The Approved Suffering Protocol

I Introduction: The Instrument and the Therapy

In 2023, the FDA rejected MDMA-assisted therapy for PTSD despite Phase 3 clinical trial data showing 67% of participants no longer met diagnostic criteria for PTSD after treatment, compared to 32% in the placebo group. For context: the FDA's own approval threshold for psychiatric medications is typically an effect size that would produce far smaller numerical differences. Antidepressants have been approved on response rates that independent meta-analyses have characterized as clinically indistinguishable from placebo in mild-to-moderate depression — the specific evidence for this is documented in Section 3.3, which reviews the Kirsch et al. and Cipriani et al. meta-analyses in detail.

The FDA's stated reason for rejection was not insufficient efficacy. It was that the trial design was flawed — specifically, that participants could not be reliably blinded to whether they had received MDMA or placebo, and that the therapist's role in the therapy made isolation of the compound's effect impossible. These are real methodological concerns. They are also, this paper argues, a description of the instrument's limitations rather than the therapy's failure.

The RCT, as currently constructed and enforced by regulatory bodies, assumes that the active element of a therapy is a discrete compound whose effect can be isolated from context, relationship, and subjective experience. This assumption is valid for aspirin. It is not valid for a therapy whose mechanism of action is neuroplasticity enhancement during a guided experiential process. The instrument cannot reach the phenomenon. The phenomenon is not, therefore, evaluated — it is excluded by design.

This is the Engineered Incompetence pattern. The RCT instrument was built for a specific class of interventions. It produced genuine value within that class — it screened out thalidomide, established dose-response curves, caught false positives before they harmed populations. The regulatory infrastructure grew around it. Careers and institutions were built on it. And then a different class of intervention arrived — one whose mechanism is experiential rather than pharmacological in the narrow sense — and the instrument could not evaluate it. The regulatory body did not adapt the instrument. It rejected the therapy.

Fifty years of Schedule I classification meant that the research to develop better instruments was itself made illegal. By the time the therapy returned to clinical investigation, the regulatory framework had calcified around the old instrument. The result: a treatment with some of the strongest efficacy data in psychiatric history cannot reach the patients it would help, while the approved treatments for the same conditions have effect sizes that meta-analyses have characterized as marginal.

II The RCT: What It Was Built For

2.1 The Design Logic

The randomized controlled trial was developed as a solution to a specific problem: how do you know whether a treatment's effects are caused by the treatment itself, rather than by expectation, regression to the mean, natural recovery, or confounding variables? The answer — brilliant in its simplicity — is to randomly assign participants to treatment and control groups, blind both participants and evaluators to group assignment, and measure outcomes with standardized instruments. Any difference between groups, at sufficient statistical power, can be attributed to the treatment.

This design makes two foundational assumptions. First: that the active element of treatment can be isolated from context. Second: that meaningful blinding is achievable — that participants cannot reliably determine whether they received the active treatment or the placebo. Both assumptions hold for most pharmacological interventions. A participant taking a blood pressure medication cannot tell from their subjective experience whether they have the active compound or the sugar pill. The design is sound for the phenomenon it was designed to probe.

2.2 Where the Assumptions Break Down

Psychedelic-assisted therapy violates both foundational assumptions, structurally and unavoidably.

On isolation: the therapy's mechanism is not separable from the experience it produces. MDMA-assisted therapy for PTSD, as conducted in the MAPS Phase 3 trials, involves three preparation sessions, two eight-hour MDMA sessions conducted with two trained therapists present, and three integration sessions. The therapists are not administering a compound and observing outcomes. They are co-creating the conditions for a specific type of experiential processing — trauma recontextualization within a state of reduced fear response and enhanced prosocial bonding. The compound does not produce the therapeutic outcome alone. The compound, the relationship, the setting, and the integration together produce it. Isolating the compound removes the therapy.

