MDMA and brain activity during neurocognitive performance: An overview of neuroimaging studies with abstinent ‘Ecstasy’ users

MDMA (3,4-methylenedioxymethamphetamine, commonly 'Ecstasy') has returned to prominence with rising prevalence among young adults and higher-strength tablets/powders in recent years. Preclinical studies have shown that repeated high-dose MDMA exposure reduces central serotonin function and may cause loss of serotonergic terminals; early human molecular imaging likewise reported reduced serotonin transporter (SERT) availability in several brain regions. At the same time, human neuropsychological studies report inconsistent neurocognitive findings: some show deficits (notably in verbal memory and other aspects of memory/executive function) while others do not, and it remains uncertain which factors (dose, polydrug use, abstinence duration, task selection) determine observed impairments. Roberts and colleagues set out to review empirical human neuroimaging studies that examined brain activity during neurocognitive performance in abstinent recreational MDMA/ecstasy users. The review covers molecular imaging (PET, SPECT), haemodynamic functional imaging (fMRI, fNIRS), and electrophysiological measures (EEG/ERPs, sensory-evoked potentials), with the goal of relating imaging-indexed serotonergic or metabolic changes to performance on cognitive tasks and to identify common patterns and gaps in the literature.

The paper is presented as an overview of empirical neuroimaging research linking MDMA/ecstasy exposure to brain activity during cognitive tasks. The extracted text does not clearly report a formal methods section: it does not specify search strategies, databases, date ranges, exact inclusion/exclusion criteria, or a fixed number of included studies, so it appears to be a narrative rather than a fully systematic review or quantitative meta-analysis. Instead, the authors organise the review by imaging modality and cognitive domain. They summarise molecular imaging studies that assessed SERT or 5-HT2A receptors and metabolic activity, functional haemodynamic studies using fMRI and fNIRS during tasks such as n-back working memory, associative memory, response inhibition and semantic tasks, and electrophysiological studies using EEG (including loudness dependence of auditory evoked potentials, event-related potentials and visual/ auditory evoked potentials). Many primary studies compared abstinent MDMA users to healthy controls and sometimes to polydrug control groups; a minority of studies included prospective or longitudinal designs, biochemical confirmation of recent use (hair analysis), or multiple confounder controls. The review highlights methodological heterogeneity across primary studies (sample sizes, polydrug composition, task selection, dose metrics) but does not present pooled statistical analyses or a formal risk-of-bias assessment in the extracted text.

