Thank you! It is so very important to highlight such progress not only because of whatever intrinsic value it may have (regarding 'the facts of the matter') but also because it provides a crucial reminder of how little is yet known. It is valuable to see both things framed in this positive, future-focused, way.
Sometimes it feels like every researcher in psychiatry should have the physicist A.A. Michelson's (in)famous quotation from the end of the 19th century proudly displayed: “it seems probable that most of the grand underlying principles have now been firmly established.” (Over the next few decades (and in no small part due to Michelson’s own work), fundamental physical theory underwent its most dramatic changes since the times of Galileo and Newton.)
I'll go ahead and bet on 'flopped' by 2036. With confidence. But flopping isn't actually a flop. We move forward in tiny steps, and the failure of each step brings on the next, and each next step bears the fingerprint of the previous one.
This was a good read and I hope it improves treatment. Type 2 sounded really familiar to me as an AuDHDer and I wonder if that suggests similar features that could help explain why autistic people are most likely to have some sort of psychotic features at some point in our lives and why autistic catatonia patients are likely to be harmed rather than helped by antipsychotics.
Also, my motor inhibition is truly awful but I've never had a clinician evaluate or care about that, even though I think it's probably more closely related to my most-severe executive functions than the things like mood, anxiety, sleep, or focus that clinicians always try to steer the conversation toward. Like, my problem is that I cannot inhibit repetitive behaviors like scrolling no matter how much I want to, and I cannot initiate voluntary behaviors without external structure or stimulation factors that I can't control, and I, like many other people with autistic inertia, think those are due to the same underlying mechanism. Guanfacine was a miracle cure for that for about a week, but efficacy has been greatly reduced (alongside side effects) with tolerance. Anyway, I was excited that these researchers are taking impaired motor inhibition as an important thing to look for.
This isn't a criticism of what B-SNIP has accomplished, the findings feel genuinely significant. But it seems like the question of who gets to have 'intact underlying neurology' may not be separable from the question of who has had enough material stability to protect their nervous system in the first place. For those of us coming from peer movement and community mental health backgrounds, the absence of social determinants from even the most sophisticated biological frameworks is a recurring and consequential silence, and one worth naming even when, maybe especially when, the science itself is this compelling.
On the scope point, I agree that a program like B-SNIP brackets the social aspects by its methodological decisions. The biotypes were built agnostically from cognition and electrophysiology. And that by itself doesn’t tell us whether “material conditions” are playing a role or not.
The second is a stronger, causal claim about the role of social context, and here the data are interesting. B-SNIP’s own group has begun clustering the “exposome” (trauma, substance use, socioeconomic status) and testing environmental risk against the biotypes. [https://www.nature.com/articles/s41598-025-14438-6 & https://pubmed.ncbi.nlm.nih.gov/39777534/ ] I didn’t go into it in the post because I am still getting familiar with that literature. If social deprivation were simply damaging nervous systems into the deficit groups, we’d expect the strongest exposure–symptom coupling in Biotypes 1 and 2, the neurobiologically compromised ones. The findings are different. Exposure burden is fairly flat across biotypes, and the coupling between environmental risk and symptom severity concentrates in Biotype-3, the group with relatively preserved cognition and where the biomarker panel cannot tell apart from healthy controls. This suggests a pathway into a psychosis subtype where the social-psychological mechanisms play a stronger role while cognition and electrophysiology are relatively preserved. Another factor to consider: relatives of probands carry the same bio-factor patterns, so these signatures have substantial familial loading and stronger genetic associations.
So while B-SNIP consortium have their work cut out for them in terms of figuring out the social context and biotype relationships, I think those invested in the social determinants of mental health side of things have their work cut out for them too in understanding how these scientific developments update our understanding of psychoses and the tendency to invoke lack of material stability as a default explanation for neurological dysfunctions.
Thank you, this was a fascinating read! I’m curious what biotypes we might see for individuals who have psychosis-like symptoms that appear to arise from distinct processes or cluster with other mental disorders. For example people with auditory hallucinations arising from PTSD, or paranoia or delusions in the context of extreme anxiety.
Yes, would be good to investigate. I suspect they’ll be more like biotype 3 (since they generally don’t exhibit neurocognitive deficits at the same level as schizophrenia spectrum) but a small proportion may have biotype 1 or 2 like features, or perhaps entirely new biotypes!
Hello, thanks for describing this in-depth work emerging in the research world. It certainly aligns with what I see in my own practice as a psychologist. The categories of bipolar with psychosis and schizoaffective with bipolar presentation seem to co-exist in the same person, and shift back and forth over time; eventually (sometimes) even resolving given the depth of therapy over the long-term which can shift how previous trauma can be processed - which then has an effect on cognitive style. But even when the paranoia, auditory hallucinations, and fixed delusions are worked through (when this is possible), there still seem to be neurological difficulties remaining - migraines, seizure like activity, etc. Have you encountered this in your practice?
Also, wondering if you explain abit more about the visual / optical aspects of the assessments, and what is known about visual disturbances? Thanks much.
Totally, I see a lot of residual neurocognitive deficits even when hallucinations and delusions resolve.
