The concept of verbal self-regulation, or verbal mediation, originated in behavioristic analyses of thinking, but was later extended to cognitivistic analyses. In both applications, the research that was generated was usually deficient in ecological validity. In addition, in some of the research verbal self-regulation was inferred when an overt analogue--vocalizing aloud--occurred before the instrumental behavior and not when the vocalizing occurred after the instrumental behavior. The reasoning is faulty because the overt analogue could be mediated rather than mediating. A new line of research, concerning knowledge about self-regulation, is suggested. (Author/SS)