Rationale/Objectives: Recent theoretical models suggest that the central executive may not be a unified structure. The present study explored the nature of central executive deficits in ecstasy users. Methods: In Study One, 27 ecstasy users and 34 nonusers were assessed using tasks to tap memory updating (computation span; letter updating) and access to long-term memory (a semantic fluency test and the Chicago word fluency test). In Study Two, 51 ecstasy users and 42 nonusers completed tasks that assess mental set switching (number/letter and plus/minus) and inhibition (random letter generation). Results: MANOVA revealed that ecstasy users performed worse on both tasks used to assess memory updating, and on tasks to assess access to long-term memory (C-and S-letter fluency). However, notwithstanding the significant ecstasy-group related effects, indices of cocaine and cannabis use were also significantly correlated with most of the executive measures. Unexpectedly, in Study Two, ecstasy users performed significantly better on the inhibition task producing more letters than nonusers. No group differences were observed on the switching tasks. Correlations between indices of ecstasy use and number of letters produced were significant. Conclusions: The present study provides further support for ecstasy/polydrug related deficits in memory updating and in access to long-term memory. The surplus evident on the inhibition task should be treated with some caution as this was limited to a single measure and has not been supported by our previous work.