In a prosaccade task, a person looks at a central point, then a target flashes in their peripheral vision. They simply look toward it as quickly as possible. This measures basic visual orienting.
In an antisaccade task, now the person must deliberately look away from the target to the opposite side. This requires suppressing the automatic reflex to look at a sudden stimulus and then generating a voluntary eye movement in the other direction. It measures inhibitory control and the ability to override prepotent responses.
people with schizophrenia, on average, make more antisaccade errors (reflexively looking toward the target before correcting), and their error rates are substantially elevated compared to healthy controls. This is pretty well replicated, but not specific or sensitive to schizophrenia (also seen in some schizoaffective and bipolar cases, and biological relatives)
Antisaccade performance depends partly on a circuit involving the dorsolateral prefrontal cortex (DLPFC)
Smooth pursuit eye movement dysfunction is also seen in SZ spectrum. When tracking a moving object, pursuit signal is “broken up”, the eyes fall behind the target, then catch up with small rapid saccades (catch-up saccades).
Thank you for this — the clearest account of B-SNIP I've read. What struck me is not that the biotypes differ but where. BT2 and BT3 deviate on essentially one brain response, and it's the same one. BT1 deviates on three — N100 down, P300 down, saccades slowed — and the one component it leaves alone is that same P200. Four stages, and P200 is carrying all the weight.
A frame that might make that less strange. Suppose perception happens in two steps: information is first organized until it settles into a single coherent state, and only then does that finished state get broadcast — frontal systems notice something has arrived, judge its significance, and pass it on to memory, language, and movement. The broadcast is the familiar P300 complex, from about 300 ms. The settling happens before it.
Read BT1 that way and its profile gets simple. If the trouble is in building the state rather than finishing it, the brain still finishes on schedule — just with thin material. So the closing response looks normal while what comes before and after is weak. One problem, three deviations, and the spared component is the one you'd predict. BT2 and BT3 then aren't milder BT1s; they're a different problem, sitting right at the closing step, with everything either side intact.
Which raises a question I'd put forward tentatively: might P200 be where a moment actually becomes conscious — the point at which experience arrives and can be broadcast? If so, the biotypes look less like three severities and more like failures at different stages. Two things could sink it. BT1's spared P200 sits between two suppressed neighbors, and averaged waveforms can make a component look preserved when it isn't. And the finding you cite in the comments — that environmental risk couples most tightly to symptoms in BT3 — suggests that group's P200 may be state rather than trait. Both testable in data the consortium already holds.
Thank you! It is so very important to highlight such progress not only because of whatever intrinsic value it may have (regarding 'the facts of the matter') but also because it provides a crucial reminder of how little is yet known. It is valuable to see both things framed in this positive, future-focused, way.
Sometimes it feels like every researcher in psychiatry should have the physicist A.A. Michelson's (in)famous quotation from the end of the 19th century proudly displayed: “it seems probable that most of the grand underlying principles have now been firmly established.” (Over the next few decades (and in no small part due to Michelson’s own work), fundamental physical theory underwent its most dramatic changes since the times of Galileo and Newton.)
I'll go ahead and bet on 'flopped' by 2036. With confidence. But flopping isn't actually a flop. We move forward in tiny steps, and the failure of each step brings on the next, and each next step bears the fingerprint of the previous one.
Thanks Michael! I like the Michelson's reference. Very apt.
This was a good read and I hope it improves treatment. Type 2 sounded really familiar to me as an AuDHDer and I wonder if that suggests similar features that could help explain why autistic people are most likely to have some sort of psychotic features at some point in our lives and why autistic catatonia patients are likely to be harmed rather than helped by antipsychotics.
Also, my motor inhibition is truly awful but I've never had a clinician evaluate or care about that, even though I think it's probably more closely related to my most-severe executive functions than the things like mood, anxiety, sleep, or focus that clinicians always try to steer the conversation toward. Like, my problem is that I cannot inhibit repetitive behaviors like scrolling no matter how much I want to, and I cannot initiate voluntary behaviors without external structure or stimulation factors that I can't control, and I, like many other people with autistic inertia, think those are due to the same underlying mechanism. Guanfacine was a miracle cure for that for about a week, but efficacy has been greatly reduced (alongside side effects) with tolerance. Anyway, I was excited that these researchers are taking impaired motor inhibition as an important thing to look for.
Interesting ideas to consider! Have you ever undergone comprehensive neuropsychological testing?
This isn't a criticism of what B-SNIP has accomplished, the findings feel genuinely significant. But it seems like the question of who gets to have 'intact underlying neurology' may not be separable from the question of who has had enough material stability to protect their nervous system in the first place. For those of us coming from peer movement and community mental health backgrounds, the absence of social determinants from even the most sophisticated biological frameworks is a recurring and consequential silence, and one worth naming even when, maybe especially when, the science itself is this compelling.
i’m still curious about this
Sascha, thank you for bringing this up.
On the scope point, I agree that a program like B-SNIP brackets the social aspects by its methodological decisions. The biotypes were built agnostically from cognition and electrophysiology. And that by itself doesn’t tell us whether “material conditions” are playing a role or not.
