Ecstasy and Samadhi: Recent Research on Psychedelics and Meditation

During the 1960s, philosopher of science Mary Hesse developed an influential conception of scientific explanation, in which theoretical innovation is driven by metaphorical redescription—that is, breakthroughs in explanation result from novel analogies and the gradual refinement of these analogies. Hesse’s go-to example is a once-revolutionary analogy in physics, between billiard balls and gas molecules: the movement of gas molecules is analogous to the vectors and collisions of stripes and solids on a pool table. Like all analogies, this model is imperfect. There is not a one-to-one correspondence between the properties of the balls and the properties of the molecules. Some properties are shared by the two analogues—this is the positive analogy of the model. Other properties belong to one analogue and not the other—this is the negative analogy of the model (e.g., billiard balls are not microscopic; gas molecules don’t have hard resin shells). And then there’s the neutral analogy, those properties which could be positive or negative, shared or exclusive, but that we just don’t know about. So for Hesse, scientific explanation is a metaphorical process in which negative analogies are discarded and neutral analogies prompt investigators to ask new questions and create new experiments.

The central model of all this, the discourse within and around Psychedelic Sangha, is the analogy—positive, negative, neutral—between psychedelics and meditation. This discourse is not limited to scientific explanation, of course—it’s often colloquial, anecdotal, speculative, or pragmatic—but even then, we’re simply making and refining analogies between psychedelic and meditative modes. An obvious positive analogy between psychedelic substances and meditation practices is that both are capable of disrupting self-consciousness. This shared property is the starting point for a collaborative, interdisciplinary research paper, published in September 2018 by Frontiers in Psychology, entitled “Psychedelics, Meditation, and Self-Consciousness” (PMSC). It’s a groundbreaking work which aggregates a dizzying array of recent investigations into the neurophysiology and phenomenology of experiences of “self-loss” induced by both meditation and drugs.

One of the co-authors of PMSC is Aviva Berkovich-Ohana of the University of Haifa, a neurobiologist whose primary area of research is the effects of mindfulness training. Dr. Berkovich-Ohana was kind enough to answer a few of my questions by email. She told me that the beginnings of PMSC go back to The Sense of Self, a conference held at the University of Oxford in May 2017, where the authors (including philosopher and conference organizer Raphaël Millière, as well as psychopharmacologist and leading psychedelic researcher Robin Carhart-Harris) met and “through discussing our results we realized the many converging aspects, warranting a mutual comparison.” Convergences and divergences, positive and negative analogies—there’s a lot of nuance to unpack in this dense paper, but here are a couple of key points:

Neither meditation nor psychedelic states are simple, uniform categories.

Self-consciousness is a multidimensional construct which is divisible into several aspects or variables. Therefore, “self-loss” is multidimensional and can take several forms.

Xōchipilli and Buddha.

In Mary Hesse’s account of scientific models, the two subjects or analogues are usually regarded as “systems” of commonplaces—i.e., sets of shared facts and beliefs. Off the bat, with the first point that neither meditation nor psychedelic states are simple, uniform categories, Berkovich-Ohana and her colleagues problematize any straightfoward comparisons between these two “systems.” We’re dealing with a set of analogies, with a model, that’s much more complex than “gas molecules behave like billiard balls.” For instance, PMSC opens with the acknowledgment that there are “more than 100 varieties of meditation ... [a] remarkable diversity of traditions,” before digging into recent attempts at a “cross-cultural classification of standard meditation styles validated by functional and structural neuroanatomical data.” In particular, PMSC builds upon the results of a 2016 “meta-analysis” by Kieran Fox, et al., which divides the “neural correlates” of various meditation practices into four common styles: focused attention (e.g., śamatha meditation), open monitoring (e.g., vipassanā meditation), loving-kindness meditation, and mantra recitation. Each style of meditation activates and deactivates different regions of the brain, but all four styles affect the insular cortex in one way or another, which indicates a central role for the “attentional control of bodily awareness, and awareness of the breathing in particular, during various contemplative practices.”

The heart of PMSC is the second point, which is really two related conclusions: Self-consciousness is a multidimensional construct which is divisible into several aspects or variables. Therefore, “self-loss” is multidimensional and can take several forms. Berkovich-Ohana, et al., begin with a “greatly simplified” model of self-consciousness and self-loss, a model with two dimensions or axes: “narrative” selfhood and “embodied” (or “multisensory”) selfhood. The narrative dimension includes two variables: the frequency of self-related thought, and access to autobiographical information. The embodied dimension of selfhood consists of three variables: body ownership, bodily awareness, and self-location.