On blinding: MDMA at therapeutic doses produces unmistakable physiological and psychological effects — elevated heart rate, pupil dilation, heightened emotional sensitivity, feelings of closeness and empathy. The placebo condition produces none of these. In the MAPS Phase 3 trials, 80–90% of participants correctly identified whether they had received MDMA or placebo. This is not a trial design failure. It is a property of the compound. You cannot blind participants to whether they have experienced a profound alteration of consciousness.

The FDA's rejection criteria — that blinding was compromised and therapist effects could not be isolated — accurately describe properties of the therapy. They do not describe flaws in the trial design that a better-designed trial could fix. A better-designed trial would have the same properties, because those properties are features of the therapy being evaluated, not artifacts of the trial design.

III The Efficacy Data the Instrument Cannot Capture

3.1 MDMA-Assisted Therapy for PTSD

The MAPS Phase 3 trial is the most rigorously conducted study of MDMA-assisted therapy to date. It enrolled 90 participants with severe, treatment-resistant PTSD — many of them military veterans and first responders for whom existing treatments had failed. The results:

Outcome Measure

MDMA Group

Placebo Group

Significance

No longer meets PTSD criteria (CAPS-5)

67%

32%

p < 0.001

Functional impairment loss (SDS)

71%

48%

p < 0.05

Clinically significant improvement

88%

60%

p < 0.001

Serious adverse events

No drug-related SAEs

—

Safety confirmed

The 67% vs. 32% remission gap is not a marginal effect. It is one of the largest effect sizes ever recorded in a psychiatric clinical trial for a severe, treatment-resistant condition. For comparison: the approval threshold that has been applied to SSRIs for depression involves response rate differences that have been measured in single-digit percentages in independent meta-analyses.

3.2 Psilocybin for Treatment-Resistant Depression

Multiple independent research groups have published psilocybin data for treatment-resistant depression — patients for whom two or more adequate antidepressant trials have failed:

Study

Institution

Response Rate

Duration of Effect

Carhart-Harris et al. (2021)

Imperial College London

57% remission at 3 weeks

Durable at 6-month follow-up

Davis et al. (2021)

Johns Hopkins

71% response at 4 weeks

Maintained at 12 months in 54%

Compass Pathways Phase 2b (2022)

25mg dose group

29% remission at 3 weeks

Superior to 1mg and 10mg doses; Phase 3 results pending — replication challenges reported as of 2024

SSRI comparison (Carhart-Harris 2022)

Imperial College

Psilocybin non-inferior at 6 weeks

Better sustained response profile

Treatment-resistant depression by definition has not responded to the treatments the current regulatory apparatus approves. These patients are offered the approved treatments, which have not worked for them, and are not offered the treatments for which the evidence base — while imperfect by RCT standards — is the strongest available. The instrument determines what counts as evidence. The patients bear the cost.

3.3 What the Approved Treatments Actually Show

The epistemic context matters. The treatments that have successfully navigated the RCT approval pathway for depression and PTSD are not as effective as the approved label implies. A series of meta-analyses has challenged the clinical significance of SSRI efficacy:

Kirsch et al. (2008, PLoS Medicine): meta-analysis of FDA-submitted data found that SSRIs outperform placebo on the Hamilton Depression Scale by an average of 1.8 points out of 52. The clinical significance threshold is generally set at 3 points. For mild-to-moderate depression, the mean drug-placebo difference fell below this threshold.

Cipriani et al. (2018, The Lancet): the largest antidepressant meta-analysis ever conducted (522 trials, 116,000+ participants) confirmed that SSRIs work better than placebo — but found substantial variation in effect size and noted that for individual patients, predicting who will respond remains impossible.

Hengartner & Plöderl (2018): found that when active placebos are used — placebos that mimic side effects, thus preserving blinding — SSRI effect sizes approach zero for most patients.

The takeaway is not that SSRIs are useless. Some patients benefit substantially. The takeaway is that the regulatory framework has approved treatments with modest average effect sizes while blocking evaluation of treatments with effect sizes that dwarf them — because the approved treatments fit the instrument and the novel treatments do not.

