Self-Experiments with Psychoactive Substances: A Historical Perspective

Passie and colleagues situate their chapter within a long but under‑recognised tradition of self‑experimentation (SE) with psychoactive substances by physicians, psychologists and chemists. They note that SEs have been used for more than a century to explore subjective effects, refine therapeutic procedures, gain philosophical insight, and sometimes to compensate for personal deficits. The authors argue that substances producing complex alterations of consciousness (for example classic hallucinogens and entactogens) attracted far more SE activity than drugs with more predictable ‘‘simple’’ effects (for example many stimulants or benzodiazepines), because complex subjective states are harder to describe and therefore especially amenable to first‑person investigation. This chapter aims to provide a historical overview of SEs with selected psychoactive classes (laughing gas/nitrous oxide, cannabis, cocaine, hallucinogens, entactogens, and some dissociatives), to classify types of SEs, to document motives and consequences, and to consider ethical and safety issues. The authors limit the scope to SEs conducted by medically or scientifically trained individuals and emphasise the contrast between controlled SEs performed in clinical settings and the ‘‘wild’’ or informal SEs that became more common with the rise of new psychoactive substances (NPS) from the 1960s onward.

The extracted text presents this work as a historical, narrative review rather than a formal systematic review; the authors do not provide a clear search strategy, database list, or explicit inclusion/exclusion criteria in the provided extract. They state that more than 100 publications describing SEs with psychoactive substances exist and that one prior review had identified 137 medical SEs, but the chapter emphasises qualitative synthesis and illustrative case examples rather than a quantitative meta‑analysis. To manage scope, Passie and colleagues restricted coverage in two ways: they focused on substances that tend to produce ‘‘complex’’ experiential effects (hallucinogens, entactogens, cannabis, some dissociatives and cocaine) and they restricted SEs to those performed by physicians, psychologists or medicinal chemists, excluding most lay, artistic or purely recreational reports. The authors adopt a tripartite typology of SEs—controlled SEs (planned, clinically supervised, documented), uncontrolled SEs (less documented, variable supervision), and ‘‘wild’’ SEs (primarily recreational or exploratory, often informal and undocumented). Much of the chapter is organised historically, with substance‑by‑substance narrative summaries, notable first‑person reports and discussion of later developments such as the emergence of NPS and internet forum reporting. Where relevant, the authors draw on regulatory and toxicological considerations (for example requirements for preclinical toxicology in modern clinical studies) and on examples of analytical work informed by SEs (for instance controlled self‑administration studies that helped characterise the activity of some synthetic cannabinoids). The extract does not report a formal risk‑of‑bias assessment of included reports, nor does it present pre‑specified outcome measures or meta‑analytic methods.

