The emergence of a conditioned cocaine-induced hyperlocomotion was examined in C57BL/6J mice using a procedure that has not been used previously. Two days after a session of preexposure to the test chambers under saline, a first group of mice (cocaine-cued) received five once-daily injections of 10-mg/kg sc cocaine every other day (on the odd days of the chronic treatment period) and a saline injection on the 5 days following each cocaine injection day (on the even days of the treatment period), in all cases before being placed in the test chamber. Another group of mice (saline-cued) received 10 injections of saline on both the even and the odd days in the same context, and a third group of mice (cocaine-uncued) received five injections of saline on the even days in the test context and five injections of cocaine on the odd days in an alternative context. On the odd days sessions, the cocaine-cued group showed significant repeated increases in locomotion without behavioural sensitisation being induced, whereas the saline-cued levels of locomotion remained on baseline levels. On the first even session, the three groups did not differ from each other and showed lower levels of locomotion than on the preexposure session. During the two following even sessions, the cocaine-cued group showed an increase in locomotion that levelled off on the two remaining sessions, whereas the saline-cued and the cocaine-uncued groups (which presented comparable values) exhibited significantly lower levels of locomotion. That pattern of successive placebo responses resembles the typical S-shaped development of a Pavlovian conditioned response, albeit the increase described here was quite rapid. The protocol used here may provide an additional method for the experimental analysis of stimulant-induced conditioned placebo activity.