IV. The Fifty-Year Gap: Legislative Capture as Instrument Lock-In

4.1 The Timeline

Psychedelic research did not fail in the 1960s and get shelved because the evidence was negative. It was legislatively terminated before the evidence could accumulate. The timeline is specific:

Year

Event

1943

Albert Hofmann synthesizes LSD; early psychiatric research begins in Switzerland and Canada

1950s–60s

Hundreds of clinical studies on LSD and psilocybin. Bill Wilson (AA founder) advocates LSD for alcoholism treatment. Positive results for anxiety, depression, addiction.

1965–68

Media-driven moral panic around LSD. Nixon administration begins anti-drug campaign. Research defunded at federal level.

1970

Controlled Substances Act. LSD and psilocybin classified Schedule I: 'no accepted medical use, high potential for abuse.' Research effectively halted.

1970–2000

Thirty-year research gap. Institutional knowledge lost. A generation of researchers never trained in the methodology. Patients treated with approved alternatives.

2000

FDA authorizes first modern psilocybin study (Johns Hopkins). Research resumes under intensely restricted conditions.

2010s

MAPS Phase 2 MDMA trials. Hopkins psilocybin depression studies. Evidence base begins rebuilding from scratch.

2021–23

MAPS Phase 3 trials complete. FDA issues Breakthrough Therapy Designation, then rejects approval citing trial design concerns.

4.2 The Political Origin of Schedule I

John Ehrlichman, Nixon's domestic policy chief, gave an interview in 2016 — published in Harper's — that has become one of the most cited admissions in drug policy history. His account of the War on Drugs:

EHRLICHMAN, 2016 (HARPER'S MAGAZINE)

'The Nixon campaign in 1968, and the Nixon White House after that, had two enemies: the antiwar left and Black people. We knew we couldn't make it illegal to be either against the war or Black, but by getting the public to associate the hippies with marijuana and Blacks with heroin, and then criminalizing both heavily, we could disrupt those communities. We could arrest their leaders, raid their homes, break up their meetings, and vilify them night after night on the evening news. Did we know we were lying about the drugs? Of course we did.'

Baum conducted this interview with Ehrlichman in 1994. Ehrlichman died in 1999; the notes were published in Harper's in 2016 — twenty-two years after the interview occurred. This transcription gap has been raised by critics. The quote is widely cited and its authenticity has not been formally disputed; Baum is a credentialed journalist with documented access. Readers should weigh this context. Even setting the Ehrlichman quote aside, the documented history of Schedule I's enactment — the simultaneous termination of research funding and assertion of 'no accepted medical use' — establishes the self-fulfilling structure of the classification on its own terms.

Schedule I classification was not a scientific determination. It was a political instrument. The 'no accepted medical use' designation for psilocybin and LSD was asserted at the moment research was being terminated — creating a self-fulfilling definition. Once Schedule I, research requires DEA Schedule I licenses, which are rarely granted and subject to extraordinary restrictions. The classification made the evidence difficult to produce, then cited the absence of evidence as justification for the classification.

This is the Instrument Capture Loop operating at the legislative level. The regulatory instrument was not just inadequate — it was weaponized to prevent the development of alternative instruments. The fifty-year gap is not a story of science proceeding cautiously. It is a story of science being legally prohibited while patients died from inadequately treated PTSD, treatment-resistant depression, and addiction.

V The Instrument Mismatch: Why the RCT Cannot Evaluate This Therapy

5.1 Experience-Dependence as a Structural Incompatibility

The RCT's core assumption — that the active element of treatment is isolable from context — is not a practical limitation waiting for better trial design. It is a categorical boundary. Some therapies are fundamentally experience-dependent: the mechanism of action is the experience itself, and altering the experience to make it RCT-compatible destroys the therapy.

