Cannabis use and familial vulnerability to psychosis have been associated with social cognition deficits. This study examined the potential relationship between cannabis use and cognitive biases underlying social cognition and functioning in patients with first episode psychosis (FEP), their siblings, and controls.

We analyzed a sample of 543 participants with FEP, 203 siblings, and 1168 controls from the EU-GEI study using a correlational design. We used logistic regression analyses to examine the influence of clinical group, lifetime cannabis use frequency, and potency of cannabis use on cognitive biases, accounting for demographic and cognitive variables.

FEP patients showed increased odds of facial recognition processing (FRP) deficits (OR = 1.642, CI 1.123–2.402) relative to controls but not of speech illusions (SI) or jumping to conclusions (JTC) bias, with no statistically significant differences relative to siblings. Daily and occasional lifetime cannabis use were associated with decreased odds of SI (OR = 0.605, CI 0.368–0.997 and OR = 0.646, CI 0.457–0.913 respectively) and JTC bias (OR = 0.625, CI 0.422–0.925 and OR = 0.602, CI 0.460–0.787 respectively) compared with lifetime abstinence, but not with FRP deficits, in the whole sample. Within the cannabis user group, low-potency cannabis use was associated with increased odds of SI (OR = 1.829, CI 1.297–2.578, FRP deficits (OR = 1.393, CI 1.031–1.882, and JTC (OR = 1.661, CI 1.271–2.171) relative to high-potency cannabis use, with comparable effects in the three clinical groups.

Our findings suggest increased odds of cognitive biases in FEP patients who have never used cannabis and in low-potency users. Future studies should elucidate this association and its potential implications.

Moderate-to-vigorous physical activity (MVPA) is beneficial for health, and reducing sedentary behavior (SB) is recommended in international guidelines. People with mental illnesses are at higher risk of preventable diseases than the general population, partly attributable to lower MVPA and higher SB. Self-determination theory provides a framework for understanding how motivation regulates behavior. This study aimed to evaluate the contribution of different forms of motivation for physical activity (amotivation, controlled, autonomous) to MVPA and SB in people with mental illnesses.

Cross-sectional self-reported and accelerometer-derived MVPA and SB in people with a range of mental illnesses across four countries were pooled for analysis (Australia, Belgium, England, Uganda). Motivation for physical activity was measured using the Behavioural Regulation in Exercise Questionnaire (BREQ). Regression analyses were used to investigate the association of MVPA and SB with amotivation, controlled, autonomous motivations, controlling for mental health and demographic variables.

Autonomous motivation was associated with 31% higher self-reported MVPA, and amotivation and controlled motivation were associated with 18% and 11% lower self-reported MVPA, respectively (n = 654). In contrast, controlled motivation was positively associated with SB (n = 189). Having physical comorbidities or an alcohol use disorder was associated with lower MVPA (n = 318). Sub-analyses with accelerometer-derived MVPA and SB (n = 139 and n = 145) did not reveal any associations with motivational forms.

Findings with an international sample support the universal relevance of motivation in promoting health-related behavior. Strategies for facilitating autonomous motivation should be utilized by health professionals seeking to support people with mental illnesses to become physically active.

Psychedelic drugs are a focus of interest in the treatment of depression and other disorders but there are longstanding concerns about possible adverse psychiatric consequences. Because the relevant literature is largely informal, the seriousness of these risks is difficult to evaluate.

Searches were made for case reports of schizophrenia-spectrum, affective or other psychiatric disorders after use of psychedelic drugs. Case reports of flashbacks were also searched for. Individuals with recent use of other drugs (apart from cannabis and alcohol) and/or a previous history of major psychiatric disorder were excluded. Symptoms were tabulated using the Syndrome Check List of the Present State Examination (PSE-9).

