On The Promise and Paradox of Non-Psychedelic Psychedelics

Arthur Juliani
8 min readJun 14


Psychedelic therapy is emerging as a promising new treatment for depression, anxiety, addiction, and other mental health issues [1]. Many proponents are even envisioning a world in which psychedelic therapy, which is administered once or twice over the course of a month would replace the current status-quo of taking an SSRI daily over the course of months or years. For this to happen psychedelic therapy needs to be broadly accessible — both financially and logistically. Unfortunately that isn’t yet the case. What a typical psychedelic therapy treatment consists of today is often a preliminary introductory session with a therapist on the first day followed by a second day of taking the drug and having the “trip” experience under the supervision of a therapist which is then followed by a third day of working again with a therapist to integrate the insights gained from the experience into one’s life. In most instances there is not one but two trained therapists present each day. While this treatment is often deeply meaningful for those that go through it [2], and increasingly being born out as being effective [3], it is also quite expensive to administer [4]. This has led many to seek out ways to provide the benefits of psychedelic therapy without the psychedelic therapy. A seeming contradiction which actually has more promise than you might at first think.

A drug which has all the remarkable antidepressant effects of psychedelic therapy without the bothersome “hallucinations” is a tantalizing prospect, especially for pharmaceutical companies and the healthcare industry. Such a treatment would enable them to provide significantly less expensive and more straightforward treatment for depression and anxiety than what is required for traditional psychedelic therapy. It is likewise an appealing prospect for individuals who might not feel comfortable with or psychologically healthy enough to do the heavy-duty internal work that often accompanies a psychedelic trip. It may also be desirable for individuals with family or personal histories of psychosis who may otherwise be prevented from participating in traditional psychedelic therapy. This kind of innovation would allow for psychedelic treatment to scale in a way which is difficult under the current paradigm. The question of course is whether the development of such a drug is possible.

One good reason to think that developing such a drug is possible is the fact that a number of research labs have claimed to have done just that. The development of these drugs follows a relatively common blueprint: a research group starts with a known psychedelic drug such as psilocybin or LSD. They then modify the molecular structure of the drug in order to remove the “hallucinogenic” effect of the drug. This most commonly involves changing the drug’s action at the serotonin 5-HT2A receptor, which is believed to be the main mechanism by which the subjective effects of psychedelics take place. Once a candidate drug has been created, it is typically verified in rodent models. This is done by looking for the “head twitch response” which in classical psychedelics is normally a sign that the drug they have ingested is psychedelic [5]. If there is no “head twitch” then the drug is presumed to be non-hallucinogenic.

After a drug has been verified as likely not hallucinogenic, it is then tested for its potential antidepressant effects. Here again, rather than testing directly in humans, rodents are often used as a model animal. This is not ideal, given that depression in humans means something quite different than depression in a rodent, but there are some useful analogues. Effects consistent with resolving symptoms of depression are measured either in the rodent’s ability to perform various depression-analogue tasks such as the forced-swim task [6], or in its ability to induce neuroplasticity in the brain of the animal, measured by growth of new dendrites or dendritic spines on neurons [7]. If a reasonable subset of these properties is present (or ideally all of them), the drug is declared a potential non-psychedelic psychedelic and a candidate for clinical investigation. A number of such drugs have been identified in the past few years, including Tabernanthalog [8], 2-Br-LSD [9], and others.

Why does it seem possible to dissociate (at least in rodents) the subjective effects of psychedelics from their antidepressant effects? A paper came out last week in the journal Nature Neuroscience which provides some valuable clues [10]. The paper describes the discovery that psychedelics such as LSD and psilocin seem to bind directly to a receptor in the brain which is essential for neurogenesis: TrkB. The authors of that paper demonstrated that even in the absence of 5-HT2A agonism (the mechanism responsible for the “hallucinations” and other subjective effects), as long as the TrkB binding took place there was neurogenesis in the target cell. Compared to SSRIs, which also bind to TrkB, psilocin and LSD had 1000x greater affinity for the receptor. The authors hypothesize that this mechanism may be responsible for the rapid and long-lasting antidepressant effects of psychedelics.

So it seems that with the discovery of psychedelics binding to TrkB there is a clear path forward for the groups interested in developing non-psychedelic psychedelics. If you assume that the antidepressant effects of psychedelics are the result of their ability to induce neurogenesis, and consider the fact that the “hallucinogenic” effect of psychedelics seems to be mediated by a completely different molecular pathway than the neurogenesis-inducing effects, it seems hypothetically possible to develop a drug which only acts to induce neurogenesis, and thus only provides the antidepressant effect without the “trip.” That foundational assumption — that the antidepressant effect of psychedelics is only a result of neurogenesis — is a very big one though, and has been challenged by a number of individuals on various grounds.

