O efeito de variáveis do procedimento de pareamento estímulo-estímulo sobre a frequência de respostas vocais de crianças com Transtorno do Espectro Autista

Abstract

The explanation of the basic processes underneath the development of verbal behavior remains a challenge for Behavior Analysis. Stimulus-stimulus pairing (SSP) is a procedure that has been investigated and that aims to increase the frequency of vocal responses for children with Autism Spectrum Disorder (ASD) by presenting vocal sounds paired with items of interest to the child (edibles, tangibles or social). The results of previous studies were quite diverse as to the effectiveness of the procedure. For this thesis, three studies were conducted with the objective of a) evaluating the effect of variables present in the SSP on the frequency of target vocal responses; b) investigating the existence of correlations between characteristics of the participants and the efficiency of the SSP; and c) assessing possible similarities between the effects of the SSP and other pavlovian conditioning procedures. All children from the three studies were assessed for their verbal repertoire with The Behavior and Language Assessment Form (BLAF) before the first session. In Study 1, the effects of the vocal pattern during pairings (monotonic or maternal) were investigated. Nine children with ASD, between 3.9 and 8.2 years old, participated. Two sounds were selected per child, one target sound for each vocal pattern. The first session served as a baseline and consisted of presenting target sounds without pairing. In the following ten sessions, the target sounds were presented in the corresponding sound patterns and paired with favorite items (toys or social stimuli) for each child. Overall, target response rates increased at a low magnitude, below 0.25 R / min. Rates greater than 0.25 R / min were obtained with five out of nine participants, in up to three sessions for each. No differences were found depending on the sound pattern. In Study 2, the objective was to assess the effects of Forward and Backward conditioning on the target response rates. Twelve children with ASD, aged 2-5 years old, participated. In this study there were two target sounds, one for each type of conditioning, and a control sound was added to be presented without being paired with preferred stimuli. Response rates, as in Study 1, were below 0.25 R / min in most sessions. The difference between the frequencies of Forward and Backward responses was statistically significant (Wilcoxon test - T = 55, p = 0.006). The correlation analysis of two behavioral derived measures of individual data of the children revealed two negative correlations, between the children's age and the total level of responses, and between the BLAF and the conditioning index. In Study 3, the objective was to assess the effect of different inter trial intervals (ITIs) on the frequency of target responses. Eight children with ASD, between 2.2 and 7.3 years old, participated. Two target sounds were used with each child, each paired with a different ITI, short (20 s on average) or long (60 s on average). As in the other two studies, target response rates were low. No differences were found in response rates depending on the duration of the ITIs. As for the effectiveness of the SSP, the three studies replicated previous results in which the procedure was effective in increasing the occurrence of vocalizations for part of the children, but not for all of them. When there was an increase in frequency, it was of low magnitude. Unlike the type of conditioning, the sound pattern and the duration of the ITI did not produce statistically significant differential effects on the response rate. The results of Study 1 suggest that the sound pattern may be irrelevant to the efficiency of the SSP, at least for children with ASD in the same age range, with similar repertoires. The results of Study 2 (a) clearly demonstrate that the order in which the stimuli are presented in the pairing is essential for the increase in frequency of target responses, (b) indicate that the Backward pairing may have produced inhibitory effects, similar to what is produced with other preparations of Pavlovian conditioning (e.g., autoshaping) and (c) suggests that the correlations found between behavioral derived measures and characteristics of the participants (age and BLAF score), might help to establish possible predictors of the procedure's success in applied settings. The results of Study 3 contrast with the results of animal studies. Possibly this was due to the insufficient difference between the ITIs used in the present study, significantly less than that used with animals. Other general implications of these results for the understanding of the initial speech development and study limitations are discussed.