Understanding How MDMA Works: Insights from Psychoanalytic Therapists' Experiences

Psychedelic-assisted psychotherapy has re-emerged as a highly promising strategy for treating a range of psychiatric disorders [1, 2]. Among psychedelic compounds, the evidence base relating to the entactogen MDMA is particularly advanced, with two Phase III trials supporting its efficacy and safety in post-traumatic stress disorder (PTSD) [3,4]. Yet, despite meeting or exceeding the typical requirements for a treatment to be considered empiricallysupported, the US Food and Drug Administration (FDA) did not approve a new drug application for MDMA-assisted psychotherapy (MDMA-AP) in 2024, citing various limitations of the relevant MDMA-AP trials. Additionally, the role of the non-pharmacological components of MDMA-AP and related psychedelic treatments remains uncertain [5,6]. Existing trial designs – which typically compare MDMA + psychotherapy with placebo + psychotherapy – do not disentangle drug from psychotherapy effects. Moreover, because the FDA does not regulate psychotherapy, a key component of the proposed treatment falls outside of current regulatory processes. Additionally, some researchers have questioned the necessity of co-administered psychotherapy in psychedelic assisted psychotherapies [5]. In the case of MDMA-AP, scepticism regarding the psychotherapy component may be driven, in part, by the relative unfamiliarity of psychotherapy researchers with the inner-directedtherapy model used in the main trials [3,4,7], which draws on components of somatic and internal family systems psychotherapies [8–10]. Although other models of psychotherapy have also been proposed as adjuncts to MDMA treatments, these are often narrowly based around cognitive behavioural conceptualisations of psychopathology [11] and continue to focus on PTSD [cf.12]. Indeed, cognitive behavioural therapy has been proposed as a default mode of therapy in psychedelic-psychotherapy combinations by Yaden and colleagues [13]. These authors’ suggestion that cognitive behavioural (and related) therapies should be considered the default, however, relied on somewhat outmoded caricatures of alternative models of psychotherapy (particularly psychoanalysis). Moreover, their suggestions appear grounded in component-level descriptions of therapeutic strategies and their interactions with psychedelics (e.g., “openness, awareness, and engagement could conceivably potentiate psychedelic sessions” [13], p.7), rather than process-level mechanistic explanations of how MDMA might interact (additively or synergistically) with these components. Such process-focused explanations might be exemplified by statements like “MDMA-induced reductions in defensive avoidance may alter the participant’s epistemic stance – for example, by promoting a more trusting attitude towards their therapist – thus encouraging therapeutic engagement and enabling other change processes, like reconsolidation-updating or extinction of traumatic memories”. In the absence of more precisely specified mechanistic hypotheses of this kind, it is difficult to determine how (or which of) these interactions should be tested experimentally and statistically. As such, we believe the proposal for cognitive behavioural therapy primacy in psychedelic-assisted psychotherapy is, at least, premature.

Current accounts of MDMA’s psychological mechanisms of action focus on descriptive behavioural psychopharmacology, for example, its effects on self-compassion, openness, empathy, and fear responding. While these behavioural or subjective variables may form the components (e.g., mediators) of a mechanistic model of MDMA’s therapeutic effects [14], their temporal dynamics (i.e., causal ordering), relational embedding, and interactions with psychotherapeutic support remain underspecified, making it difficult to formulate discriminating predictions.

To summarise, current descriptions of MDMA-induced psychological phenomena remain theoretically underdeveloped and insufficiently integrated into existing theories of psychological change. The lack of a shared lexicon or validated process measures further contributes to the conceptual gaps between descriptive behavioural-psychopharmacological findings, process-oriented psychotherapeutic studies and clinical trial research. Without a well-developed understanding of MDMA’s mechanisms of action, it is difficult to establish which non-pharmacological elements of MDMA-AP are most essential, how these elements might be prioritised and optimised, and which elements (if any) might be unhelpful or even harmful (including, but not limited to, psychotherapeutic modality, relational stance, exposure to traumatic material, dosing-day support, as well as preparation provision and post-dosing meaning-making). These questions are, of course, based on the assumption that at least some psychotherapy is needed in MDMA treatment, which, as noted above, remains a contentious point in psychedelic medicine research [15].

For this study, we devised an innovative qualitative protocol aimed at developing a theory of MDMA’s therapeutic mechanisms of action by investigating its effects in a specific participant group: psychotherapists with verified credentials in psychoanalytic/psychodynamic psychotherapy. Our participants will have limited past experience with MDMA and will be administered the drug in a controlled research setting under clinical supervision. Due to their training – a compulsory component of which is completion of extensive personal therapy – these participants are likely to be highly skilled in self-reflection, to possess an advanced capacity for describing their own cognitive and affective processes, and to be able to apply theoretical frameworks to understand mental phenomena in general, and psychopathology and its treatment in particular. The detailed qualitative exploration of participants’ first-hand experiences of MDMA will, therefore, provide a credible, clinically informed basis for developing a psychological model of MDMA’s action.

The objectives of this study are, therefore:

To simulate aspects of the use of MDMA in recent clinical trials [1], participants will receive up to 120 mg MDMA in each of two dosing sessions, separated by at least one week. No formal psychotherapeutic procedures (e.g., those intended to direct trauma or experience processing) will be employed.