We found 17 case reports of schizophrenia spectrum disorder, 17 of affective disorder (depression, mania, or both), 3 cases of anxiety, 1 of depersonalization, and 1 of unclassifiable illness. The states could develop after a single use of the drug (5/17 schizophrenia; 6/17 affective disorder), and duration was highly variable. Recovery was the rule in cases of affective disorder but not in schizophrenia spectrum disorder. Twelve of 29 cases of flashbacks showed psychiatric symptomatology definitely outlasting the attacks, mainly anxiety (5 cases) and depression (8 cases). Flashback symptoms resolved within twelve months in approximately half of the cases but in a few persisted for years.

Reliable descriptions of schizophrenia spectrum disorder and major affective disorder after psychedelic drug use disorder exist but are relatively uncommon. Flashbacks are sometimes but not always associated with psychiatric symptomatology, mainly anxiety or depression.

Psychiatric diagnosis is based on categorical diagnostic classification, yet similarities in genetics and clinical features across disorders suggest that these classifications share commonalities in neurobiology, particularly regarding neurotransmitters. Glutamate (Glu) and gamma-aminobutyric acid (GABA), the brain's primary excitatory and inhibitory neurotransmitters, play critical roles in brain function and physiological processes.

We examined the levels of Glu, combined glutamate and glutamine (Glx), and GABA across psychiatric disorders by pooling data from 121 1H-MRS studies and further divided the sample based on Axis I disorders.

Statistically significant differences in GABA levels were found in the combined psychiatric group compared with healthy controls (Hedge's g = −0.112, p = 0.008). Further analyses based on brain regions showed that brain GABA levels significantly differed across Axis I disorders and controls in the parieto-occipital cortex (Hedge's g = 0.277, p = 0.019). Furthermore, GABA levels were reduced in affective disorders in the occipital cortex (Hedge's g = −0.468, p = 0.043). Reductions in Glx levels were found in neurodevelopmental disorders (Hedge's g = −0.287, p = 0.022). Analysis focusing on brain regions suggested that Glx levels decreased in the frontal cortex (Hedge's g = −0.226, p = 0.025), and the reduction of Glu levels in patients with affective disorders in the frontal cortex is marginally significant (Hedge's g = −0.172, p = 0.052). When analyzing the anterior cingulate cortex and prefrontal cortex separately, reductions were only found in GABA levels in the former (Hedge's g = − 0.191, p = 0.009) across all disorders.

Altered glutamatergic and GABAergic metabolites were found across psychiatric disorders, indicating shared dysfunction. We found reduced GABA levels across psychiatric disorders and lower Glu levels in affective disorders. These results highlight the significance of GABA and Glu in psychiatric etiology and partially support rethinking current diagnostic categories.

Epidemiological samples provide opportunity to understand the development of mental health trajectories to better understand whether such epidemiological data can help to plan and modify service delivery for youth mental health. Variation between countries is not well understood and thus applying evidence from other countries to national strategies limits support service policy and planning. We therefore examine developmental patterns of youth mental health across different countries using the Growing Up in Ireland (GUI) Cohorts, with comparison to existing UK longitudinal cohort data (Millennium Cohort Study, MCS; Growing up in Scotland, GUS).

Youth mental health problems within each cohort across development (5–17/18 years) were assessed using parent reported Strengths and Difficulties Questionnaire (SDQ) scores. Using latent growth curve analyses, we examined trajectories of emotional, conduct, and hyperactivity problems for boys and girls, separately for each cohort.

Across cohorts, we observed similar developmental patterns for emotional, conduct, and hyperactivity problems. However, the GUI emotional problems in Ireland emerged earlier than in the UK. By adolescence, GUI emotional scores were similar to the UK, suggesting that the differences in emotional problems between the ROI and UK had narrowed by adolescence. Covariates also had different associations with youth mental health trajectories across cohorts.

Utilizing multiple nationally representative cohort longitudinal datasets can help inform clinically meaningful conclusions and potential recommendations on population level multi-tiered service needs and development in the area of child and adolescent mental health support and future provision.