Molecular imaging: Multiple PET and SPECT studies reported reduced SERT availability in current MDMA users relative to controls, particularly in cortical regions and hippocampus. For example, one PET study using [11C]DASB found significantly lower SERT binding in occipital, parietal, temporal cortices, anterior and posterior cingulate, dorsolateral prefrontal cortex (DLPFC) and hippocampus, with a negative correlation between hippocampal SERT and duration of use. A larger study of 49 moderate users and 50 controls found SERT reductions across the cerebral cortex and hippocampus, with poorer performance on complex neurocognitive tests (e.g. Trail Making Test-B, California Verbal Learning) correlating with lower SERT in insular cortex and hippocampus. Resting FDG-PET in 19 male MDMA users showed decreased regional cerebral glucose metabolism in bilateral DLPFC and inferior parietal cortex, bilateral thalamus, right hippocampus, right precuneus, right cerebellum and pons; verbal learning and delayed recall correlated with lower metabolism in frontal and parietal regions, and lifetime MDMA dose was associated with lower metabolism in several frontal and temporal regions. Overall, molecular imaging findings are consistent in showing serotonergic and metabolic alterations following repeated MDMA exposure, and several studies report associations between these markers and memory performance, though correlations are not universally observed. fMRI: Functional magnetic resonance imaging studies produced a mixed but recurrent pattern: many investigations reported equivalent behavioural performance between MDMA users and controls, yet altered BOLD responses in users. Several studies found increased activation (hyperactivation) in prefrontal, parietal and hippocampal regions during working memory, inhibition and recognition tasks, which investigators interpreted as compensatory recruitment or reduced neural efficiency. For example, some reports noted greater BOLD in left medial/superior frontal gyri, left thalamus and right hippocampal gyrus during memory tasks. Conversely, other studies—especially at higher task difficulty—observed reduced activation in regions such as the angular gyrus, inferior temporal gyrus and striate cortex in MDMA users. A longitudinal n-back study reported increased parietal activation over 18 months in continuing users (but not abstainers), with the increase correlating with higher one-night MDMA doses, suggesting dose-related change over time. Hippocampal findings were particularly inconsistent: some studies showed elevated hippocampal activation with equivalent performance, others showed reduced hippocampal activity during episodic or associative memory tasks; a prospective study of minimal-exposure users reported decreases in left parahippocampal encoding activity at 12 months in those who continued ecstasy use, correlated with interim ecstasy but not amphetamine use. Sample sizes were frequently small and several studies included polydrug users, complicating attribution to MDMA alone. fNIRS: Only three fNIRS studies are reported. Two showed increased oxygenated haemoglobin (oxy‑Hb) responses over prefrontal cortex voxels in ecstasy users during inhibitory control and word fluency tasks, despite equivalent performance; oxy‑Hb increases correlated with ecstasy use indices (lifetime dose, recent use, frequency) after controlling for alcohol and cannabis. One study reported lower oxy‑Hb in left and right DLPFC during a multitasking stressor in ecstasy polydrug users compared with drug‑naive controls. Authors generally interpret the common findings of increased oxy‑Hb as evidence of greater cognitive effort or compensatory recruitment in serotonin‑rich prefrontal regions. EEG / evoked potentials / ERPs: Studies of the loudness dependence of auditory evoked potentials (LDAEP), an index sometimes related to central serotonergic function, found stronger intensity dependence in currently abstinent long‑term MDMA users than in cannabis users or drug‑naive controls; in one study the LDAEP magnitude correlated with cumulative lifetime MDMA use. Visual evoked potential studies reported reduced intermediate (P200) and late (P300) component amplitudes in heavy users and lower P300 amplitudes even in moderate users, with early components preserved. ERP work during cognitive tasks commonly revealed altered component amplitudes or latencies (e.g. smaller central P300, altered N2 or P200) in MDMA users while behavioural performance remained similar to controls. Some ERP findings correlated with use intensity or with other behavioural traits (for example, higher aggression scores correlated with pronounced intensity dependence), but results across studies were heterogeneous: some reported no P300 differences, others reported reductions or increases, and many studies were small and included polydrug populations. Across modalities, a recurrent empirical pattern is altered neuronal activation in MDMA users—often increased haemodynamic or electrophysiological responses despite preserved task performance—interpreted by primary authors as compensatory recruitment. However, the directionality and regional specificity of effects varies across studies, and many studies are limited by small samples, polydrug use confounds, heterogeneous task selection, and variable dose metrics (e.g. lifetime episodes vs nightly dose).

Roberts and colleagues interpret the converging molecular imaging evidence as indicating consistent alterations of the serotonin system following recreational MDMA use, with many studies reporting reduced SERT availability in cortical and hippocampal regions. Functional imaging and electrophysiological studies frequently show altered neural responses during cognitive tasks—often increased haemodynamic or ERP amplitudes in prefrontal and temporal regions—despite equivalent task performance, which is taken to reflect compensatory recruitment or reduced neural efficiency. The authors emphasise that these functional changes are broadly consistent with serotonergic adaptations observed in molecular imaging, though the extracted text notes that existing data cannot resolve whether changes represent irreversible neurotoxicity or reversible neuroadaptation. Key limitations highlighted by the review include small and heterogeneous samples across studies, frequent inclusion of polydrug users, inconsistent dose metrics (lifetime episodes versus nightly dose), variability in cognitive tasks (many tasks may not target domains most sensitive to MDMA effects), and a relative paucity of studies that combine functional imaging with direct markers of central serotonin function. The review also notes inconsistencies in the direction of imaging responses (increased versus decreased activation) and limited investigation of domains most robustly impaired in MDMA users (for example delayed verbal recall). For future research the authors recommend better‑powered studies with more homogeneous user samples, use of function‑specific cognitive tasks that reliably elicit deficits observed in behavioural studies (such as declarative and prospective memory), and multimodal approaches that pair functional neuroimaging with molecular imaging, genetics or pharmacological challenge of the serotonergic system to clarify mechanistic links. They also suggest prospective designs to disentangle pre‑existing differences from drug‑related change and to examine dose‑response relationships.

The review concludes that chronic recreational MDMA/ecstasy use is associated with haemodynamic and electrophysiological changes during neurocognitive performance that often reflect recruitment of additional neural resources to maintain comparable task performance. These functional alterations align with molecular imaging evidence of reduced serotonin signalling after repeated MDMA exposure, yet current data do not answer whether the changes reflect neurotoxicity or adaptive neuroplasticity. Methodological heterogeneity and small samples limit the strength of conclusions; the authors call for future multimodal, better‑powered studies and note the public‑health relevance given increasing MDMA strength, recommending harm‑reduction efforts targeted at novice users.