Björn Schlier
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Björn Schlier has a total of 157 co-authors. Here are the top co-authors by number of shared publications:
- Tania Marie Lincoln
Universität Hamburg
30 shared publications Tania M. Lincoln
Univ Hamburg, Inst Psychol, Fac Psychol & Human Movement Sci, Clin Psychol & Psychotherapy, Hamburg, Germany
10 shared publicationsClara Strauss
School of Psychology, University of Sussex, Brighton, UK
3 shared publications- Edo Sebastian Jaya
Faculty of Psychology, Universitas Indonesia, Depok, Indonesia
3 shared publications Felix Strakeljahn
3 shared publications- Frank Larøi
Department of Biological and Medical Psychology, University of Bergen, Bergen, Norway
3 shared publications
Latest Publications
- Open Access
- 20 October 2026
- Published by Center for Open Science in PsyArXiv
Towards precision psychotherapy: Leveraging session-level data to predict session-by-session treatment outcomes
Theoretical Background: Clinical support tools can potentially enhance cognitive behavioral therapy (CBT) by providing session-by-session treatment recommendations. To enable such tools, prediction models are needed that can accurately track therapy outcomes based on assessable session data. This study developed models that predict session-by-session mental health levels and identified variables that influence these predictions.Methods: We analyzed data from a monocentric, naturalistic therapy-process study of outpatients with affective, anxiety, and psychotic disorders (n=325) participating in up to 24 weekly sessions of CBT in routine care in Germany. Within a nested cross-validation approach, we used mixed-effect random forest models to predict session-by-session scores on a brief self-report measure of mental health (i.e., symptoms, mood, distress, functioning) assessed in the next session. Included predictors were baseline psychopathology, type of CBT interventions conducted in the prior session, and patient- and therapist-reported mechanisms of change (e.g., cognitive change, self-understanding, coping) assessed in the prior session. Variable importance was examined using SHAP values.Results: The mixed-effects random forest predicting next-session outcome achieved an R²=0.25 across the test folds. The most influential predictors were baseline psychopathology and prior session self-reported behavioral mechanisms of change, i.e., engagement in meaningful/positive activities and coping.Conclusion: Session-by-session prediction of mental health outcomes using routine outcome monitoring data yielded promising predictive accuracy. Self-reported mechanisms of change are particularly valuable for predicting session-by-session mental health levels. These findings may suggest that activating behavioral mechanisms of change in therapy may have a stronger impact on short-term mental health than activating cognitive mechanisms of change.
- 20 October 2026
- Published by Center for Open Science in PsyArXiv
Open AccessTowards precision psychotherapy: Leveraging session-level data to predict session-by-session treatment outcomes
Theoretical Background: Clinical support tools can potentially enhance cognitive behavioral therapy (CBT) by providing session-by-session treatment recommendations. To enable such tools, prediction models are needed that can accurately track therapy outcomes based on assessable session data. This study developed models that predict session-by-session mental health levels and identified variables that influence these predictions.Methods: We analyzed data from a monocentric, naturalistic therapy-process study of outpatients with affective, anxiety, and psychotic disorders (n=325) participating in up to 24 weekly sessions of CBT in routine care in Germany. Within a nested cross-validation approach, we used mixed-effect random forest models to predict session-by-session scores on a brief self-report measure of mental health (i.e., symptoms, mood, distress, functioning) assessed in the next session. Included predictors were baseline psychopathology, type of CBT interventions conducted in the prior session, and patient- and therapist-reported mechanisms of change (e.g., cognitive change, self-understanding, coping) assessed in the prior session. Variable importance was examined using SHAP values.Results: The mixed-effects random forest predicting next-session outcome achieved an R²=0.25 across the test folds. The most influential predictors were baseline psychopathology and prior session self-reported behavioral mechanisms of change, i.e., engagement in meaningful/positive activities and coping.Conclusion: Session-by-session prediction of mental health outcomes using routine outcome monitoring data yielded promising predictive accuracy. Self-reported mechanisms of change are particularly valuable for predicting session-by-session mental health levels. These findings may suggest that activating behavioral mechanisms of change in therapy may have a stronger impact on short-term mental health than activating cognitive mechanisms of change.