To explore the association of cardiovascular-kidney-metabolic (CKM) health with the risk of depression and anxiety and to investigate the joint association of CKM health and social connection with depression and anxiety.

This prospective cohort study included 344 956 participants from the UK Biobank. CKM syndrome was identified as a medical condition with the presence of metabolic risk factors, cardiovascular disease, and chronic kidney disease, and was classified into five stages (stage 0–4) in this study. Loneliness and social isolation status were determined by self-reported questionnaires. Cox proportional hazards models were applied for analyses.

Compared with participants in stage 0, the HRs for depression were 1.17 (95% CI 1.10–1.25), 1.40 (95% CI 1.33–1.48), and 2.14 (95% CI 1.98–2.31) for participants in stage 1, 2–3, and 4, respectively. Similarly, participants in stage 2–3 (HR = 1.20, 95% CI 1.14–1.26) and stage 4 (HR = 1.63, 95% CI 1.51–1.75) had greater risks of incident anxiety. We found additive interactions between loneliness and CKM health on the risk of depression and anxiety. Participants simultaneously reported being lonely and in stage 4 had the greatest risk of depression (HR = 4.44, 95% CI 3.89–5.07) and anxiety (HR = 2.58, 95% CI 2.21–3.01) compared with those without loneliness and in stage 0. We also observed an additive interaction between social isolation and CKM health on the risk of depression.

Our findings suggest the importance of comprehensive interventions to improve CKM health and social connection to reduce the disease burden of depression and anxiety.

Unpredictability is a core but understudied dimension of adversities and has been receiving increasing attention recently. The effects of unpredictability on psychopathology and the underlying neural mechanisms, however, remain unclear. It is also unknown how unpredictability interacts with other dimensions of adversities in predicting brain development and psychopathology of youth.

We applied cluster robust standard errors to examine how unpredictability was associated with the developmental changes in resting-state functional connectivity (rsFC) of large-scale brain networks implicated in psychopathology, as well as the moderating role of deprivation, using data from the Adolescent Brain Cognitive Development (ABCD) study, which included four measurements from baseline (mean ± s.d. age, 119.13 ± 7.51 months; 2815 females) to 3-year follow-up (N = 5885).

After controlling for threat, unpredictability was associated with a smaller increase in rsFC within default mode network (DMN) and a smaller decrease in rsFC between cingulo-opercular network (CON) and DMN. Neighborhood educational deprivation moderated the associations between unpredictability and changes in rsFC within DMN and fronto-parietal network (FPN), as well as between CON and DMN. A smaller decrease in rsFC between CON and DMN mediated the association between unpredictability and externalizing problems. Neighborhood educational deprivation moderated the indirect pathway from unpredictability to externalizing problems via a smaller decrease in CON-DMN rsFC.

Our findings shed light on the neural mechanisms underlying the associations between unpredictability and adolescents' psychopathology and the moderating role of deprivation, highlighting the significance of providing stable environment and abundant educational opportunities to facilitate optimal development.

Mental disorders and physical-health conditions frequently co-occur, impacting treatment outcomes. While most prior research has focused on single pairs of mental disorders and physical-health conditions, this study explores broader associations between multiple mental disorders and physical-health conditions.

Using the Norwegian primary-care register, this population-based cohort study encompassed all 2 203 553 patients born in Norway from January 1945 through December 1984, who were full-time residents from January 2006 until December 2019 (14 years; 363 million person-months). Associations between seven mental disorders (sleep disturbance, anxiety, depression, acute stress reaction, substance-use disorders, phobia/compulsive disorder, psychosis) and 16 physical-health conditions were examined, diagnosed according to the International Classification of Primary Care.