Knee replacement surgery cannot be evaluated with a placebo-controlled RCT. Psychotherapy cannot be meaningfully blinded. Dietary interventions cannot control for all confounders in a free-living population. These are not failures of science — they are properties of the phenomena being studied. The scientific response to experience-dependence is not to insist on the blinded RCT or reject the therapy. It is to develop methodological frameworks appropriate to the phenomenon.

The methodological frameworks exist. They have been used in surgical outcomes research, behavioral therapy research, and complex intervention evaluation for decades. They include:

Active control designs — comparing MDMA-assisted therapy to the best available alternative therapy, rather than to placebo

Expectancy-controlled designs — measuring and controlling for expectancy effects statistically rather than through blinding

Dismantling studies — systematically varying components (compound dose, session length, therapist contact, integration intensity) to identify the active elements

Adaptive platform trials — Bayesian designs that update allocation based on accumulating evidence, suitable for multi-component interventions

SMART designs (Sequential Multiple Assignment Randomized Trials) — developed specifically for multi-stage treatment protocols

None of these designs require abandoning scientific rigor. They require adapting the instrument to the phenomenon. The FDA's position — that MDMA-assisted therapy cannot be approved without a blinded RCT — is not a scientific requirement. It is an institutional requirement masquerading as one.

5.2 The Therapist Effect as Feature, Not Flaw

The FDA's concern about therapist effects — that participants in the active MDMA group received a qualitatively different therapeutic relationship during the eight-hour sessions, making it impossible to attribute outcomes to the compound alone — is real. It is also a description of the therapy's mechanism, not a limitation of the trial.

MDMA produces a state of reduced amygdala reactivity to threat stimuli, enhanced oxytocin-mediated prosocial bonding, and a window of increased neuroplasticity during which traumatic memories can be reprocessed without the normal fear response. The therapist's role is to provide the relational context within which this reprocessing occurs. The compound alone does not produce the therapeutic outcome. The compound plus the relationship plus the integration does.

This is not a confound to be controlled away. It is the therapy. Insisting on isolating the compound from the relationship is equivalent to insisting that a surgeon's skill be isolated from the surgical procedure when evaluating surgical outcomes. The FDA does not require this for surgery. It requires it for psychedelic therapy because the RCT instrument was designed for discrete compounds, not integrated therapeutic modalities.

VI The Human Cost of Instrument Lock-In

6.1 PTSD

Post-traumatic stress disorder affects an estimated 13 million Americans in any given year. The veteran population bears a disproportionate burden: the VA estimates that 11–20% of veterans who served in Iraq and Afghanistan have PTSD in a given year. Suicide rates among veterans with PTSD are substantially elevated compared to the general population.

The current FDA-approved pharmacological treatments for PTSD are two SSRIs: sertraline (Zoloft) and paroxetine (Paxil). Meta-analyses of their efficacy in PTSD show response rates of approximately 50–60%, with remission rates substantially lower. Treatment-resistant PTSD — defined as failing to respond to two adequate SSRI trials — is common and has no approved pharmacological alternative.

The MAPS Phase 3 data show 67% remission in a treatment-resistant population. This is not a marginal improvement over existing options. For patients who have failed the approved treatments, it may be the difference between recovery and permanent disability — or between life and death. These patients are currently denied access to this treatment because the approval instrument cannot evaluate the therapy properly.

6.2 Treatment-Resistant Depression

Major depressive disorder is the leading cause of disability worldwide. An estimated 30% of patients do not respond adequately to first-line antidepressant treatment. Of those who receive adequate trials of multiple antidepressants, approximately 10–20% are classified as treatment-resistant. For treatment-resistant depression, current options include electroconvulsive therapy (ECT), transcranial magnetic stimulation (TMS), and ketamine infusions — all with significant limitations, accessibility barriers, or transient effects.

