Both conditioning and expectancy models have been offered in recent years as explanations for the placebo response. Following our earlier work on conditioning placebo responses in human subjects the current study examined the relative contribution made by conditioning and verbal expectancy. Group 1 received a Combined Expectancy and Conditioning Manipulation; group 2 received Expectancy Alone; group 3, Conditioning Alone; and group 4 was the control group. Subjects' responses were compared with and without a placebo cream, using iontophoretic pain stimulation. The results suggest that conditioning was more powerful than verbal expectancy in creating a placebo response.