A major concern of this report was to compare the effect of three verbal reinforcement combinations (VRCs) on the rate at which children learn a simultaneous two-choice discrimination problem. The experiments were designed to test the following: (1) whether relative effectiveness of the VRCs was dependent upon task complexity or subject age; (2) what were the effects of the VRCs, task complexity, and overtraining on the learning of a discrimination reversal problem in children; (3) what were the relative effects of the delay of the VRCs; and (4) whether the VRCs would differentially affect the retention of a simultaneous discrimination problem. Findings indicate that regardless of sex or age or problem complexity, the VRCs involving the use of the term "wrong" led to rates of learning superior to those produced by use of the "right-nothing" combination. Similar effects were produced for a discrimination reversal problem. Delay of the VRCs seriously interfered with the learning of a simultaneous discrimination problem in children. Further stimulus research is suggested. (Author/CJ)