Sobre a observa？？o enquanto procedimento metodológico na análise do comportamento: positivismo lógico, operacionismo e behaviorismo radical

behavior analysis adopts two distinct viewpoints about observation as a methodological procedure. in the experimental analysis, it tends to accept only the description of the functional relations among observable variables. in interpretive analysis, however, it infers functional relations among observable and unobservable variables. the radical behaviorist stance about this topic becomes clearer when examined from the influence of operationism over psychology, especially in the 30's and 40's. this paper aims to accomplish such an examination, based on the specialized literature about the topic and on skinner's texts. it concludes that the analysis of private events proposed by skinner: (1) has an interpretive nature; (2) refers, partly at least, to publicly unobservable events; (3) in some measure, employs introspection as a method; (4) to a certain extent, makes inferences about the private world of the other.