When a psychedelic binds the 5-HT2A receptor, it doesn't flip a single switch. It activates multiple downstream signaling pathways simultaneously, and different compounds activate those pathways in different proportions. The field's standard tool for measuring this is the β-factor, a number that expresses how much a given compound favors one pathway over another, relative to serotonin, or any control one chooses. A positive β-factor means the compound leans toward β-arrestin recruitment; a negative one means it leans toward Gq protein activation. The problem is what the β-factor doesn't tell you.
It's calculated from a ratio of pathway activities, and ratios destroy absolute scale. A compound that weakly activates both pathways, say 8% on one arm and 12% on the other, produces the same β-factor as a compound that strongly activates both at 50% and 75%. The direction of preference is identical. The actual degree of receptor engagement is not. You lose that information the moment you divide.
This matters because the neuroplastogen field has made β-arrestin bias load-bearing in its mechanistic arguments. The core claim, made most explicitly around compounds like tabernanthalog, is that favoring β-arrestin over Gq allows 5-HT2A engagement without psychedelia, while still driving the plasticity that makes these compounds therapeutically interesting. A high β-factor toward β-arrestin is offered as evidence that this separation is achieved.
But a compound can score any β-factor at negligible efficacy. If neither pathway is meaningfully activated, the bias ratio is a description of nothing. The plasticity argument only holds if signaling is actually occurring at a level sufficient to drive downstream effects. The β-factor cannot tell you whether that threshold is met.
Eline Pottie, whose lab at Ghent University has developed some of the most rigorous receptor-proximal assays for 5-HT2A signaling, put it plainly in a recent conversation with Cascades Analytics: the β-factor is more qualitative than quantitative. Her assays use split-luciferase technology to measure whether signaling proteins physically interact with the receptor, a direct, receptor-specific readout rather than a downstream cellular response that could originate anywhere. Each experiment produces a full concentration-response curve for both the Gq and β-arrestin arms. The β-factor can be derived from those curves, but the curves themselves contain far more useful information than the ratio does.
Even setting aside the measurement problem, there is a deeper issue: the link between receptor-proximal signaling bias and the plasticity endpoints that matter therapeutically has not been established. The β-factor encodes a preference between Gq and β-arrestin at the receptor level. The therapeutic bet is on what happens much further downstream: new dendritic spines, increased synaptic density, changes in cortical connectivity. The most recent data have complicated this further: direct TrkB binding may account for part of the plasticity signal independent of 5-HT2A activation entirely, and intracellular rather than surface receptors appear to mediate plasticity-promoting effects in at least some contexts. Neither finding maps cleanly onto the Gq versus β-arrestin framework the neuroplastogen field is built on.
“The link between receptor-proximal signaling bias and the plasticity endpoints that matter therapeutically has not been established.”
The more coherent version of the neuroplastogen argument isn't that β-arrestin drives plasticity directly. It's that β-arrestin bias is a tuning mechanism: by shifting the compound's preference away from Gq, you keep Gq activation below the threshold that triggers hallucinations while still engaging it enough to drive plasticity downstream. The bet is that there is a window between those two thresholds, and that β-arrestin-biased partial agonism is how you stay in it. Preclinical data support this framing: Gq efficacy, not β-arrestin recruitment, predicts psychedelic-like effects in animal models, and a threshold level of Gq activation appears necessary to produce the head-twitch response used as a behavioral proxy for psychedelia.
That framing is more coherent, but it makes the β-factor problem worse. If the therapeutic claim depends on Gq activation landing within a specific efficacy window, then the absolute Gq Emax is the number that matters. The β-factor, being a ratio, tells you nothing about where that Emax sits. A compound could be β-arrestin biased and have a Gq Emax of 5%, well below any plausible plasticity threshold. Another could have a Gq Emax of 60%, potentially above the psychedelic threshold regardless of its bias score. The ratio looks the same. The pharmacology is completely different. And the field hasn't established where either threshold sits, which means even a compound with a well-characterized absolute Gq Emax cannot yet be confidently positioned in or out of the therapeutic window.
There is a further problem that rarely gets mentioned. The same preclinical data showing that Gq drives psychedelic-like effects also showed that β-arrestin-biased agonists induce receptor downregulation and tachyphylaxis. A compound designed to maximize β-arrestin engagement relative to Gq may therefore progressively desensitize the very receptor it depends on, undermining the durability of any therapeutic effect. This is not a theoretical concern: it is the known pharmacology of β-arrestin-mediated GPCR internalization. The neuroplastogen field has not addressed it.
The β-factor is a legitimate pharmacological descriptor. It characterizes where a compound sits in signaling space and allows meaningful comparisons across chemical series. What it isn't is a functional prediction. Any mechanistic claim built on it alone should come with the underlying numbers: what is the absolute potency and maximal effect on each pathway, in what cell type, at what receptor expression level? Without those, the β-factor tells you which way a compound leans, not whether it does anything worth caring about.
References
- Moliner, R., Girych, M., Brunello, C.A., Kovaleva, V., Biojone, C., Enkavi, G. and Castrén, E. (2023). Psychedelics promote plasticity by directly binding to BDNF receptor TrkB. Nature Neuroscience, 26, 1032–1041.
- Vargas, M.V., Dunlap, L.E., Dong, C., Carter, S.J., Tombari, R.J. and Olson, D.E. (2023). Psychedelics promote neuroplasticity through the activation of intracellular 5-HT2A receptors. Science, 379(6633), 700–706.
- Viol, L., Bhatt, D.L., Bhattacharya, A. and Bhattacharya, S. (2023). Identification of 5-HT2A receptor signaling pathways associated with psychedelic potential. Nature Communications, 14, 8101.