Of 112 mental-disorder/physical-health condition pairs, 96% of associations yielded positive and significant ORs, averaging 1.41 and ranging from 1.05 (99.99% CI 1.00–1.09) to 2.38 (99.99% CI 2.30–2.46). Across 14 years, every mental disorder was associated with multiple different physical-health conditions. Across 363 million person-months, having any mental disorder was associated with increased subsequent risk of all physical-health conditions (HRs:1.40 [99.99% CI 1.35–1.45] to 2.85 [99.99% CI 2.81–2.89]) and vice versa (HRs:1.56 [99.99% CI 1.54–1.59] to 3.56 [99.99% CI 3.54–3.58]). Associations were observed in both sexes, across age groups, and among patients with and without university education.

The breadth of associations between virtually every mental disorder and physical-health condition among patients treated in primary care underscores a need for integrated mental and physical healthcare policy and practice. This remarkable breadth also calls for research into etiological factors and underlying mechanisms that can explain it.

Borderline personality disorder (BPD) is a debilitating condition characterized by pervasive instability across multiple major domains of functioning. The majority of persons with BPD engage in self-injury and up to 10% die by suicide – rendering persons with this condition at exceptionally elevated risk of comorbidity and premature mortality. Better characterization of clinical risk factors among persons with BPD who die by suicide is urgently needed.

We examined patterns of medical and psychiatric diagnoses (1580 to 1700 Phecodes) among persons with BPD who died by suicide (n = 379) via a large suicide death data resource and biobank. In phenotype-based phenome-wide association tests, we compared these individuals to three other groups: (1) persons who died by suicide without a history of BPD (n = 9468), (2) persons still living with a history of BPD diagnosis (n = 280), and (3) persons who died by suicide with a different personality disorder (other PD n = 589).

Multivariable logistic regression models revealed that persons with BPD who died by suicide were more likely to present with co-occurring psychiatric diagnoses, and have a documented history of self-harm in the medical system prior to death, relative to suicides without BPD. Posttraumatic stress disorder was more elevated among those with BPD who died by suicide relative to the other PD group.

We found significant differences among persons with BPD who died by suicide and all other comparison groups. Such differences may be clinically informative for identifying high-risk subtypes and providing targeted intervention approaches.

Genetic vulnerability to mental disorders has been associated with coronavirus disease-19 (COVID-19) outcomes. We explored whether polygenic risk scores (PRSs) for several mental disorders predicted poorer clinical and psychological COVID-19 outcomes in people with pre-existing depression.

Data from three assessments of the Australian Genetics of Depression Study (N = 4405; 52.2 years ± 14.9; 76.2% females) were analyzed. Outcomes included COVID-19 clinical outcomes (severe acute respiratory syndrome coronavirus 2 [SARS-CoV-2] infection and long COVID, noting the low incidence of COVID-19 cases in Australia at that time) and COVID-19 psychological outcomes (COVID-related stress and COVID-19 burnout). Predictors included PRS for depression, bipolar disorder, schizophrenia, and anxiety. The associations between these PRSs and the outcomes were assessed with adjusted linear/logistic/multinomial regressions. Mediation (N = 4338) and moderation (N = 3326) analyses were performed to explore the potential influence of anxiety symptoms and resilience on the identified associations between the PRSs and COVID-19 psychological outcomes.

None of the selected PRS predicted SARS-CoV-2 infection or long COVID. In contrast, the depression PRS predicted higher levels of COVID-19 burnout. Anxiety symptoms fully mediated the association between the depression PRS and COVID-19 burnout. Resilience did not moderate this association.

A higher genetic risk for depression predicted higher COVID-19 burnout and this association was fully mediated by anxiety symptoms. Interventions targeting anxiety symptoms may be effective in mitigating the psychological effects of a pandemic among people with depression.

The hippocampal formation represents a key region in the pathophysiology of schizophrenia. Aerobic exercise poses a promising add-on treatment to potentially counteract structural impairments of the hippocampal formation and associated symptomatic burden. However, current evidence regarding exercise effects on the hippocampal formation in schizophrenia is largely heterogeneous. Therefore, we conducted a systematic review and meta-analysis to assess the impact of aerobic exercise on total hippocampal formation volume. Additionally, we used data from a recent multicenter randomized-controlled trial to examine the effects of aerobic exercise on hippocampal formation subfield volumes and their respective clinical implications.