- Open Access
- 20 October 2026
- Published by Center for Open Science in PsyArXiv
The predictive value of trans related stressors and resilience factors for mental health outcomes – a machine learning approach
Background: Trans people experience elevated levels of mental health problems compared to the general population. Trans-specific psychosocial stressors and resilience factors have been shown to account for part of the variance in mental health. However, most studies have included only a limited subset of relevant stressors and resilience factors, leaving unclear both the amount of variance that can be explained by a comprehensive predictor set and which individual stressors and resilience factors are most informative for mental health outcomes.Methods: A total of 447 adult trans individuals participated in a cross-sectional online study and completed measures assessing trans-specific stressors and resilience factors, as well as well-being, social functioning, and overall mental health problems. Prediction models were estimated using random forests within a nested cross-validation framework. Shapley additive explanations were applied to identify the most influential predictors for each mental health outcome.Results: Prediction models based on psychosocial stressors and resilience factors, and demographic data achieved R² = 26% for well-being, R² = 32% for social functioning, and R² = 35% for overall mental health problems. Gender dysphoria, rumination about others’ perception, and internalized transnegativity consistently ranked among the five most important predictors across all models.Discussion: Trans-specific stressors and resilience factors explain a substantial proportion of variance in mental health outcomes. Nevertheless, incorporating additional predictors, including general risk and protective factors for mental health, may further enhance predictive accuracy and thereby improve mental health prediction in this population.
- 20 October 2026
- Published by Center for Open Science in PsyArXiv
Open AccessThe predictive value of trans related stressors and resilience factors for mental health outcomes – a machine learning approach
Background: Trans people experience elevated levels of mental health problems compared to the general population. Trans-specific psychosocial stressors and resilience factors have been shown to account for part of the variance in mental health. However, most studies have included only a limited subset of relevant stressors and resilience factors, leaving unclear both the amount of variance that can be explained by a comprehensive predictor set and which individual stressors and resilience factors are most informative for mental health outcomes.Methods: A total of 447 adult trans individuals participated in a cross-sectional online study and completed measures assessing trans-specific stressors and resilience factors, as well as well-being, social functioning, and overall mental health problems. Prediction models were estimated using random forests within a nested cross-validation framework. Shapley additive explanations were applied to identify the most influential predictors for each mental health outcome.Results: Prediction models based on psychosocial stressors and resilience factors, and demographic data achieved R² = 26% for well-being, R² = 32% for social functioning, and R² = 35% for overall mental health problems. Gender dysphoria, rumination about others’ perception, and internalized transnegativity consistently ranked among the five most important predictors across all models.Discussion: Trans-specific stressors and resilience factors explain a substantial proportion of variance in mental health outcomes. Nevertheless, incorporating additional predictors, including general risk and protective factors for mental health, may further enhance predictive accuracy and thereby improve mental health prediction in this population.
- Open Access
- 12 October 2026
- Published by Wiley in JCPP Advances
Paranoia and social anxiety: Structural and temporal relationships across adolescence
Background: Social anxiety (SA) and paranoia (PA) are commonly experienced during adolescence, but the nature of their co‐development is largely unknown. Greater clarity could significantly advance our understanding of both conditions, and of their progression into serious mental illness. Methods: Utilising the Neuroscience in Psychiatry Network 2k cohort data, an accelerated cohort design study in 14–22 year olds, pre‐registered hypotheses examined: (1) via measurement models, whether SA and PA were statistically distinguishable; (2) whether SA at a given time‐point was significantly associated with subsequent PA at the next time point; and (3) common co‐developmental trajectories through adolescence (e.g., low in both; high SA, low PA, etc.). Results: (1) Measurement models indicated that two and three factor models, with SA and PA as related but separate constructs, best fit the data across all age groups, (2) in cross‐lagged panel analysis, SA was a more significant predictor of subsequent (approximately 1 year later) PA than vice versa; (3) Six common patterns in how SA and PA developed together over time (i.e., co‐developmental trajectories) were detected, whereby low levels of each were most common, followed by clusters with elevated SA but not PA, followed by mid and descending SA and mid PA, or elevated levels on each. Conclusion: SA and PA are commonly comorbid in adolescence. More evidence was obtained for the SA‐fuelling‐PA hypothesis than vice versa, but nonetheless we also detected a PA‐to‐SA pathway in the youngest age group which requires further investigation. Understanding the social and cognitive factors that account for escalation from SA into PA is a key future direction.
