design, with respect to arranging relations targeted for training. Two, typically-developing toddlers were taught compound action-object labels in either an OV or NOV matrix training design. Results suggest that an OV matrix design facilitates recombinative generalization more effectively than a NOV design.

No description available.

Verbal behavior is a critical repertoire for children with autism spectrum disorder to acquire. Tacts—verbal behavior evoked by nonverbal stimuli—are important for communicating about the world around oneself. Noun-verb tacts are part of a robust tact repertoire and may be addressed during applied behavior analytic intervention. When acquiring noun-verb tacts, it is important that the speaker learn to respond to many variations of stimuli like novel combinations of learned nouns and verbs, which is called recombinative generalization. One strategy to teach multi-component targets, such as the noun-verb tact, and lead to recombinative generalization is matrix training. Matrix training is a framework utilized to arrange targets that can be combined in order to facilitate recombinative generalization by teaching a subset of combinations and then probing others. With three children with ASD, we used matrix training and evaluated the acquisition of trained and novel combinations of noun-verb tacts with GIFs as stimuli arranged in three matrices. We used a concurrent multiple probe design across sets, and our results indicated that all participants acquired trained noun-verb tact targets in the presence of the GIFs. The degree of recombinative generalization varied across participants, but each participant demonstrated recombinative generalization with some stimuli. We analyzed responding during generalization probes to identify possible sources of stimulus control. We discussed the errors that were emitted when testing for recombinative generalization and provided suggestions for future research on matrix training and recombinative generalization.

Recombinative generalization is a generative outcome that involves responding to novel stimulus combinations, and it can be facilitated through an instructional approach called matrix training. A learner's history with constituent stimuli and the arrangement of combination stimuli within the instructional matrix may affect the likelihood of recombinative generalization. To investigate this further, the current project assessed recombinative generalization with novel combinations of abstract stimuli by programming specific training histories for undergraduate student participants. The matrix training conditions were: (a) trained constituents with overlap training, (b) untrained constituents with overlap training, (c) trained constituents with nonoverlap training, and (d) untrained constituents with nonoverlap training. We evaluated whether and the extent to which recombinative generalization occurred in each matrix training condition in comparison to a condition that included training the constituents and providing a word-order rule. Finally, we compared the training trials in experimental conditions to directly training all constituents and combinations. The results suggested both overlap conditions and the trained constituents with nonoverlap condition produced recombinative generalization, and the trained constituents with nonoverlap condition was the most efficient. These results could inform the training order and stimulus arrangements practitioners employ to program for recombinative generalization.

A common goal of instructional techniques is to teach skills effectively and efficiently. Matrix training techniques are both effective and efficient as they allow for the emergence of untrained responding to novel stimulus arrangements, a phenomenon known as recombinative generalization. However, it is unclear which type of matrix arrangement best promotes recombinative generalization. The current study compared two common matrix training approaches, an overlapping (OV) design and a non-overlapping (NOV) design, with respect to arranging relations targeted for training. We conducted a replication evaluation of a Wilshire and Toussaint study, and taught two typically-developing preschoolers compound object-action labels in Spanish and used either an OV or NOV matrix training design. Results from both studies demonstrated the participant trained with an OV design produced recombinative generalization and participants trained with a NOV design produced significantly low levels of emergence or none at all. These results suggest that an OV matrix design facilitates recombinative generalization more effectively than a NOV design. Implications for instructional arrangements are discussed.

No description available.