The meta-analysis comprised six studies that investigated the influence of aerobic exercise on total hippocampal formation volume compared to a control condition with a total of 186 people with schizophrenia (100 male, 86 female), while original data from 29 patients (20 male, 9 female) was considered to explore effects of six months of aerobic exercise on hippocampal formation subfield volumes.

Our meta-analysis did not demonstrate a significant effect of aerobic exercise on total hippocampal formation volume in people with schizophrenia (g = 0.33 [−0.12 to 0.77]), p = 0.15), but our original data suggested significant volume increases in certain hippocampal subfields, namely the cornu ammonis and dentate gyrus.

Driven by the necessity of better understanding the pathophysiology of schizophrenia, the present work underlines the importance to focus on hippocampal formation subfields and to characterize subgroups of patients that show neuroplastic responses to aerobic exercise accompanied by corresponding clinical improvements.

ADHD symptoms are associated with emotional problems such as depressive and anxiety symptoms from early childhood to adulthood, with the association increasing with age. A shared aetiology and/or a causal relationship could explain their correlation. In the current study, we explore these explanations for the association between ADHD symptoms and emotional problems from childhood to adulthood.

Data were drawn from the Twins Early Development Study (TEDS), including 3675 identical and 7063 non-identical twin pairs. ADHD symptoms and emotional symptoms were reported by parents from childhood to adulthood. Self-report scales were included from early adolescence. Five direction of causation (DoC) twin models were fitted to distinguish whether associations were better explained by shared aetiology and/or causal relationships in early childhood, mid-childhood, early adolescence, late adolescence, and early adulthood. Follow-up analyses explored associations for the two subdomains of ADHD symptoms, hyperactivity-impulsivity and inattention, separately.

The association between ADHD symptoms and emotional problems increased in magnitude from early childhood to adulthood. In the best-fitting models, positive genetic overlap played an important role in this association at all stages. A negative causal effect running from ADHD symptoms to emotional problems was also detected in early childhood and mid-childhood. When distinguishing ADHD subdomains, the apparent protective effect of ADHD symptoms on emotional problems in childhood was mostly driven by hyperactivity-impulsivity.

Genetic overlap plays an important role in the association between ADHD symptoms and emotional problems. Hyperactivity-impulsivity may protect children from emotional problems in childhood, but this protective effect diminishes after adolescence.

First-episode schizophrenia (FES) is a progressive psychiatric disorder influenced by genetics, environmental factors, and brain function. The functional gradient deficits of drug-naïve FES and its relationship to gene expression profiles and treatment outcomes are unknown.

In this study, we engaged a cohort of 116 FES and 100 healthy controls (HC), aged 7 to 30 years, including 15 FES over an 8-week antipsychotic medication regimen. Our examination focused on primary-to-transmodal alterations in voxel-based connection gradients in FES. Then, we employed network topology, Neurosynth, postmortem gene expression, and support vector regression to evaluate integration and segregation functions, meta-analytic cognitive terms, transcriptional patterns, and treatment predictions.

FES displayed diminished global connectome gradients (Cohen's d = 0.32–0.57) correlated with compensatory integration and segregation functions (Cohen's d = 0.31–0.36). Predominant alterations were observed in the default (67.6%) and sensorimotor (21.9%) network, related to high-order cognitive functions. Furthermore, we identified notable overlaps between partial least squares (PLS1) weighted genes and dysregulated genes in other psychiatric conditions. Genes linked with gradient alterations were enriched in synaptic signaling, neurodevelopment process, specific astrocytes, cortical layers (layer II and IV), and developmental phases from late/mid fetal to young adulthood. Additionally, the onset age influenced the severity of FES, with discernible differences in connection gradients between minor- and adult-FES. Moreover, the connectivity gradients of FES at baseline significantly predicted treatment outcomes.

