30 undergraduates were given a Taffel-type verbal conditioning task under double reinforcements, which involve both direct and vicarious reinforcements. Each subject was assigned to one of the three conditions of verbal reinforcement combinations, RW, RN, and NW. After the operant trials the subjects heard tape-recorded responses of the model's trials with vicarious reinforcements, and then performed their own trials with direct reinforcements. The model's and the subject's trials were alternated six times per 10 trials. The result that RW and NW were superior to RN was consistent with the previous findings under the direct and the vicarious reinforcements.