Aims The health hazards and prevalence of ecstasy use have been documented in two decades of research, but no review reporting on potentially modifiable antecedents of use is available. The aim of this study was to integrate systematically research identifying cognitive correlates of ecstasy use. Such research has the potential to identify targets for evidence-based interventions designed to discourage use. Methods The databases PsycINFO and MedLine were searched, inclusion criteria applied to resulting hits, and descendency and ancestry approaches applied to the selected publications. Reported associations between cognitive determinants, including intention to use and ecstasy use measures, were synthesized by calculating a weighted mean effect size, r. Results The pattern of associations lent support both to the theory of planned behaviour (TPB) and the expectancy approach as descriptions of potentially useful determinants. Attitudes were associated most strongly with intention and use, followed by subjective norm and perceived behavioural control. Conclusions Consideration of the strength of associations and the potential modifiability of identified cognitions suggests that evidence-based interventions to discourage ecstasy use should target negative expectancies, perceived behavioural control and anticipated regret, and consider tailoring perceived behavioural control elements.