Programa informatizado de ensino pré-matemático com crianças autistas com e sem deficência intelectual

Abstract

The quality of mathematics teaching in Brazilian schools has generated discussions about procedures and professional training. Teaching mathematics through conditional discrimination can be organized into a network of stimulus relations, among which a set of tasks can be taught directly and others evaluated, without direct teaching. Subitization refers to an elementary component in mathematics teaching that can serve as a basis for learning more complex behaviors. Thus, the objective of the study was to evaluate the effect of a teaching procedure composed of conditional discrimination of mathematical concepts (larger/smaller, much/little, equal/different), number-quantity relationship and subitization-based probes, with autistic children and young people with and without intellectual disabilities (ID), applied by specialized professionals, after school. Four autistic students aged six to 15 years with and without ID, enrolled in the regular education system, who presented performance equal to or lower than 61% of correct answers in the Protocol for Recording and Assessment of Mathematical Skills (PRAHM) participated in the study. The teaching procedure was organized in two stages. In the first, the teaching sequence consisted of three groups, namely: equal and different, larger and smaller, and much and little. The second stage consisted of numerical stimuli and their respective quantities. The tasks were presented in blocks: a) AB relation (identification of the attribute/Arabic number, based on the spoken attribute/number; b) AC (identification of quantity, based on the spoken quantity); c) mixed block between AB and AC. Tests were performed with manipulable stimuli to assess the generalization of learning in the probes. To this end, an individual design scheme with multiple probes was carried out between the stimulus groups to assess the results. The assessment of the input repertoire shows that the four participants replicated the pedagogical gap in relation to basic knowledge of Mathematics, considering the school year. Data analysis showed that the number of exposures to the procedure, as the child or young person reached the learning criterion, decreased, in relation to the initial sessions, on average from 2.5 to 1.25. The two behaviors that presented the highest number of correct answers, on average 80% correct answers in the post-test compared to the pre-test, were counting and the numerical sequence relationship. In contrast, the behaviors of identifying differences and the number-quantity concept were those with the highest number of attempts, on average 341, requiring adjustments to the procedure. It is concluded that the procedure planned through the teaching of conditional discriminations favored the acquisition of pre-mathematical behaviors in 25 applications of approximately seven minutes each, showing that through an individualized teaching procedure it was possible to guarantee the learning of such skills.