This paper reports results of two studies that sought to teach generative reading skills to a large group of Brazilian children who were exhibiting protracted failure in school. Inspired by Skinner's analysis of verbal relations and minimal verbal units, the methodology took advantage of certain characteristics of Portuguese. Many words in this language are comprised of two-letter syllabic units (e.g., BO+LA= ball, CA+BO= handle, LA+TA= can) that can be recombined to form new words (e.g., BOCA= mouth, BOTA= boot), thus establishing a route to generative reading via recombinative generalization. Such syllabic units were incorporated within curricular framework that used matching-to-sample and learning by exclusion methods to teach matching relations involving pictures, printed and spoken words, and printed and spoken syllables. Study 1 was conducted within a university-based learning center that maintained certain aspects of laboratory conditions. It showed that teaching textual relations between dictated and printed syllables could control procedurally the inter- and intra-participant variability observed in past studies that lacked this feature -resulting in virtually universally positive teaching outcomes. Study 2 was conducted in a public school programs that applied the same basic training methodology. Positive training outcomes in an experimental group were approximately 3-5 times greater than that in a placebo control group. Together, these studies illustrate that the functional analysis in Verbal Behavior is having a direct impact in educational science in Brazil. It has led to procedures that can be effectively translated from the laboratory to the community via delivery systems that can be implemented in the developing world.
The inference that others are subject to private events is almost universal among humans. Since no one has direct access to another person's private events, we have proposed this inference occurs because of: 1) The almost simultaneous occurrence a child's private kinesthetic stimuli and the visual stimuli produced by another person's motor act during imitation of motor acts; 2) The similarity between the child's vocal behavior and that of another person during vocal imitation; and 3) The stimulus equivalence that occurs when the child's behavior and similar behavior of others are given the same name. We have proposed that perspective taking is a very common activity in our daily lives and that performance on false belief tests is a special case of perspective taking. In our analysis of the prerequisites for successful predictions on false belief tests we have considered false belief tests as primarily predictions concerning the behavior of others in situations in which discriminative stimuli are available to the child being tested and not to the protagonist about whom the child is to make a prediction. Predictions about other's behavior are made on the basis of three types of prior observations and descriptions: (a) observation and descriptions of the behavior of a specific individual in similar situations; (b) observation and descriptions of the behavior of many different people in similar situations; and (c) observation and descriptions of one's own behavior in similar situations. Success on the false belief tests necessitates discrimination between the stimuli available to the child and those available to the protagonist.