Can evidence ever really change something that is rooted in ideology?

Can evidence ever really change something that is rooted in ideology?

An example of the ever-growing chasm in psychedelic-assisted therapy

Last weekend, while scrolling on various social media sites, I came across a couple of commentaries on recent publications about psilocybin-assisted therapy for PTSD (see references below).

The first commentary I came across was on the recent Dougherty et. al. (2026) publication in the Journal of Psychopharmacology. The finding cited in the publication that the commentary post focused on was that ‘on average, 78% of the time was filled with silence’. The ‘78% of the time’ refers to the sessions where participants were administered psilocybin (as oppose to different types of sessions within the ‘treatment protocol’, such as preparation or integration/follow up, etc., where psilocybin is not administered).

The problem with this commentary was that it was ALL it focused on, conveniently omitting the important nuances (I’ll get to it shortly).

The second commentary post I came across was that on the findings by Armstrong et. al. (2026) who found that ‘participant PTSD symptoms improved significantly during the preparatory therapy phase, before the administration of psilocybin, and those who improved more before dosing had more positive outcomes at one month after dosing.’ Again, ‘dosing’ here refers to the psilocybin administration session. It further discussed that expectancy was not found to have played a role in the outcomes of the therapy tested.

*

Just in case it hasn’t become clear yet, here’s my explicit disclosure about where I sit in this debate: I am strictly on the side of ‘therapy assisted by psychedelics’ rather than ‘psychedelic therapy’ (like we would say, for example, ‘antibiotics therapy’, meaning that we just hand people the substance and off they go). If, over time, the evidence contrary to my position becomes overwhelming, I will do my best to swallow the bitter pill and change my stance (even typing out the entertaining of this possibility is making me squirm) but, until then, I stick firmly with it. I suppose, as per main title of this article, I’m digging in my heels as per my ideology.

Also, I am all for cognitive liberty. However, in this particular instance – where people are looking for relief from suffering related to mental health struggles – cognitive liberty isn’t what people seek.

*

Now that my personal declarations of ideological biases are out of the way, I return to the aforementioned articles and the commentary posts that alerted me to their existence.
It is, indeed, true that Dougherty et. al. (2026) reported that 78% of the psilocybin session was silent, i.e., the participant did not interact much with the therapists in the room during the hours of acute experience. My issue with the commentary that first introduced me to the existence of this publication is that they failed to inform their audience of the rest of the findings in the qualitative element of the same study, which were as follows:

‘[…] (1) support was minimally enacted, yet EXPERIENTIALLY SALIENT’ (capitalisation my own)

‘(2) autonomy was promoted through the introspective psychedelic state and non-directive support’

‘(3) primary modes of support during altered states of consciousness included REASSURANCE AND VALIDATION’ (capitalisation my own)

I urge you to make mental note of these three points, for they will become relevant – not that they aren’t relevant in and of themselves – later, when I discuss the second study.

The commentary post either omitted the nuance entirely or put it in a considerably smaller font. Another thing that the poster failed to mention is that all of the 14 authors of this study declared a ‘potential conflict of interests’ by means of, among others, being current or past Compass Pathways (a for-profit company that funded and sponsored this particular study) employees or having shares in the company.

The final thing that the poster omitted to share with their followers was a full reference to the original publication on which they based their post. I think that we should always practice transparency and empowering our readers to use their own agency by sharing the original sources we base our writing on, however, incidentally, this omission led to me being able to observe another curiosity. When I entered some key words into a search engine in the effort to access the original publication, five out of the first ten links were to stock market-related websites rather than to the journal in which this study was published.

I am not implying any intentional disingenuity on the part of the poster of this first commentary, but their chosen way of presenting this study to their followers had the effect – judging by the comment section – of furthering the convenient- and desirable-by-some narrative that I can only sum up as ‘attaching therapy to psychedelics has always been unnecessary and we should just fit them neatly into a well-established model where we just prescribe people substances and that’s that.’

I think being selective in what and how we choose to report and either actively steering the narrative into a specific direction or leaving room for it to unfold in the way that a bigger picture is lost is irresponsible at best, and dangerous at worst.

While I am undecided about where I sit in terms of the particulars of up-to-date protocols for psychedelic-assisted therapy, the fact is that what may be good for a seasoned psychonaut, may not be good for a person seeking help for PTSD or treatment-resistant depression.

The second study – one by Armstrong et. al. (2026) – I was alerted to by another commentary post is in perfect contrast to how Dougherty’s et. al. (2026) study was presented. While their findings cannot be directly compared because the researchers were looking at and measuring different things, inferences can be drawn from juxtaposing the two studies. Armstrong et. al. (2026) found something that validated my bias (I refer you back to the declaration of my ideological bias above). They found that the preparation stage of the psilocybin-assisted therapy protocol ALONE, BEFORE (capitalisation my own) the psilocybin was even taken, significantly reduced participants’ PTSD scores. Furthermore, ‘those who improved more before dosing had more positive outcomes at one month after dosing.’

