This study reports how placebo analgesia was produced by conditioning whereby the intensity of electric stimulation was surreptitiously reduced in order to examine the contribution of psychological factors of suggestibility and expectancy on placebo analgesia. This strategy was used in order to manipulate expectancy for pain reduction. The magnitudes of the placebo effects were estimated after a manipulation procedure and during experimental trials in which stimulus intensities were reset to original baseline levels. Individual differences in suggestibility, verbal expectancy for drug efficacy and manipulation procedure for pain reduction were tested as possible mediators of placebo analgesia. The following dependent variables were measured: (a) subjective expectancy for drug efficacy in pain relief, (b) expected pain intensity and unpleasantness, (c) concurrent pain intensity and unpleasantness and (d) remembered pain intensity and unpleasantness. Statistically significant placebo effects on sensory and affective measures of pain were obtained independently of the extent of the surreptitious lowering of stimulus strength during manipulation trials. The pairing of placebo administration with painful stimulation was sufficient to produce a generalized placebo analgesic effect. However, verbal expectancy for drug efficacy and individual differences in suggestibility were found to contribute significantly to the magnitude of placebo analgesia. The highest placebo effect was shown by the most pronounced reductions in pain ratings in highly suggestible subjects who received suggestions presumed to elicit high expectancy for drug efficacy. The results also demonstrated that placebo effects established on remembered pain were at least twice as great as those obtained on concurrent placebo effects. This was mainly because baseline pain was remembered as being much more intense than it really was. Moreover, remembered placebo effects, like the concurrent placebo effects, were highly correlated with expected pain scores obtained just after manipulation trials. These results indicate that multiple factors contribute to the placebo effect, including suggestibility, expectancy and conditioning, and that the judgement of placebo analgesia is critically determined by whether pain relief is assessed concurrently or after treatment.