Revisão de literatura e investigação experimental sobre o comportamento de lembrar avaliado em procedimentos de Delayed Matching-to-Sample (DMTS)

Abstract

The general objetive of this work was to investigate possible contingencies involving the remembering in Delayed Matching-to-Sample (DMTS) procedures. The first study presents a review of experimental research articles, carried out in the Journal of Experimental Analysis of Behavior. The manipulated variables, the effects of these variables on remembering, methodological aspects used in the procedures and discussions were examined. In most experiments pigeons was subject. Some variables may favor remembering such as manipulations in the model stimulus (increased ratio of observation responses, longer duration, greater discrepancy between stimuli, different observation responses) and manipulations in consequences (greater magnitude, greater quantity, longer presentation time, higher probability of production, shorter delay). It is concluded that the investigation of the variables present in studies on remembering allowed the identification of factors that favor such behavior and of methodological aspects that can contribute to further research on this phenomenon. It is suggested that further revisions be made and that experimental research with humans be conducted to investigate principles found with non-humans. The second study investigated the effects of teaching DMTS tasks with different delay intervals (0s, 2s, 4s and 6s) in two stimulus modalities on choice responses in human participants. 16 participants were submitted to the tasks; half started in the Auditory Condition and did the Visual Condition on the second day, and half followed the reverse order. The procedure did not consistently lead to stimulus generalization results, not being sufficient to assess whether in human participants the delay interval can control recall in DMTS tasks. Regarding the sensory modalities, the Auditory Condition allowed a higher percentage of correct answers in the training and a greater number of generalization gradients. It is suggested that further studies be carried out with changes in procedures to assess how the delay interval can control the participants' choice responses and provide data for the understanding of remembering as a behavior under the control of temporal dimensions