The psilocybin data for treatment-resistant depression consistently shows response rates exceeding those of available alternatives. The Johns Hopkins study reported 71% response at four weeks. The sustained nature of the effect — with many participants maintaining response at 12-month follow-up after a single or two-session treatment — suggests a mechanism of lasting neural reorganization rather than ongoing pharmacological suppression.

Patients in this category are currently offered treatments with lower efficacy and greater side effect burden than treatments the regulatory instrument cannot approve. The instrument is not protecting them. It is protecting itself.

VII Devil's Advocate: The Case for the Current Framework

Every paper in the Engineered Incompetence series is required to present the strongest possible opposing argument and engage it seriously before responding. The arguments below are the ones this paper's thesis must actually defeat. The Institute's publishing rationale — why these papers are not submitted to mainstream peer review — is documented in Paper 1, Section 6.2 of this series.

7.1 The Safety Argument: The RCT Protects Populations From Harm

The regulatory framework built around the RCT exists for a reason rooted in catastrophe. Thalidomide was prescribed to pregnant women for morning sickness in the late 1950s and early 1960s, causing an estimated 10,000 cases of severe birth defects. The tragedy drove the 1962 Kefauver-Harris Amendment, which required proof of both safety and efficacy before drug approval. The RCT's current dominance is a direct consequence of that lesson.

MDMA, specifically, has documented neurotoxicity at high doses in animal studies. Enthusiastic adoption of MDMA-assisted therapy before the full safety profile is established risks repeating the pattern: a therapy that appears effective in short trials produces harms that only emerge with broader use and longer time horizons. The FDA's caution is not bureaucratic obstruction — it is the correct institutional response to a history of harm caused by premature approval.

Furthermore, MDMA has significant abuse potential. Expanding access through an approval pathway risks diversion to recreational use at a scale that causes societal harm exceeding the therapeutic benefit. The Schedule I classification exists not only as a research barrier but as a legitimate harm-reduction mechanism.

This is the strongest version of the institutional argument and it requires the most careful engagement.

Response to the Safety Argument

The thalidomide argument establishes that regulatory caution has value. It does not establish that the current instrument is correctly calibrated for all therapeutic modalities. Thalidomide's harms were physical and dose-dependent — properties that the RCT framework can detect. The concerns about MDMA-assisted therapy are of a different character.

On neurotoxicity: the animal studies showing MDMA neurotoxicity involved doses many times higher than therapeutic administration and repeated-dose protocols that bear no resemblance to the MAPS approach (two sessions, months apart). The Phase 3 safety data — 90 participants, no drug-related serious adverse events — does not support the neurotoxicity concern at therapeutic doses. A regulatory posture that requires more safety data is reasonable; a regulatory posture that rejects efficacy data while citing safety concerns that the safety data do not confirm is not.

On abuse potential: the therapeutic protocol — two eight-hour sessions with trained therapists, months apart, in a clinical setting — is not a pathway to recreational use. The compound as administered in therapy is a different intervention than the compound consumed recreationally. Approval for therapeutic use does not require Schedule II reclassification to permit recreational use. These are distinct regulatory questions that the FDA conflates in its analysis.

The deeper point: the safety argument is an argument for rigorous evaluation, not for the RCT specifically. The MAPS trials included extensive safety monitoring that the RCT design permitted. The concern about therapist effects and blinding is an efficacy evaluation concern, not a safety concern. The FDA's rejection on these grounds conflates safety rigor (where the RCT contributes) with efficacy evaluation instrument fitness (where it does not).

7.2 The Methodological Slippery Slope Argument

If the FDA creates alternative approval pathways for therapies that argue their mechanism of action is incompatible with the RCT, this creates a category that can be exploited by less scrupulous actors. Homeopathy practitioners, supplement manufacturers, and proponents of discredited therapies would use the same argument — 'our treatment works through mechanisms the RCT cannot evaluate' — to seek approval for interventions with no genuine evidence base. The integrity of the approval system depends on consistent application of a rigorous standard.

This is a serious concern and it is the one that requires the most nuanced response.