Self-related thought includes de se (“of oneself”) thoughts, i.e., thoughts involving the first-person concept, usually expressed using the first-person pronoun. De se thoughts include reflection on one’s identity and beliefs, as well as more mundane mind-wandering. Self-related thought also includes “mental time travel to the future,” such as imagining one’s life trajectory, or simply pondering what to do this weekend. The second variable of narrative self-consciousness is autobiographical memory retrieval. According to Berkovich-Ohana, et al., there is ample anecdotal evidence that psychedelic drugs can disrupt access to abstract information about oneself, resulting in a sort of “reversible retrograde amnesia.” These self-referential thoughts, imaginings, and memories weave the “narrative of our daily lives” and the “stories we tell ourselves about the kind of person we are or want to be.”

In addition to the “narrative self,” PMSC follows the lead of philosophers such as Shaun Gallagher and Thomas Metzinger (who both participated in the Oxford Sense of Self conference) and proposes a “minimal” or “embodied” self with the three “somatosensory” aspects mentioned above: body ownership, bodily awareness, and self-location. These aspects are closely related but “can come apart in special cases.” For example, “it is unclear whether meditation-induced states can involve a complete loss of bodily awareness (including interoceptive awareness), while it does seem that they can involve a loss of self-location.” There are at least a couple of glaring omissions with this picture of embodied selfhood. One is temporal self-location, which PMSC consolidates into spatial self-location without much elaboration, simply noting that many reports of “drug-induced ego dissolution” (DIED) describe a loss of sense of temporal duration, and then “tentatively” suggesting that “some drug-induced and meditative states might lack both spatial and temporal self-locating content.” Another omission is the sense of agency, which is intentional on the part of the authors: the “sense of agency is an ambiguous concept ... [a] more sophisticated model could take into account additional parameters, such as the sense of agency.”

However, this two-dimensional model is “at most as a helpful idealization which reduces the dimensionality of the notion of self-consciousness.” Besides its oversimplification, it can be misleading, since it “suggests that the loss of body ownership, bodily awareness and self-location are all degrees of self-loss that can be ordered along the same [embodied] dimension.” The same goes for self-related thoughts and autobiographical memory along the narrative dimension. So instead Berkovich-Ohana, et al., propose a more complex model, a six-dimensional state space in which all these cognitive and somatosensory aspects of selfhood are independent variables which meditation or psychedelics can disrupt either in isolation or in combination. Each aspect of self-consciousness may be differently affected by psychedelics or meditation, or by specific psychedelic drugs and specific meditation practices. That is, each specific psychedelic drug or meditation style may have its own profile with regard to the aspects it disrupts. This leads us to the conclusion that self-loss is multidimensional and can take several forms: “A conscious state in which one of these aspects is radically disrupted may be described as ‘selfless’ in one respect, although a subject undergoing such a state could retain other forms of self-consciousness.” Self-loss is not a single, unequivocal phenomenon.

A tentative multidimensional model of self-loss. Global states of consciousness are plotted on the radar chart according to their score on six dimensions (using an arbitrary scale), representing the degree to which they involve (1) a sense of body ownership, (2) awareness of bodily sensations, (3) awareness of spatial self-location, (4) rich phenomenology, (5) access to semantic autobiographical information, and (6) self-related thoughts. Regions in the radar chart represent idealized examples of global states of consciousness, including an ordinary state during wakefulness (dotted white line), two examples of meditation-induced states and two examples of drug-induced states. The region in dark green is an example of a typical meditative state with increased bodily awareness (via attentional focus on the breath), slightly decreased overall phenomenal richness (via visual-auditory deprivation) and decreased frequency of self-related thoughts. The region in light green is an example of a “selfless” state described by experienced meditators, with a cessation of self-related thought, a loss of body ownership, agency and self-location, and significant reductions in bodily awareness and phenomenal richness. The region in dark purple is an example of a state induced by a moderate dose of psychedelic drugs such as LSD or psilocybin, with increased bodily awareness (modulated by salient and unusual bodily sensations) and increased phenomenal richness (via decreased sensory gating and vivid perceptual abnormalities). The region in light purple is an example of drug-induced ego dissolution with a loss of narrative and multisensory aspects of self-consciousness, but rich sensory content. This chart illustrates how even states of “total self-loss” can differ from a phenomenological perspective between meditation (light green line) and psychedelics (light purple line). In addition, it shows that some states of consciousness induced by both meditation practice and psychedelics can also score higher than baseline on certain dimensions of self-consciousness, in particular bodily awareness. Overall, states of “self-loss” are the exception rather than the norm for both modes of induction. Finally, it should be noted that the phenomenology of altered states induced by meditation and psychedelics may considerably change over time, sometimes very quickly; consequently, the idealized states plotted on this figure should be considered as phenomenological “snapshots” at a given time. For example, “Psychedelic state 1” (dark purple line) could be part of the same drug-induced experience as “Psychedelic state 2” (light purple line), assuming that the phenomenology dynamically shifts toward the peak of the experience (during the transition to drug-induced ego dissolution).

