This Week in Psychiatry — Jun 7, 2026

Welcome to This Week in Psychiatry. This week we're covering 9 notable papers spanning new insights into psychosis, innovations in therapeutics from brain circuits to digital aids, and key challenges in delivering care for youth and individuals with substance use disorders. Let's dive in.

A central goal in psychosis research is to better predict outcomes for individuals at clinical high risk. A study in *JAMA Psychiatry* explored resting-state EEG microstates as a potential biomarker [6]. In a large cohort from the North American Prodrome Longitudinal Study, researchers found that individuals at high risk showed reduced temporal stability of these microstates compared to healthy controls, reflecting more rapid shifting between brain network states. This instability was even more pronounced in those who later converted to psychosis and was associated with an earlier time to conversion and greater positive symptoms. Interestingly, while the high-risk group had cognitive deficits, lower microstate stability was linked to better attention scores, suggesting it might partly reflect a compensatory mechanism. These findings suggest EEG microstate instability shows promise as a prognostic tool to help stratify risk. Moving from prediction to presentation, a paper in *The American Journal of Psychiatry* asks whether cannabis-associated psychosis is a distinct clinical entity [9]. In a prospective study comparing males hospitalized for first-episode psychosis with and without recent cannabis exposure, researchers found significant differences. At admission, the cannabis-associated group had fewer negative symptoms, more prominent depressive and manic symptoms, and a distinct EEG signature suggesting lower cortical excitatory-inhibitory balance. Despite similar overall cognitive performance at admission, the cannabis group showed significant cognitive improvement after four weeks of inpatient treatment, which was not seen in the non-cannabis group. Critically, follow-up revealed that resuming cannabis use was strongly associated with rehospitalization, highlighting its role in illness recurrence. These findings support the idea that cannabis-associated psychosis may be a distinct subtype, which could have important implications for diagnosis and treatment planning.

This week also saw several papers on novel therapeutic approaches, spanning from large-scale brain circuits down to the molecular level. A review in *Biological Psychiatry* examines the current state of magnetic resonance-guided high-intensity focused ultrasound, or HIFU, for psychiatric disorders [2]. This incisionless technology allows for millimeter-precision thermal ablation of brain circuits. Early trials for severe, treatment-resistant OCD and major depression show clinically meaningful symptom reduction and sustained quality of life improvements with a favorable safety profile. However, the authors caution that the field is still young. Challenges include patient-specific skull characteristics that can limit energy delivery, heterogeneous outcomes, and a lack of consensus on optimal targets. Complementing this, a study in *The American Journal of Psychiatry* aims to refine targeting for an existing neuromodulation technique, transcranial magnetic stimulation, for nicotine addiction [8]. By analyzing brain lesion locations that led to smoking cessation and TMS stimulation sites that reduced withdrawal, researchers identified a common underlying brain circuit. They found that TMS sites with stronger functional connectivity to this lesion-derived circuit were associated with greater improvements in nicotine withdrawal symptoms. This convergence of lesion and TMS data points to a specific, testable circuit in areas like the frontopolar and posterior parietal cortex that could be targeted to enhance TMS efficacy for smoking cessation. Beyond neuromodulation, a randomized trial in *Nature Medicine* evaluated a novel intervention for LGBTQ+ youth at risk of self-harm [1]. Participants were randomized to receive safety planning alone or safety planning plus Purrble, a socially assistive robot designed to support emotion regulation through tactile interaction. Over 13 weeks, the group with the robot reported significantly fewer difficulties with emotion regulation and also had lower symptoms of anxiety and depression compared to the control group. However, the intervention did not have a significant effect on self-harm itself. This suggests that such tools may be a scalable way to support emotion regulation, but their direct impact on high-risk behaviors needs further study. Finally, digging into the fundamental biology of neurodegeneration, a report in *Science Translational Medicine* provides a mechanistic link between TDP-43 pathology, seen in ALS and frontotemporal dementia, and neuronal dysfunction [5]. Researchers identified that the loss of nuclear TDP-43 leads to errors in RNA splicing, creating so-called cryptic exons in genes critical for synaptic function and membrane excitability. Using human stem cell-derived neurons, they showed this leads to impaired synaptic transmission. Importantly, they demonstrated that using antisense oligonucleotides to suppress these splicing errors could partially or almost fully rescue the neuronal deficits. This work establishes cryptic splicing as a direct driver of neuronal dysfunction and points toward a potential therapeutic strategy for these devastating diseases.