Response to the Slippery Slope Argument

The slippery slope argument has empirical weight in regulatory contexts. The response is not to dismiss it but to specify what distinguishes MDMA-assisted therapy from the therapies the argument is actually worried about.

The distinguishing criterion is falsifiable mechanistic specificity. Homeopathy cannot specify a mechanism through which water memory produces clinical outcomes. MDMA-assisted therapy can specify the mechanism through which MDMA enables therapeutic processing: reduced amygdala reactivity to trauma cues, documented in fMRI studies; enhanced oxytocin release, measured in serum; increased neuroplasticity in the hours following administration, evidenced by enhanced BDNF signaling. The mechanism is specific, falsifiable, and independently supported by the basic neuroscience literature.

A regulatory framework that requires mechanistic specificity as the criterion for alternative evaluation pathways would admit MDMA-assisted therapy and exclude homeopathy. This is not a category that is difficult to define — it is a category that requires definitional precision the FDA has so far avoided providing. The absence of that precision is itself a product of institutional inertia, not genuine methodological uncertainty.

The alternative approval pathway this paper proposes — active control designs with expectancy measurement, dismantling studies, and adaptive platform trials — produces evidence that can be evaluated and potentially falsified. It is not an exemption from scientific scrutiny. It is a different scientific instrument for a different class of phenomenon. Regulatory science has developed these instruments for surgery, behavioral therapy, and device evaluation. The extension to experience-dependent pharmacological interventions is a matter of institutional will, not scientific impossibility.

VIII The Alternative: An Instrument That Matches the Therapy

8.1 A Framework for Experience-Dependent Intervention Evaluation

The goal is not to lower the evidentiary standard. It is to make the evidentiary standard accessible to the class of therapy being evaluated. The following framework preserves rigor while matching the instrument to the phenomenon:

Design Element

Current RCT Requirement

Proposed Alternative

Control condition

Inert placebo

Active control: best available existing therapy (e.g., prolonged exposure therapy for PTSD)

Blinding

Double-blind: participant and evaluator blinded

Expectancy measurement: assess and statistically control for expectancy effects; use outcome evaluators blinded to treatment assignment

Outcome isolation

Attribute outcomes to compound alone

Dismantling design: vary compound dose, therapist contact, integration intensity systematically across arms to identify active components

Trial structure

Fixed protocol

Adaptive platform trial: Bayesian allocation based on accumulating outcomes; permits protocol refinement during trial

Primary outcomes

Symptom scale reduction (CAPS-5, HAM-D)

Symptom scale reduction plus: functional impairment (SDS), quality of life (WHOQOL), sustained response at 12 months

Therapist effects

Controlled out / confound

Measured and reported: therapist fidelity ratings, session quality ratings, therapeutic alliance scales — treated as moderating variables

8.2 Rescheduling as a Research Infrastructure Requirement

Schedule I classification currently requires DEA Schedule I research licenses for any investigation of psilocybin, MDMA, or LSD. These licenses are restrictive, slow to obtain, and limited to specific research sites. The practical effect is that the research infrastructure for psychedelic-assisted therapy exists at approximately ten institutions in the United States, compared to thousands of sites capable of conducting SSRI trials.

Rescheduling MDMA and psilocybin to Schedule II — the same classification as cocaine, fentanyl, and methamphetamine, all of which have accepted medical uses — would not legalize recreational use. It would permit the development of a broader research infrastructure and, eventually, prescribing by licensed providers under appropriate training requirements. The rescheduling conversation is ongoing; the DEA and FDA are engaged in review processes that may produce movement in 2026–2027. This paper notes that rescheduling is a prerequisite for the research infrastructure this proposal requires.

IX Conclusion

The RCT framework has produced genuine value in protecting populations from ineffective and harmful treatments. This paper does not dispute that value. It disputes the application of a specific instrument — designed for single-compound, placebo-blinded, context-independent interventions — to a class of therapy whose mechanism of action is definitionally context-dependent.

