Observation training: Evaluating a procedure for generating self -rules in the absence of reinforcement
To conduct a behavioral analysis of the phenomenon known as Observation Learning, one needs to reconcile the research in that area with currently known principles and concepts of Behavior Analysis. In doing so, it became clear that a repertoire of rule-stating and rule-following may account for research result in that area. A computer-based sorting task and verbal behavior questionnaire were employed to test whether an Observation Training procedure could increase accurate sorting and accurate rule statements with respect to the task in the absence of reinforcement for those responses. Observation Training included ten trials of a data collection task that provided feedback only for accurate data collect responses. Four variations of Observation Training were tested. Results indicate that there were improvements in accuracy in both repertoires as a result of Observation Training. Variability of the results was discussed as an outcome of varying histories of reinforcement for rule stating and rule-following prior to this experiment. Additionally, results indicate that the interaction between the repertoires of rule stating and rule-following can account for results observed in the Observational Learning literature.