Retrieval-extinction memory reactivation procedures have been used to prevent the return of learned fear and drug seeking in preclinical models. These procedures first reactivate the original memory with a brief cue exposure (i.e., retrieval) session, and then disrupt memory reconsolidation by conducting extinction training within the reconsolidation window. The original memory is thought to be updated with the new information conveyed by extinction learning, resulting in a persistent therapeutic effect beyond that observed with extinction training alone (i.e., no retrieval). Here, we attempted to replicate the therapeutic effects on cocaine seeking reported by Xue et al. (2012), and extend these findings to nicotine seeking. Rats self-administered either cocaine or nicotine with contingent cues for weeks, and were then divided into two groups. The retrieval group underwent a 10-min retrieval session wherein drug cues were available, but drug was not. Ten minutes later, they were allowed to continue cue extinction training for an additional 60 min. The no retrieval group underwent a contiguous 70-min cue extinction session. These procedures continued for weeks, followed by a test for spontaneous recovery of drug seeking. No group differences were observed on any measure of cocaine seeking, although both groups exhibited extinction and spontaneous recovery. By contrast, for nicotine seeking, the retrieval group exhibited resistance to extinction, an effect that persisted on the spontaneous recovery test. These findings underscore the importance of drug type in the outcome of retrieval-extinction procedures and moreover indicate that retrieval-extinction procedures can be detrimental to nicotine seeking.