Procedimentos para ensinar comportamento textual com base na nomeação de figuras
Abstract
This study is part of a research program that has the main goal of developing procedures to foster reading acquisition. In the present experiment we investigated the transfer of stimulus control from pictures to printed words in the emergence of textual behavior. Participants were 42 first degree students, attending a public school, who had a history of school failure. The students were able to name the pictures and the procedure was built on this repertoire: the basic procedure consisted of pairing pictures and the printed
words. Periodically, the students were presented only with printed words, to verify if
and when they were able to read the words. A control procedure consisted of pairing printed words dictated words (rather than the figures). The visual stimuli (pictures and printed words) were presented within a square in the center of the monitor screen. The dictated words, previously recorded, were presented through speakers. The student s task consisted of saying the word that corresponded to the printed word on the screen. The student s oral response finished the trial and produced differential consequences
(for either correct or incorrect responses). If no answer was given, the trial automatically ended after twenty seconds. Students were allocated to four experimental conditions. In the first condition (Simultaneous pairing), the visual stimuli (pictures and printed words) were simultaneously presented on the screen. In the second one (Fading), the picture was gradually faded-out along successive trials. In the third condition (Delay), the trial began with the printed word on the screen and the picture was
presented some seconds later; the duration of the delay was gradually increased along successive trials. In the fourth condition (Control) a dictated word was presented, rather than the picture, simultaneously with the printed word. In all of the conditions, the teaching program was the same: 50 non-abstract words were taught and tested, divided into 5 groups of 10 words. Each block of 10 training trials alternated with a block of 10 test trials in which only the printed word was on the screen. The experiment was
conducted in two phases. In Phase 1 a fixed number of training blocks was presented (5)
in order to establish how much each participant acquired of the target performance. In Phase 2, the training continued until the student reached 100% of correct answers in a group of 10 words, before proceding to the next group. A reading test with the 50 taught words (retention) and 30 new words (generalization) was conducted at the end of Phase 2. For each experimental condition Simultaneous, Fading, Delay and Control, the
median percentage of reading trained words at the end of Phases 1 and 2 were, respectively: 10 and 40%; 0 and 33.3%; 13.3 and 40%; and 6.3 and 40%. The increases in the percentage of correct reading from Phase 1 to Phase 2 were statistically
significant in all four conditions, but there were no differences among the conditions
neither in Phase 1 or Phase 2. The tests with new words, in both phases, showed that there was no reading generalization. In spite of some improvement in reading trained words, the retention of textual behavior was very low (less than 50% in all of the conditions), showing transitory effects of the teaching procedures and difficulty in the transfer of stimuli control from pictures to printed words. The inefficacy of the procedure could be due to several variables, including a blocking effect or a lack of
systematic pairing between spoken words and printed words, pointed by literature as
essential for the reading acquisition. However, data from the control condition, in which
the dictated words were paired with printed words, showed the same performance levels
and therefore rules-out this possibility and suggest the need to investigate other
variables, such as the student s entrance repertoire.