These results offer significant theoretical foundations for elucidating the intricate interplay between macroscopic functional connection gradient changes and microscopic transcriptional patterns during the onset and progression of FES.

Involuntary admissions to psychiatric hospitals are on the rise. If patients at elevated risk of involuntary admission could be identified, prevention may be possible. Our aim was to develop and validate a prediction model for involuntary admission of patients receiving care within a psychiatric service system using machine learning trained on routine clinical data from electronic health records (EHRs).

EHR data from all adult patients who had been in contact with the Psychiatric Services of the Central Denmark Region between 2013 and 2021 were retrieved. We derived 694 patient predictors (covering e.g. diagnoses, medication, and coercive measures) and 1134 predictors from free text using term frequency-inverse document frequency and sentence transformers. At every voluntary inpatient discharge (prediction time), without an involuntary admission in the 2 years prior, we predicted involuntary admission 180 days ahead. XGBoost and elastic net models were trained on 85% of the dataset. The models with the highest area under the receiver operating characteristic curve (AUROC) were tested on the remaining 15% of the data.

The model was trained on 50 634 voluntary inpatient discharges among 17 968 patients. The cohort comprised of 1672 voluntary inpatient discharges followed by an involuntary admission. The best XGBoost and elastic net model from the training phase obtained an AUROC of 0.84 and 0.83, respectively, in the test phase.

A machine learning model using routine clinical EHR data can accurately predict involuntary admission. If implemented as a clinical decision support tool, this model may guide interventions aimed at reducing the risk of involuntary admission.

Major psychiatric disorders (MPDs) are delineated by distinct clinical features. However, overlapping symptoms and transdiagnostic effectiveness of medications have challenged the traditional diagnostic categorisation. We investigate if there are shared and illness-specific disruptions in the regional functional efficiency (RFE) of the brain across these disorders.

We included 364 participants (118 schizophrenia [SCZ], 80 bipolar disorder [BD], 91 major depressive disorder [MDD], and 75 healthy controls [HCs]). Resting-state fMRI was used to caclulate the RFE based on the static amplitude of low-frequency fluctuation, regional homogeneity, and degree centrality and corresponding dynamic measures indicating variability over time. We used principal component analysis to obtain static and dynamic RFE values. We conducted functional and genetic annotation and enrichment analysis based on abnormal RFE profiles.

SCZ showed higher static RFE in the cortico-striatal regions and excessive variability in the cortico-limbic regions. SCZ and MDD shared lower static RFE with higher dynamic RFE in sensorimotor regions than BD and HCs. We observed association between static RFE abnormalities with reward and sensorimotor functions and dynamic RFE abnormalities with sensorimotor functions. Differential spatial expression of genes related to glutamatergic synapse and calcium/cAMP signaling was more likely in the regions with aberrant RFE.

SCZ shares more regions with disrupted functional integrity, especially in sensorimotor regions, with MDD rather than BD. The neural patterns of these transdiagnostic changes appear to be potentially driven by gene expression variations relating to glutamatergic synapses and calcium/cAMP signaling. The aberrant sensorimotor, cortico-striatal, and cortico-limbic integrity may collectively underlie neurobiological mechanisms of MPDs.

We aimed to identify the common types of outcome trajectories for patients with psychosis who take up specialist psychological therapy for persecutory delusions. Knowing the different potential responses to therapy can inform expectations. Further, determining predictors of different outcomes may help in understanding who may benefit.

We analyzed delusion conviction data from 767 therapy sessions with 64 patients with persistent persecutory delusions (held with at least 60% conviction) who received a six-month psychological intervention (Feeling Safe) during a clinical trial. Latent class trajectory analysis was conducted to identify groups with distinct outcome profiles. The trajectories were validated against independent assessments, including a longer-term follow-up six months after the end of therapy. We also tested potential predictors of the trajectories.