Several papers this week highlight the persistent challenges in delivering effective care within complex systems. In youth mental health, a mixed-methods systematic review in *The Lancet Psychiatry* investigated interventions to reduce the use of coercive measures like restraint and seclusion in psychiatric and residential care [3]. The synthesis of 26 studies revealed that the most effective approaches are those targeting organizational culture and staff practice. Interventions that focused on relational care, staff reflexivity, and institutional support were associated with moderate-to-large reductions in coercion, whereas child-focused behavioral interventions had smaller or more uncertain effects. This underscores that reducing coercion requires a systemic shift, not just a focus on the individual child. Also focusing on youth, a trial from the *Journal of Child Psychology and Psychiatry* evaluated a brief, non-specialist-delivered family intervention for adolescents in humanitarian settings in Lebanon [4]. The results were mixed. Compared to a waitlist control, the intervention led to significant improvements in adolescent-reported family functioning, caregiver mental health, and parenting practices. However, it did not produce a significant change in the adolescents' own psychological distress. Furthermore, the gains in family functioning and caregiver well-being dropped at the 3-month follow-up. This study shows the potential of family-systems work but also illustrates the difficulty of achieving sustained, direct impact on adolescent mental health in high-stress environments. A similar story of challenge comes from the field of substance use disorders. A large meta-analysis in *JAMA Psychiatry* pooled individual participant data from 12 randomized controlled trials of pharmacotherapies for stimulant use disorder sponsored by the United States National Institute on Drug Abuse [7]. The analysis, including over 2000 participants, found no significant difference between the pooled active medications and placebo for achieving either abstinence or reduced use of cocaine or methamphetamine. This sobering result underscores the lack of effective pharmacological options. While one medication, cabergoline, did show a signal for increasing rates of reduced cocaine use, the overall picture remains negative. The authors argue that these findings highlight the importance of evaluating reduced use as a meaningful outcome, as focusing solely on abstinence may miss important therapeutic signals, even if it didn't change the overall conclusion in this particular analysis.

If you only have time for one paper this week, make it the study in *The American Journal of Psychiatry* on cannabis-associated psychosis [9]. It provides compelling evidence that this may be a distinct clinical subtype with a different presentation, neurobiology, and course, and it powerfully reinforces the critical role of cannabis abstinence in preventing relapse in these patients.

Here are the key takeaways from this week in Psychiatry. First, in patients with first-episode psychosis, consider that a cannabis-associated subtype may present differently, with fewer negative symptoms, more mood symptoms, and a better cognitive trajectory if abstinence is maintained. Relapse is tightly linked to resuming cannabis use [9]. Second, to reduce coercive measures in youth inpatient or residential settings, focus on systemic interventions that target organizational culture, staff training, and relational care, as these are more effective than child-focused approaches [3]. Third, while still investigational, EEG microstate instability is emerging as a potential biomarker to predict psychosis conversion in high-risk individuals [6]. Fourth, for patients with nicotine addiction, TMS outcomes may be improved by targeting a specific brain circuit identified through both lesion and stimulation data, primarily involving the frontopolar and posterior parietal cortices [8]. And finally, we still lack a proven pharmacotherapy for stimulant use disorder. A major meta-analysis found no overall benefit for nine different medications, reinforcing the need for novel treatment development and highlighting the ongoing debate about using reduced use as a primary trial endpoint [7].

That's your roundup for This Week in Psychiatry. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.