The second is a stronger, causal claim about the role of social context, and here the data are interesting. B-SNIP’s own group has begun clustering the “exposome” (trauma, substance use, socioeconomic status) and testing environmental risk against the biotypes. [https://www.nature.com/articles/s41598-025-14438-6 & https://pubmed.ncbi.nlm.nih.gov/39777534/ ] I didn’t go into it in the post because I am still getting familiar with that literature. If social deprivation were simply damaging nervous systems into the deficit groups, we’d expect the strongest exposure–symptom coupling in Biotypes 1 and 2, the neurobiologically compromised ones. The findings are different. Exposure burden is fairly flat across biotypes, and the coupling between environmental risk and symptom severity concentrates in Biotype-3, the group with relatively preserved cognition and where the biomarker panel cannot tell apart from healthy controls. This suggests a pathway into a psychosis subtype where the social-psychological mechanisms play a stronger role while cognition and electrophysiology are relatively preserved. Another factor to consider: relatives of probands carry the same bio-factor patterns, so these signatures have substantial familial loading and stronger genetic associations.
So while B-SNIP consortium have their work cut out for them in terms of figuring out the social context and biotype relationships, I think those invested in the social determinants of mental health side of things have their work cut out for them too in understanding how these scientific developments update our understanding of psychoses and the tendency to invoke lack of material stability as a default explanation for neurological dysfunctions.
Thank you, this was a fascinating read! I’m curious what biotypes we might see for individuals who have psychosis-like symptoms that appear to arise from distinct processes or cluster with other mental disorders. For example people with auditory hallucinations arising from PTSD, or paranoia or delusions in the context of extreme anxiety.
Yes, would be good to investigate. I suspect they’ll be more like biotype 3 (since they generally don’t exhibit neurocognitive deficits at the same level as schizophrenia spectrum) but a small proportion may have biotype 1 or 2 like features, or perhaps entirely new biotypes!
Excellent summary thank you
Hello, thanks for describing this in-depth work emerging in the research world. It certainly aligns with what I see in my own practice as a psychologist. The categories of bipolar with psychosis and schizoaffective with bipolar presentation seem to co-exist in the same person, and shift back and forth over time; eventually (sometimes) even resolving given the depth of therapy over the long-term which can shift how previous trauma can be processed - which then has an effect on cognitive style. But even when the paranoia, auditory hallucinations, and fixed delusions are worked through (when this is possible), there still seem to be neurological difficulties remaining - migraines, seizure like activity, etc. Have you encountered this in your practice?
Also, wondering if you explain abit more about the visual / optical aspects of the assessments, and what is known about visual disturbances? Thanks much.
Totally, I see a lot of residual neurocognitive deficits even when hallucinations and delusions resolve.
In a prosaccade task, a person looks at a central point, then a target flashes in their peripheral vision. They simply look toward it as quickly as possible. This measures basic visual orienting.
In an antisaccade task, now the person must deliberately look away from the target to the opposite side. This requires suppressing the automatic reflex to look at a sudden stimulus and then generating a voluntary eye movement in the other direction. It measures inhibitory control and the ability to override prepotent responses.
people with schizophrenia, on average, make more antisaccade errors (reflexively looking toward the target before correcting), and their error rates are substantially elevated compared to healthy controls. This is pretty well replicated, but not specific or sensitive to schizophrenia (also seen in some schizoaffective and bipolar cases, and biological relatives)
Antisaccade performance depends partly on a circuit involving the dorsolateral prefrontal cortex (DLPFC)
Smooth pursuit eye movement dysfunction is also seen in SZ spectrum. When tracking a moving object, pursuit signal is “broken up”, the eyes fall behind the target, then catch up with small rapid saccades (catch-up saccades).
Thank you for this — the clearest account of B-SNIP I've read. What struck me is not that the biotypes differ but where. BT2 and BT3 deviate on essentially one brain response, and it's the same one. BT1 deviates on three — N100 down, P300 down, saccades slowed — and the one component it leaves alone is that same P200. Four stages, and P200 is carrying all the weight.
A frame that might make that less strange. Suppose perception happens in two steps: information is first organized until it settles into a single coherent state, and only then does that finished state get broadcast — frontal systems notice something has arrived, judge its significance, and pass it on to memory, language, and movement. The broadcast is the familiar P300 complex, from about 300 ms. The settling happens before it.
Read BT1 that way and its profile gets simple. If the trouble is in building the state rather than finishing it, the brain still finishes on schedule — just with thin material. So the closing response looks normal while what comes before and after is weak. One problem, three deviations, and the spared component is the one you'd predict. BT2 and BT3 then aren't milder BT1s; they're a different problem, sitting right at the closing step, with everything either side intact.
Which raises a question I'd put forward tentatively: might P200 be where a moment actually becomes conscious — the point at which experience arrives and can be broadcast? If so, the biotypes look less like three severities and more like failures at different stages. Two things could sink it. BT1's spared P200 sits between two suppressed neighbors, and averaged waveforms can make a component look preserved when it isn't. And the finding you cite in the comments — that environmental risk couples most tightly to symptoms in BT3 — suggests that group's P200 may be state rather than trait. Both testable in data the consortium already holds.