- 12 October 2026
- Published by Wiley in JCPP Advances
Open AccessParanoia and social anxiety: Structural and temporal relationships across adolescence
Background: Social anxiety (SA) and paranoia (PA) are commonly experienced during adolescence, but the nature of their co‐development is largely unknown. Greater clarity could significantly advance our understanding of both conditions, and of their progression into serious mental illness. Methods: Utilising the Neuroscience in Psychiatry Network 2k cohort data, an accelerated cohort design study in 14–22 year olds, pre‐registered hypotheses examined: (1) via measurement models, whether SA and PA were statistically distinguishable; (2) whether SA at a given time‐point was significantly associated with subsequent PA at the next time point; and (3) common co‐developmental trajectories through adolescence (e.g., low in both; high SA, low PA, etc.). Results: (1) Measurement models indicated that two and three factor models, with SA and PA as related but separate constructs, best fit the data across all age groups, (2) in cross‐lagged panel analysis, SA was a more significant predictor of subsequent (approximately 1 year later) PA than vice versa; (3) Six common patterns in how SA and PA developed together over time (i.e., co‐developmental trajectories) were detected, whereby low levels of each were most common, followed by clusters with elevated SA but not PA, followed by mid and descending SA and mid PA, or elevated levels on each. Conclusion: SA and PA are commonly comorbid in adolescence. More evidence was obtained for the SA‐fuelling‐PA hypothesis than vice versa, but nonetheless we also detected a PA‐to‐SA pathway in the youngest age group which requires further investigation. Understanding the social and cognitive factors that account for escalation from SA into PA is a key future direction.
- Open Access
- 30 September 2026
- Published by Center for Open Science in PsyArXiv
What this trial tells us about modified manualised psychodynamic psychotherapy for schizophrenia—and what it cannot
Montag and colleagues report a randomised trial of modified psychodynamic psychotherapy for schizophrenia (MPP-S) plus treatment as usual (TAU) versus TAU alone. We welcome efforts to strengthen the evidence base for psychodynamic psychotherapy. Given the limited access to psychological treatment for people with psychosis and the, at best, moderate effects of existing pharmacological and psychological interventions this is a valuable effort to advance evidence-based treatment in psychosis. Several methodological decisions, however, complicate interpretation of its findings.
- 30 September 2026
- Published by Center for Open Science in PsyArXiv
Open AccessWhat this trial tells us about modified manualised psychodynamic psychotherapy for schizophrenia—and what it cannot
Montag and colleagues report a randomised trial of modified psychodynamic psychotherapy for schizophrenia (MPP-S) plus treatment as usual (TAU) versus TAU alone. We welcome efforts to strengthen the evidence base for psychodynamic psychotherapy. Given the limited access to psychological treatment for people with psychosis and the, at best, moderate effects of existing pharmacological and psychological interventions this is a valuable effort to advance evidence-based treatment in psychosis. Several methodological decisions, however, complicate interpretation of its findings.
- Open Access
- 17 September 2026
- Published by Center for Open Science in PsyArXiv
What are the merits of identifying as neurodivergent? Assessing the effects of neurodiversity labels on the evaluation of and stigmatizing attitudes towards people exhibiting autistic traits.
