There is increasing debate over the unidimensionality of the construct "drive" in theories of behavior. The earliest drive theory postulated a simple entity which increased or decreased as a function of external or internal stimulation and affected behavior monotonically. Duffy and Malmo have recently hypothesized that the effects of drive are not monotonic, but curvilinear; high and low drive inhibit behavior, while moderate drive releases it. The present study examined the effects of different drive levels and sources on verbal productivity, disfluency, lexical redundancy ratios and distributions of lexical redundancy. The study examined the effects of perceived interviewer status, type of evaluation anticipated by interviewees (positive or negative) and specificity of interview question (high or low) on interviewees' verbal productivity and lexical redundancy. Subjects were undergraduates; interviewers were graduate and undergraduate students. Interviewer status and anticipated positive or negative evaluation were induced through visual and written cues. Disfluency results tend to support the Mahl-Osgood hypothesis; results for MSTTR, mean and types support the Hull-Spence hypothesis; and results for verbal productivity, standard deviation, coefficient of variation, and variance fail to support either. (CHK)