Perhaps the most consequential change between the simple two-dimensional model and the six-dimensional “radar chart” is the addition of “phenomenal richness or sparsity” (i.e., the degree of sensory gating or the “bandwidth” of conscious contents) as the sixth variable. For me, this evokes earlier models of selfless experience, such as Mircea Eliade’s distinction between ecstasis (shamanic transport or “turning outward”) and enstasis (yogic samādhi or “turning inward”), or Roland Fischer’s EEG-based “cartography of ecstatic and meditative states” (below), which is a continuum running from hyperarousal to hypoarousal. This has huge implications for analogies between psychedelics and meditation. Indeed, “phenomenal richness” seems to be a primary variable which tunes the other aspects of self-consciousness. I asked Dr. Berkovich-Ohana about this, and she replied: “Your interpretation is correct. This feature might indeed be a key player in the typology of altered states of consciousness (ASCs). There are different ‘kinds’ of ASCs, some expanding to include more stimuli, the other ‘collapsing internally’ ... It seems to be about time to write a paper distinguishing between ‘emptiness’ and ‘expansion’, proposing different neural mechanisms for both. I’m happy to guide a serious PhD or a postdoc on this topic!”

One big implication of “phenomenal richness or sparsity” as a variable of self-loss becomes apparent in PMSC’s distinction between “nondual awareness” (NDA) and “pure consciousness” (PC): states of non-dual awareness are not necessarily states of cessation (i.e., samādhi) which lack determinate phenomenological contents! According to Berkovich-Ohana, et al.:

“It is an open question whether NDA differs from pure consciousness, and in what respect ... There has been considerable debate about whether purely conscious states are even possible ... purely conscious states could be defined as states of extreme absorption involving high sensory gating, whose sparse phenomenal content has little overlap with the rich phenomenology of ordinary wakeful experience ... non-dual awareness might not entail pure consciousness, because states of total self-loss (lacking narrative and multisensory self-consciousness) need not have very sparse phenomenal content.”

This suggestion is sure to be controversial among meditation practitioners in those Buddhist and Hindu traditions which emphasize the attainment of pure consciousness, or anyone who insists on defining “self-loss” in absolute terms.

Xōchipilli and Buddha.

It’s tempting to see sensory gating or “phenomenal richness or sparsity” as a sharp line dividing psychedelic from meditative experience, the same line that Eliade drew between shamanism and yoga, ecstasy and samādhi. Indeed, this seems to be the position of those Buddhists and yogis who argue that psychedelic ecstasy is doubling down on the delusion of the senses, rather than transcending it. However, as we’ve just seen with PMSC, things aren’t that simple. Specific drugs or styles of meditation can affect each variable of self-consciousness differently, and this includes phenomenal richness. There are anomalies and outliers. For example, studies have found that the “open monitoring” style of meditation (“bringing attention to the present moment and openly observing mental contents without getting caught up in focusing on any of them”) actually deactivates the right thalamus, a region of the brain that filters out sensory stimuli—that is, open monitoring seems to decrease sensory gating and therefore increase phenomenal richness. And then there’s the infamous psychedelic 5-MeO-DMT. According to some anecdotal evidence, 5-MeO-DMT, rather than inducing a rich sensory phenomenology and vivid hallucinations, may “induce states of radical absorption reminiscent of Samadhi practice.” PMSC pulls quotes from Erowid.org trip reports (as well as R. A. Masters’ Darkness Shining Wild) which describe the peak 5-MeO-DMT experience as one of “emptiness, nothingness or void ... a cessation of thoughts, extreme sensory deprivation and a complete loss of self-consciousness.” So to reiterate, states of self-consciousness and self-loss (even the “total” self-loss of nondual awareness) are multidimensional and complex, and neither meditation nor psychedelics are simple, uniform categories. Our analogy still needs a lot of work.

ψ संघ

I asked Dr. Berkovich-Ohana if she had any plans for taking this line of research further, and she responded: “I certainly plan to take it a further step, using neurophenomenology (a combination of neural measurement and phenomenology—as done in several other papers of mine, e.g. Ataria, Dor-Ziderman, & Berkovich-Ohana, 2015; Dor-Ziderman, Ataria, Fulder, Goldstein, & Berkovich-Ohana, 2016). I would start with some case studies, in collaboration with current clinical studies in Israel. I will be happy to recruit postdocs with background in cognitive studies, from USA or Canada—there is a current call for fellowships, for the interested—please send me an email.”

ψ संघ