Ten studies comparing long-term verbal memory, short term verbal memory and attention in Ecstasy users and non-Ecstasy user controls were examined via meta-analytic process (comparing standardized scores for means and standard deviations, correcting for small samples through a process not described in the paper). Controls were sometimes non-drug users, as in Bhattachary and Powell 2001, sometimes cannabis-users, as in Croft et al. 2001, and sometimes polydrug users, as in Reneman et al. 2001a and 2001b). Studies were selected only if subjects reported abstaining from Ecstasy and other illicit drugs for at least a week prior to study day. When studies featured more than one group of Ecstasy users, comparisons were made between the group with the highest lifetime Ecstasy consumption and non-Ecstasy using controls. In this analysis, the author considered RAVLT immediate recall (verbal list-learning study) as an example of short-term memory, and RAVLT delayed recall an example of long-term memory. Attention was defined as a measure of sustained vigilance. Verbaten also performed weighted regressions, with total lifetime Ecstasy consumption serving as the predictor, and separate weighted regressions with total lifetime cannabis consumption serving as the predictor. Number of null results needed to reduce the power of each meta-analysis was calculated via fail-safe N. Analysis for short-term verbal memory (across 10 studies) found a significant effect size. Fail-safe N was calculated to be 14 (meaning, 14 failures to find results would be sufficient to alter effect size). Neither lifetime Ecstasy consumption (assessed across seven studies) nor lifetime cannabis consumption (assessed in three studies) reduced effect size, indicating that lifetime ecstasy and cannabis consumption did not influence impairment in short-term memory. An analysis of the ten studies examining long-term memory found a significant effect size, and fail-safe N was 9. Weighted regression with lifetime Ecstasy consumption (in ten studies) did not reduce effect size, an indication that reduced long-term verbal memory was not influenced by lifetime Ecstasy consumption. However, a weighted regression with lifetime cannabis use (in four studies) found that the effect size was no longer significant after taking lifetime cannabis use into account, so that reduced long-term memory in Ecstasy users may be due at least in part the effects of lifetime cannabis use. While meta-analysis of sustained attention (in eight studies) also found a significant effect size, a fail-safe N of 5 suggests that that this difference between Ecstasy users and non-Ecstasy users inconclusive or countered by non-published findings. Weighted regression taking lifetime Ecstasy consumption into account (six studies) and lifetime cannabis consumption into account (two studies) failed to find any change in effect size due to either of these variables. Meta-analysis of performance (number of errors), across eight studies, found a significant effect size, but a fail-safe N of 3 strongly suggests that these findings are inconclusive or less definite. Weighted regressions examining effects of lifetime Ecstasy consumption (across seven studies) or cannabis use (across three studies) found no indications that either lifetime Ecstasy consumption or lifetime cannabis consumption influenced performance. Findings were interpreted to indicate that regular Ecstasy use affects short-term verbal recall, and may affect long-term verbal recall, though lifetime cannabis use is at least partly and perhaps mostly responsible for long-term memory differences in the studies analyzed. With respect to attention and performance, regular Ecstasy use may affect these areas, but the number of contradictory findings needed to invalidate effect size were low enough to be affected by unpublished studies reporting on effect. The author found a stepwise relationship between lifetime Ecstasy consumption and short term and long-term memory, and not a continuous relationship. Despite finding a stepwise relationship, the author does not believe that a single exposure to Ecstasy reduces verbal memory, since no studies have assessed effects of single doses, and since lifetime Ecstasy consumption was not associated with likelihood of finding an effect (for example, comparing Croft et al. 2001, Verkes et al. 2001). Hence the author proposes the existence of a "threshold" dose or exposure that impairs short-term verbal memory once it is crossed. Ecstasy users in the studies examined had often taken more amphetamine and cocaine than non-user controls, as noted in data presented in this report, and so it is possible that findings attributed to Ecstasy use may instead be partly or wholly due to amphetamine or cocaine use. To date, this is the only published attempt to perform a meta-analysis on studies of neurocognitive function in Ecstasy users. (A review using a similar means of comparison, but without using a similar means of study selection, can be found in Baggott et al. 2001 . All comparisons used the test with the largest indicator of group difference, and comparisons were made between Ecstasy users and non-user controls). There are unacknowledged differences in studies examined in this report, including failure to blind examiners to subject condition (Rodgers et al. 2000) and reporting data from the same or similar samples (Reneman 2001a and 2001b). Meta-analytic processes assume no systematic variation in publication, yet it is very likely that a systematic bias exists for publishing findings of impaired cognition in Ecstasy users, considering the desire to highlight negative effects of psychoactives along with a general bias for publishing significant findings. Lastly, at least four very recent publications (Curran et al. 2003; Daumann et al. 2000; Simon et al. 2002; Thomasius et al. 2003) have failed to find memory differences in current Ecstasy users. Two of three studies comparing current Ecstasy users with former users found differences in former users only (Curran et al. 2003; Thomasius et al. 2003). These findings were not included in the meta-analysis, and suggest that the relationship between repeated Ecstasy use and impaired verbal memory remains complex and inconclusive. Nevertheless, this paper is valuable as a preliminary attempt at a quantitative review rather than a review based on summarizing study findings.