From the historical material synthesised, the authors report that SEs with psychoactive substances became culturally visible from the mid‑eighteenth century and expanded through the nineteenth and early twentieth centuries. Early medical SEs included experiments with nitrous oxide, cannabis and cocaine; systematic SEs of hallucinogens and related compounds proliferated from the late nineteenth century into the twentieth century. Notable first‑person investigators identified in the narrative include Humphrey Davy and Thomas Beddoes (nitrous oxide), Horace Wood and Weir Mitchell (cannabis and mescaline), Sigmund Freud (cocaine), Albert Hofmann (LSD), Alexander Shulgin (many novel phenethylamines and tryptamines), and numerous psychiatrists and psychotherapists who both tested and later used psychedelics in clinical contexts. The authors describe three SE categories and characterise their typical features: controlled SEs tended to be planned, documented, dose‑defined and sometimes supervised by an observer; uncontrolled SEs lacked standardisation and documentation; and ‘‘wild’’ SEs were informal, often recreational, and became widespread with the emergence of NPS. They report that controlled SEs historically yielded rich first‑person descriptions of subjective phenomena—insights into altered perception, ego dissolution, mystical‑type experiences and other complex states—while acknowledging that such reports lack the methodological rigour of blinded, controlled trials. Substance‑specific findings include historical first reports and characteristic experiential patterns: nitrous oxide produced fleeting revelatory and analgesic effects; cannabis produced dose‑dependent alterations in thought, mood and memory with considerable inter‑individual variability; cocaine induced strong euphoria, increased self‑confidence and, in some cases, dependence among investigators; mescaline and LSD generated vivid visual and cognitive alterations and were frequently explored by psychiatrists to model psychosis or for therapeutic insight; DMT produced intense, often entity‑encountering experiences especially after non‑oral routes of administration; entactogens such as MDMA and MDA were first recognised through chemists’ SEs and later adopted by psychotherapists; ketamine and some dissociatives were self‑tested for their dissociative and purportedly transpersonal effects but have been associated with problematic dependence in some investigators. The authors note a shift over time: scientifically rigorous, controlled SEs dominated until the late twentieth century, after which ethical standards and regulatory frameworks curtailed investigator self‑administration, while ‘‘wild’’ SE practices expanded alongside the proliferation of NPS. The review highlights a range of motives for SEs (personal curiosity, methodological preparation for clinical studies, therapeutic training, philosophical inquiry, harm‑reduction testing of unknown substances) and documents harms observed in the literature, including bias in reporting, loss of objectivity, dependence and tragic outcomes such as neurotoxicity from MPTP and cases of ketamine dependence and fatality linked to uncontrolled use. The authors also report that SEs have occasionally provided valuable toxicological and pharmacokinetic information (for example controlled oral testing of a synthetic cannabinoid to assess psychoactivity and metabolism).

Passie and colleagues interpret their review as demonstrating a long and heterogeneous tradition of SEs with psychoactive substances that has yielded meaningful descriptive insights into subjective drug effects and, at times, informed therapeutic practice. They emphasise that SEs were particularly valuable when investigators sought nuanced, first‑person descriptions of complex alterations in perception, emotion and self‑experience that are difficult to capture with objective measures alone. At the same time, the authors stress the scientific and ethical limitations of SEs: single‑subject reports are vulnerable to bias, lack blinding and control, and therefore cannot substitute for randomized, double‑blind, placebo‑controlled trials when estimating average effects or safety profiles. Historical changes in research ethics—most notably post‑Nuremberg principles and modern institutional review processes—have reduced investigator self‑administration in formal research settings. The authors warn that ‘‘wild’’ SEs, especially those involving novel NPS and internet‑mediated anecdotal reporting, pose public‑health concerns because of unknown toxicology, variable purity, and lack of clinical oversight. They identify a typology of motivations (exploratory versus compensatory) and argue that substances differ in the likelihood of provoking problematic investigator behaviour: drugs that produce reliable euphoria and reward (for example cocaine, some amphetamines, certain dissociatives) appear more likely to generate dependence and loss of control among some experimenters, whereas ‘‘consciousness‑expanding’’ compounds were less often associated with repeated abuse by investigators. The authors acknowledge that their review is selective by design—the chapter excludes many lay or artistic self‑experiments and focuses on medically framed SEs—and they recommend that, where SEs are deemed necessary, controlled and well documented procedures with adequate toxicological data and ethical oversight are preferable. Finally, they suggest a limited future role for carefully controlled SEs, for example in therapist training for psychedelic‑assisted therapies, while cautioning that the rapid growth of NPS and informal online reporting will likely drive unsafe ‘‘wild’’ SEs outside scientific governance.

The authors conclude that self‑experimentation with psychoactive substances has a continuous history of more than 125 years and has contributed both scientific and therapeutic insights as well as cautionary lessons. While controlled SEs historically yielded rich first‑person data that informed study design and therapeutic practice, methodological and ethical shortcomings mean they cannot replace controlled clinical trials. Given modern ethical requirements and the risks posed by NPS and informal ‘‘wild’’ SEs, Passie and colleagues advocate that any necessary future SEs be tightly controlled, underpinned by toxicological data and institutional review, and limited in scope—for example, as part of clinician training for psychedelic therapies—while recognising the public‑health dangers of uncontrolled experimentation outside scientific oversight.