The pattern is the Engineered Incompetence pattern. The instrument was built for one class of phenomenon. A different class of phenomenon arrived. The institution did not adapt the instrument. It applied the existing instrument, found that the phenomenon didn't fit, and classified the phenomenon as inadequately evidenced. Then it approved treatments with weaker efficacy data that did fit the instrument. Then it cited the absence of approved treatments as evidence that the evidence base for alternatives was insufficient.

Millions of patients with PTSD, treatment-resistant depression, and other conditions for which current approved treatments have limited efficacy live inside this loop. The fifty-year Schedule I classification — political in origin, self-reinforcing in structure — denied them access to treatments that might have worked. The FDA's 2023 rejection of MDMA-assisted therapy closed the loop: even when the evidence broke through despite the instrument's limitations, the instrument's assumptions were used to reject the evidence.

The instrument needs to change. The alternative framework — active control designs, expectancy measurement, adaptive platform trials — already exists and is already validated in adjacent research contexts. The patients have waited long enough.

References

1. Mitchell, J.M., et al. (2021). MDMA-assisted therapy for severe PTSD. Nature Medicine, 27, 1025–1033. [MAPS Phase 3 trial]
2. Mitchell, J.M., et al. (2023). MDMA-assisted therapy for PTSD: Secondary analyses and data from a Phase 3 randomized, double-blind, placebo-controlled trial. Science Translational Medicine, 15(680).
3. Carhart-Harris, R., et al. (2021). Trial of psilocybin versus escitalopram for depression. New England Journal of Medicine, 384, 1402–1411.
4. Davis, A.K., et al. (2021). Effects of psilocybin-assisted therapy on major depressive disorder. JAMA Psychiatry, 78(5), 481–489.
5. Kirsch, I., et al. (2008). Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Medicine, 5(2), e45.
6. Cipriani, A., et al. (2018). Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder. The Lancet, 391(10128), 1357–1366.
7. Hengartner, M.P., & Plöderl, M. (2018). Statistically significant antidepressant–placebo differences on subjective symptom-rating scales do not prove that the drugs work. Front. Psychiatry, 9, 517.
8. Carhart-Harris, R., et al. (2016). Neural correlates of the psychedelic state as determined by fMRI studies with psilocybin. PNAS, 109(6), 2138–2143.
9. Kuypers, K.P.C. (2019). The therapeutic potential of microdosing psychedelics in depression. Therapeutic Advances in Psychopharmacology, 9, 2045125319880501.
10. Compass Pathways. (2022). COMP360 psilocybin therapy Phase 2b clinical trial results. New England Journal of Medicine, 387, 1637–1648.
11. Ehrlichman, J. (2016). [Interview with Dan Baum.] Harper's Magazine, April 2016. 'Legalize it all: How to win the war on drugs.'
12. Baggott, M.J., et al. (2015). Intimate insight: MDMA changes how people talk about significant others. Journal of Psychopharmacology, 29(6), 669–677. [Mechanism of action]
13. Mithoefer, M.C., et al. (2019). MDMA-assisted psychotherapy for treatment of PTSD: Study design and rationale for Phase 3 trials based on pooled analysis of six Phase 2 randomized controlled trials. Psychopharmacology, 236(9), 2735–2745.
14. VA/DoD Clinical Practice Guideline for PTSD. (2023). Department of Veterans Affairs / Department of Defense. [Current approved treatment landscape]
15. Campbell, G., et al. (2018). The science of adaptive designs in clinical trials. Statistical Methods in Medical Research, 27(1), 3–19. [Alternative trial design framework]
16. U.S. Food and Drug Administration. (2023). Advisory Committee Meeting: MDMA-assisted therapy for PTSD. FDA Briefing Document, June 2023. [FDA rejection basis]

The Institute for Cognitive Sovereignty

Engineered Incompetence Series | Paper 2 | February 2026

Uncomfortable but Rigorous