There were four outcome trajectories: (1) Very high delusion conviction/Little improvement (n = 14, 25%), (2) Very high delusion conviction/Large improvement (n = 9, 16%), (3) High delusion conviction/Moderate improvement (n = 17, 31%) and (4) High delusion conviction/Large improvement (n = 15, 27%). The groups did not differ in initial overall delusion severity. The trajectories were consistent with the independent assessments and sustained over time. Three factors predicted trajectories: persecutory delusion conviction, therapy expectations, and positive beliefs about other people.

There are variable responses to psychological therapy for persecutory delusions. Patients with very high delusion conviction can have excellent responses to therapy, though this may take a little longer to observe and such high conviction reduces the likelihood of positive responses. A trajectory approach requires testing in larger datasets but may prove highly informative.

Short-term exposure to antipsychotics has proven to be beneficial. However, naturalistic studies are lacking regarding the long-term use of antipsychotics. This study aimed to investigate changes in use of antipsychotics over 20 years after a first-episode schizophrenia.

This study is part of the Danish OPUS trial (1998–2000), including 496 participants with first-episode schizophrenia. Participants were reassessed four times over 20 years. The main outcomes were days on medication, redeemed prescriptions of clozapine, psychiatric hospitalizations, and employment.

At the 20-year follow-up, an attrition of 71% was detected. In total, 143 out of 496 participated, with 36% (n = 51) in remission-of-psychotic-symptoms-off-medication. The lowest number of days on medication (mean [s.d.], 339 [538] days) was observed in this group over 20 years. Register data on redeemed antipsychotics were available for all trial participants (n = 416). Individuals in treatment with antipsychotics (n = 120) at the 20-year follow-up had spent significantly more days in treatment (5405 [1857] v. 1434 [1819] mean days, p = 0.00) and more had ever redeemed a prescription of clozapine (25% v. 7.8%, p = 0.00) than individuals who had discontinued antipsychotics (n = 296). Further, discontinuers had significantly higher employment at the 20-year follow-up (28.4% v. 12.5%, p = 0.00).

In a cohort of individuals with first-episode schizophrenia, 36% were in remission-of-psychotic-symptoms-off-medication. However, high attrition was detected, potentially affecting study results by inflating results from individuals with favorable outcomes. From register data, free from attrition, approximately 30% were in treatment with antipsychotics, and 70% had discontinued antipsychotics. Individuals in treatment had the least favorable outcomes, implying greater illness severity.

Machine learning (ML) has developed classifiers differentiating patient groups despite concerns regarding diagnostic reliability. An alternative strategy, used here, is to develop a functional classifier (hyperplane) (e.g. distinguishing the neural responses to received reward v. received punishment in typically developing (TD) adolescents) and then determine the functional integrity of the response (reward response distance from the hyperplane) in adolescents with externalizing and internalizing conditions and its associations with symptom clusters.

Two hundred and ninety nine adolescents (mean age = 15.07 ± 2.30 years, 117 females) were divided into three groups: a training sample of TD adolescents where the Support Vector Machine (SVM) algorithm was applied (N = 65; 32 females), and two test groups– an independent sample of TD adolescents (N = 39; 14 females) and adolescents with a psychiatric diagnosis (major depressive disorder (MDD), generalized anxiety disorder (GAD), attention deficit hyperactivity disorder (ADHD) & conduct disorder (CD); N = 195, 71 females).

SVM ML analysis identified a hyperplane with accuracy = 80.77%, sensitivity = 78.38% and specificity = 88.99% that implicated feature neural regions associated with reward v. punishment (e.g. nucleus accumbens v. anterior insula cortices). Adolescents with externalizing diagnoses were significantly less likely to show a normative and significantly more likely to show a deficient reward response than the TD samples. Deficient reward response was associated with elevated CD, MDD, and ADHD symptoms.

Distinguishing the response to reward relative to punishment in TD adolescents via ML indicated notable disruptions in this response in patients with CD and ADHD and associations between reward responsiveness and CD, MDD, and ADHD symptom severity.