One challenge to the causal link between neurogenesis and the antidepressant effect of psychedelic therapy is that the timelines don’t totally line up. In rodents the actual growth of dendrites and dendritic spines in neurons is observed between six and twenty four hours after administration of the drug [11]. This is in contrast to the rapid antidepressant effects which can onset almost immediately after the acute effects of the drug have subsided [12]. If you were going purely based on temporal order, it would make just as much sense to propose that the antidepressant effects are what induce the neurogenesis, rather than the other way around. While this likely isn’t the case, there is clearly a different mechanism at work which is underlying the change in mood separate from the induced neurogenesis.

The strongest evidence against the “neurogenesis is all you need” perspective is the fact that the strength of the clinical benefits for people undergoing psychedelic therapy seem to be strongly correlated with the actual subjective experience that they have while on the drug [13]. This was famously demonstrated in the correlation between ratings of mystical experiences and later therapeutic outcomes [14]. It wasn’t a matter of having an intense experience, it was about having a particular kind of meaningful experience which forced the participants to reevaluate some previously held beliefs about themselves and their place in the world. The entire REBUS and CANAL models of psychedelic therapy are based around this idea that these drugs work largely by allowing for the disruption and revision of previously held beliefs [15]. It doesn’t seem likely that a drug which has no subjective effect would be capable of occasioning such a revision of belief. What then is going on here?

The evidence suggests that in reality both aspects of psychedelic drug’s mechanism of action are necessary but not alone sufficient to provide the therapeutic effect which seems so promising in clinical trials. If you only have the acute subjective effect (the “trip”) and corresponding brief window of belief disruption then you might be able to see the world and yourself in new and healing ways. If those revised beliefs aren’t accompanied by neurogenesis then they will likely be lost or not be able to be integrated deeply enough into one’s daily life to make a meaningful difference. On the other hand, simply increasing the plasticity of an individual’s brain without providing the experiential context in which the right set of revised beliefs can be explored, learned, and integrated is likewise of diminished value. In a recent preprint myself and my collaborators argue that these two effects can be understood as operating on different underlying optimization landscapes, one of inference and the other of learning [16].

What evidence is there for the lack of efficacy when only one of these two mechanisms of psychedelic therapy is present? While totally anecdotal, it is interesting to note that people who have performed self-experimentation with the novel non-psychedelic psychedelic Tabernanthalog report it not being a particularly beneficial drug, even when taken over the course of multiple weeks [17]. This is in stark contrast to the impressive effects of the drug when administered to rodents. It may be that something like a mystical-type experience is indeed a missing ingredient. We can also look at the efficacy of SSRIs, which also act to increase neurogenesis through TrkB binding. Here again we find that SSRIs alone are less effective than SSRIs paired with psychotherapy, which serves to provide a context for belief examination and revision [18]. While I am not aware of studies which have done this, it would also be interesting to explore blocking TrkB binding in psilocybin or LSD and examine whether there is still an antidepressant effect. Doing so would confirm whether the inverse also holds: that the mystical experience without the increase in neuroplasticity is diminished in its therapeutic effect.

Given all of this, it seems unlikely that drugs which simply increase neurogenesis but have no phenomenological effects will be usher in a new era of treatment which is dramatically different from the current status quo of SSRIs plus psychotherapy. At best they may provide a kind of marginal benefit by having greater treatment efficacy than SSRIs (psychedelics do seem to induce neurogenesis much more reliably) while also having a less discouraging side-effect profile. If this is all that comes from this line of research it will still be a worthwhile endeavor as SSRIs are clearly not close to the optimal treatment for either depression or anxiety disorders. There is a greater potential though that psychedelics offer, and there are ways in which the classic psychedelics might still be improved upon without throwing the baby out with the bathwater.

A valuable place to look for guidance is the experiments of Alexander Shulgin, a chemist who in the 1970s synthesized over one hundred novel psychedelics based on the mescaline molecule [19]. A number of these retain the psychedelic effects, but alter them in ways which have been described as being “more gentle” or “more warm.” A well-known example is the psychedelic drug 2C-B which has often been compared to MDMA for its empathogenic effects [20]. Another lesser known drug is Ariadne, which does not produce classical psychedelic effects, but still provides phenomenally noticeable improvements in mood and motivation to those who take it [21].

Shulgin was only able to explore a small part of the entire state space of possible psychedelic-adjacent drugs. Modern drug discovery techniques make it possible to both explore that space more broadly and in a more targeted manner. One desirable direction I would like to see is toward a class of drugs which still induces the necessary acute psychedelic effects to occasion belief revision, but does so in a way that provides a kind of “no bad trip” guarantee. That of course is just one of many possibilities. We are fortunate to be at the beginning of this exploration into the next generation of mental health treatment. This next generation of treatment will likely include non-psychedelic psychedelics, but I have a feeling that they will be just part of a much wider understanding of what a psychedelic might be than what we think of today.



Arthur Juliani

Postdoctoral researcher at Microsoft. Interested in artificial intelligence, neuroscience, philosophy, and meditation. http://arthurjuliani.com/