The study objectives will be achieved through a series of in-depth interviews, narrative journaling, and other psychological assessments before and after the dosing sessions. In addition, brief interviews will be performed during each dosing session. Qualitative data will be analysed using phenomenological approaches and thematic analysis to examine changes in participants’ attitudes, self-experience, emotional processing, and relational patterns across baseline, acute and sub-acute (dosing sessions) phases and follow-up phase. The qualitative findings will be integrated with other data sources (journalling and validated questionnaires) using techniques inspired by grounded theory (such as the constant comparison approach [16]), with a view to producing a theoretically rigorous model of MDMA’s psychological mechanisms of action.

This protocol outlines a qualitative mechanistic study of the entactogen MDMA, which has shown promise in the treatment of PTSD. By combining phenomenological and thematic interviewing, RAP interviews with CCRT and reflective functioning scoring, longitudinal journaling, and psychological questionnaires, the study is designed to support a fine-grained mapping of experiential processes across acute, sub-acute, and integration phases. By purposively sampling trained psychodynamic psychotherapists and leveraging their clinical expertise and self-reflective capacities, we aim to interrogate psychological mechanisms that have been under-researched in neurobiologically focused studies, including changes in self-related processing, interpersonal patterns, reflective functioning, and meaning-making. This use of the first-person accounts of expert clinician-participants as high-fidelity reports for probing MDMA-related internal and relational phenomena is a distinctive feature of the protocol. The intended outcome is a theoretically grounded, testable model of MDMA’s psychological mechanisms of action, providing a bridge between first-person experience and constructs amenable to operationalisation in future clinical and experimental trials. We suggest that this methodological orientation adds value to the current research landscape by enabling a nuanced examination of psychological and interpersonal processes that are difficult to capture using standard quantitative and neuroscientific paradigms.

Several features strengthen the study design. First, sequential, time-anchored interviews during dosing sessions prioritise experience-near description, while the subsequent IPA interview explicitly invites longer-term meaning-making. In this way, both immediate effects and the products of longer-term integration are captured. Second, triangulation across methods (phenomenological interviews, RAP interviews, narrative journals, and descriptive psychometrics) is intended to generate testable hypotheses and strengthen analytic validity. Third, focusing on psychodynamic clinicians maximises the sample’s informationpower [20]: participants’ training in introspection and in communicating complex psychological processes increases the likelihood of detailed, nuanced accounts, thereby enhancing the efficiency and depth of qualitative inference with a modest sample size.

The study also has limitations. The open-label, non-controlled design means that expectancy and demand effects are likely to shape participants’ subjective experiences. Participants’ professional familiarity with psychotherapy processes and their theoretical commitments could both deepen insight into how MDMA achieves its effects and, conversely, constrain interpretation by privileging familiar explanatory frames over novel accounts. Our data is self-reported and is therefore vulnerable to memory and social desirability biases, despite efforts to capture in-the-moment experience. We seek to mitigate these limitations through: (i) phase-specific, idiographic analysis of interviews prior to cross-phase synthesis (2.6.1.A-C); (ii) triangulation across analysts at key stages (2.6.4); (iii) explicit attention to convergence and divergence during constant-comparison synthesis (2.6.4); and (iv) integration of experiential themes with structured indices, including CCRT and reflective functioning ratings (2.6.1.D).

An additional limitation is that our participants will not be encouraged to engage with spontaneously emerging traumatic materials during the dosing sessions. Given that PTSD is the primary indication for MDMA-AP, this may limit the insights we can gain about changes in traumatic memories processing. However, the team considered it ethically problematic to encourage participants to engage with and discuss traumatic materials when they might not, ordinarily, be inclined to do so (i.e., outside of the MDMA state). Equally, avoiding trauma processing will enable us to preserve the interpretability of mechanisms because we will not be conflating drug effects with exposure-based therapeutic processes.

Despite these constraints, we expect the study to contribute substantively to our understanding of MDMA as a therapeutic agent. A qualitative, mechanism-focused account of MDMA’s action may clarify which elements of adjunctive psychotherapy are necessary, for whom, and when, thereby supporting optimisation of psychotherapy principles and protocols. Candidate mechanisms – such as shifts in relational patterns and self–other representations, their hypothesised role, temporal ordering, and contextual conditions – can be formalised into hypotheses for experimental and clinical testing.

Beyond MDMA, the protocol offers a transferable template for qualitative mechanistic examination of other psychedelic compounds, promoting the development of experience-led, evidence-based models of psychedelic-assisted therapies grounded in first-person phenomenology. Moreover, the primary output of this project – a theory (or theories) of psychological change induced by MDMA – may have implications beyond therapeutic refinement. For example, it may generate constructs and hypotheses that may inform the development of novel quantitative tools in behavioural psychopharmacology that draw on psychoanalytic concepts likely to be affected by MDMA (e.g., questionnaires or behavioural tasks assessing loosening of defences and access to previously repressed or suppressed material; ‘corrective emotional experiences’).

Substantive amendments to the protocol will be reported transparently in subsequent publications. Such changes will be data-driven, agreed within the team, and submitted to the Research Ethics Committee for approval. Any pattern of adverse physiological or psychological events exceeding pre-specified thresholds (per our pre-specified procedure) will result in study interruption. In the event of early termination, we will document reasons and analyse accrued data with appropriate caveats. Dissemination plans include conference presentations and publication in peer-reviewed academic journals.