In Table 1 of Armstrong et. al. (2026) study, the preparation element is somewhat broken down into the following (direct quotes):

• (Prep session 1-2) ‘Establishing therapeutic alliance, enhancing psychological safety, and optimizing the set and setting. Information gathering, including personal and trauma histories and discussion of PTSD-related avoidance patterns.’

• (Prep session 3-4) ‘Education about possible perceptual, emotional, and existential experiences; normalization of challenging experiences; emphasis on curiosity, acceptance, and integration as the key mechanisms of change. Sessions also include clarification of expectations, intention-setting, boundaries, and safety procedures for dosing days.’

• (Prep session 3-4) ‘Training in diaphragmatic breathing, interoceptive awareness, and other grounding strategies to support regulation during dosing and enhancing agency.’

The reason why I cite the above three points here is because, I’d argue, that is when and how the participants’ attunement into ‘EXPERIENTIAL SALIENCE’ Dougherty’s et. al. (2026) speaks of but the commentary post failed to highlight is built. It is also when the buds of trust in the upcoming process and trust into the safety of container that is being build begin to develop which can be a deciding factor whether the participants will be able to recognise when they may need ‘REASSURANCE AND VALIDATION’ and whether they will be able to trust it. Which can be the ‘make it or break it’ thing. And, in the context of psychedelic-assisted therapy (PAT) and why people seek it in the first place, the option of ‘break it’ can have horrendous consequences.

It has always (to my knowledge but, please, point me in the right direction if I’m mistaken) been that PAT trial protocols tested non-directive ways of being (on the part of the therapists’ present) not only, but especially, during the acute sessions. There was never a suggestion that during the acute sessions analysis (as some commenters under the first commentary post spoke against as if there was ever such a suggestion) of the unfolding experience should be carried out. It has always been that the container and the safety in preparation for the medicine administration session was created in the sessions prior to the psychedelic experience and carried over into what is often referred to as ‘integration’ or ‘follow up’. Therefore, the point the commentary post that used Dougherty’s et. al. (2026) study report for reference was trying to make is moot. However, it implicitly serves the narrative that psychedelics should just be prescribed as any other medicine without a supportive container. That, in my opinion, is both unethical and dangerous and serves the status quo of affairs that stopped serving the broader society long time ago.

Also, in pursuit of equal treatment, I ought to mention that two of the eight authors of the Armstrong et. al. (2026) study declared a ‘potential competing interest’ as they are (or were at the time of publication) board members of Source Research Foundation (SRF, a non-profit organisation).

I won’t interpret the findings of either of the study or what they may indicate when juxtaposed together any further as I think the findings speak for themselves. My urge to write this commentary about the commentary was motivated by the selective and, as a result, misleading – as evidenced in the comment section – reporting of the first study. At this point, I feel I pointed to what I wanted to point, and I leave further interpretation to anyone who reads this.

But I do want to get the last few of things in:

1. Sadly, we have to dig deep and find bandwidth to not only familiarise ourselves with studies’ finding, but also to learning about who sponsors them and what discourse the way they are reported to the wider public seem to direct the readers to.

1. Most, if not all, of our wounds are acquired in a relational space. Therefore, a relational space is needed for the wounds to heal (to which the findings of the qualitative element of Dougherty’s et. al. (2026) speaks to but was mostly omitted by the commentary post).

1. Unrelated to the present article but, in my opinion, needs to be talked about in general when we’re talking about ‘science’: what and why DOESN’T get researched? Why some research ideas get funding or sponsorship easier than others and what does it mean? What ideologies drive the granting and refusal of funding and/or sponsorship for research ideas? What impact does our base ideology has on how we interpret what we see, including the presumption of good faith?

References:

Dougherty, R. F., Modlin, N. L., McGowan, N. M., Staples, P., Williams, E., Clarke, P., Shafiei, M., Leininger, C., Alti, M., Croal, M., Marwood, L., Kirlic, N., Ryslic, G. A. and Goodwin, G. M. (2026). Silence is golden: Documenting the speech production of participants and support providers in psilocybin administration sessions for the treatment of post-traumatic stress disorder. Journal of Psychopharmacology, 02698811261464988.

Armstrong, S.B., Levin, A.W., Sepeda, N.D. et al. Safety, feasibility, and preliminary clinical outcomes of psilocybin-assisted therapy for veterans with severe, treatment-resistant PTSD: an open-label pilot clinical trial. Commun Med 6, 411 (2026). https://doi.org/10.1038/s43856-026-01767-4