Sunday, August 2, 2026

Opparounds

🔬Psychiatric Research Article

Personalized fMRI-Guided TMS Targeting the Threat Neurocircuitry in PTSD: A Randomized Clinical Trial

Sanne J. H. van Rooij, Sean T. Minton, Cecilia A. Hinojosa, Timothy D. Ely, Patricio Riva-Posse, Kerry J. Ressler, Tanja Jovanovic, William M. McDonald, et al. · American Journal of Psychiatry · May 2026

Fifty adults with PTSD were randomized in a double-blind design to ten twice-daily sessions of 1-Hz TMS or sham. Rather than using a scalp landmark, the target was chosen individually: the point within the right dorsolateral prefrontal cortex showing the strongest functional connectivity to that patient's own right amygdala on baseline fMRI.

Active stimulation significantly reduced amygdala threat reactivity relative to sham, confirming the intervention did what it was designed to do at the circuit level. Clinically, both groups improved immediately after treatment, but the groups separated at three-to-six-month follow-up, with active TMS producing greater reductions in hyperarousal and in total PTSD symptoms.

💡 Why it matters

Two lessons travel beyond TMS. First, the delayed separation is a reminder that in trauma work an immediate post-treatment endpoint can hide a real effect — the sham arm's early improvement was not the whole story. Second, individualized connectivity-based targeting produced a measurable neurobiological change, which is the kind of evidence that moves neuromodulation from empirical to mechanistic. This is 50 patients at one center; treat it as a strong proof of concept, not practice change.

Read the paper → doi:10.1176/appi.ajp.20250749

📖Psychotherapy Research Article

Session-Level Emotional Processing and Subsequent Somatic Symptom Change in Online ISTDP for Treatment-Resistant Somatic Symptom Disorder: A Lagged Within-Person Secondary Analysis

Peter Lilliengren, Robert Johansson, Joel M. Town, Daniel Maroti · Journal of Psychosomatic Research · July 2026

Twenty-five patients with moderate-to-severe somatic symptom disorder who had already failed two prior interventions received up to 16 sessions of online intensive short-term dynamic psychotherapy from 17 therapists. After every session the therapist rated emotional processing — specifically the 'rise in complex feelings,' the ISTDP marker for mixed, conflicted affect breaking through defense. Somatic symptoms were tracked weekly on the PHQ-15.

Lagged multilevel models asked whether a session's emotional processing predicted the *next* week's symptoms. It did: a higher-than-usual rise in complex feelings predicted lower somatic symptom severity at the following assessment (b = -0.22, 95% CI -0.43 to -0.02). Between-person differences showed nothing (b = -0.03, p = .81) — patients who generally processed more emotion were not generally less symptomatic.

💡 Why it matters

The within-versus-between distinction is the whole finding and it is easy to miss. What predicted improvement was a patient having a better-than-their-own-average session, not being a generally emotionally expressive patient. That reframes what to look for week to week: not whether this patient is 'psychologically minded,' but whether this particular hour went deeper than their usual — which is something you can notice and pursue in real time.

Read the paper → doi:10.1016/j.jpsychores.2026.112926

💊Psychiatric Fact

Aripiprazole is a two-enzyme drug, and the label tells you exactly what to do

Aripiprazole is cleared by both CYP2D6 and CYP3A4, and because neither pathway can fully compensate for the other, the dose adjustments are unusually large and unusually specific. A strong 2D6 inhibitor — paroxetine, fluoxetine, bupropion, quinidine, duloxetine at higher doses — calls for halving the aripiprazole dose, as does being a 2D6 poor metabolizer (roughly 7 to 10 percent of people of European ancestry). A strong 3A4 inhibitor such as ketoconazole, clarithromycin, or ritonavir also warrants halving. Hit both pathways at once and the label calls for reducing to a quarter of the usual dose. In the other direction, carbamazepine or another strong 3A4 inducer roughly halves exposure and the dose is typically doubled.

Two consequences get missed in practice. First, bupropion is a potent 2D6 inhibitor, which makes the very common combination of bupropion plus adjunctive aripiprazole for depression a setup for akathisia at a dose that looks modest on paper — if a patient becomes restless on 5 mg, the interaction is a better first hypothesis than idiosyncrasy. Second, these adjustments apply to the long-acting injectables as well, where a strong inhibitor started mid-cycle cannot be undone; the aripiprazole lauroxil and monohydrate labels specify reduced doses for inhibitor use and advise against the two-month formulation with a strong 3A4 inducer entirely. Dose reduction after the injection is not an option, so the interaction has to be checked before it is given.

🛋️Psychotherapy Teaching Pearl

Psychic equivalence is not a metaphor problem — it is why your reassurance made things worse

Mentalization-based therapy names three prementalizing modes, and the one that most often derails a good intervention is psychic equivalence: the state in which inner reality is not experienced as a representation of the world but as the world itself. The thought 'you are disgusted by me' is not a fear about your reaction; it is a fact about your reaction, with the same epistemic status as the chair. Nothing you say about your actual state will be admitted as evidence, because from inside psychic equivalence there is nothing to be evidence *about*.

This is why reassurance backfires. Telling a patient in psychic equivalence that you are not disgusted does not offer an alternative view; it presents a contradiction of an established fact, and the patient must conclude either that you are lying or that they are crazy. Both conclusions worsen things. The complementary error is pretend mode, in which talk about mental states floats free of any felt experience — the articulate, insight-laden session that changes nothing — and the therapist's cue there is their own boredom.

The technical move is the same in both: stop, rewind, and explore. Interrupt the narrative, go back to the moment before mentalizing collapsed, and reconstruct it jointly and slowly, with the therapist explicitly not knowing. You are not supplying a correct account of the patient's mind or your own. You are restoring the process by which minds can be thought about at all — and only after that is running can content be addressed.

🗒️ Vignette

A 28-year-old woman with borderline personality disorder is describing a text her sister sent. Mid-sentence she stops, looks at the therapist, and says: 'You just checked the clock. You want me out of here.'

The therapist did glance at the clock. Three responses are available and two of them are wrong.

Denial — 'I wasn't watching the clock' — is false and will be caught. Reassurance — 'I don't want you out of here at all, I'm glad you're here' — is true, kind, and useless: it asks her to overwrite a fact with someone else's testimony, and in psychic equivalence that request is unmeetable.

Instead the therapist stops and rewinds. 'You are right, I looked at the clock. Let us slow this down, because something happened fast and I want to understand it. Before you saw me look — you were telling me about the text. What was happening for you right then?' She says, grudgingly, that she had been about to say she had not answered her sister for four days. 'And then you looked at me and saw me look at the clock. What went through you?' A pause. 'That you'd heard enough.'

The therapist does not correct this. 'So the sequence was: you were about to tell me something you feel bad about, and then you saw me look away, and it landed as me having heard enough. That makes a kind of sense.' Only now, with the sequence externalized and jointly held, does the therapist offer his own mind — tentatively, as one more thing to be curious about rather than as the truth: 'I can tell you what was in my head when I looked, though I am aware it may not feel like much against how certain that felt. I was thinking we had twenty minutes and I did not want to leave the sister thing half-finished.'

She is quiet, then says the thing that matters: 'I do that.' The content — the sister, the four days — is still there to work on. But it was unreachable ninety seconds ago, and it is reachable now.