BackgroundWith the neurodiversity movement gaining traction, people with autism spectrum disorders (ASD) and other diagnoses have begun self-identifying as neurodivergent and report that this manner of disclosure boosts their self-esteem and well-being. However, little is known about how self-labelling as neurodivergent changes public perception about the people self-labelling as neurodivergent. This study explores how neurodiversity labels influence stigmatizing and empowering public perceptions of people displaying ASD-typical characteristics.MethodsIn a three-armed online-experiment, 1215 participants received a text-vignette describing a male adult with everyday life challenges typically associated with ASD. The man was either labelled as neurodivergent/neurodiverse, autistic, or not labelled. Participants then rated the described man in terms of social, general, and self-regulation competencies and answered questionnaires on stigmatizing and empowering beliefs about the presented person and/or people like him.ResultsIn the neurodiversity-label and no-label conditions, ratings on social competencies and emotional stability were higher than in the autism condition. The crucial stigmatization components blame and desired social distance were significantly lower in the neurodiversity- and autism-label conditions than in the no-label condition.ConclusionAcross all public perception outcome categories, using a neurodiversity label provided the optimal balance of benefits. This highlights how self-identifying as neurodivergent (rather than by one’s diagnosis) may be a way to circumvent some public stigma when disclosing one’s challenges and warrants further investigation into the generalizability and stability of such effects.
- 17 September 2026
- Published by Center for Open Science in PsyArXiv
Open AccessWhat are the merits of identifying as neurodivergent? Assessing the effects of neurodiversity labels on the evaluation of and stigmatizing attitudes towards people exhibiting autistic traits.
BackgroundWith the neurodiversity movement gaining traction, people with autism spectrum disorders (ASD) and other diagnoses have begun self-identifying as neurodivergent and report that this manner of disclosure boosts their self-esteem and well-being. However, little is known about how self-labelling as neurodivergent changes public perception about the people self-labelling as neurodivergent. This study explores how neurodiversity labels influence stigmatizing and empowering public perceptions of people displaying ASD-typical characteristics.MethodsIn a three-armed online-experiment, 1215 participants received a text-vignette describing a male adult with everyday life challenges typically associated with ASD. The man was either labelled as neurodivergent/neurodiverse, autistic, or not labelled. Participants then rated the described man in terms of social, general, and self-regulation competencies and answered questionnaires on stigmatizing and empowering beliefs about the presented person and/or people like him.ResultsIn the neurodiversity-label and no-label conditions, ratings on social competencies and emotional stability were higher than in the autism condition. The crucial stigmatization components blame and desired social distance were significantly lower in the neurodiversity- and autism-label conditions than in the no-label condition.ConclusionAcross all public perception outcome categories, using a neurodiversity label provided the optimal balance of benefits. This highlights how self-identifying as neurodivergent (rather than by one’s diagnosis) may be a way to circumvent some public stigma when disclosing one’s challenges and warrants further investigation into the generalizability and stability of such effects.
- Open Access
- 16 September 2026
- Published by Center for Open Science in PsyArXiv
Healthy attachment for the parents? An investigation on the association between attachment parenting and parental mental health.
Introduction. Attachment parenting (AP) became popular for it promotes responsive caregiving, but has been criticized for over-extensively translating attachment research into challenging parenting standards, mirroring detrimental intensive-parenting beliefs that adversely affect parental health. Meanwhile, evidence on AP remains limited and assessments of AP-beliefs and AP-practice are lacking. This study aims to develop and validate an AP-questionnaire and test the associations between AP and parental mental health and explore the mediating role of intensive-parenting beliefs.Methods. In a preregistered cross-sectional online-survey, 1664 parents (98.5% mothers) of children aged up to six years answered an extensive item-list assessing AP-beliefs and AP implementation into practice, an intensive parenting questionnaire, and mental health measures. Confirmatory factor analyses were used for AP-questionnaire validation and correlation- and mediation-analyses for exploring the association between AP and mental health.Results. A 26-item AP-questionnaire was extracted that assesses AP-beliefs and AP- practice, both for the global AP-approach and seven individual core-principles. AP-indicators consistently showed small/minimal correlations with better mental health, whereas mediation analyses revealed parallel, opposing pathways: a positive, direct association from AP to mental health (possibly explained via parental fulfilment) and adverse indirect pathways through intensive-parenting beliefs (challenge-/self-sacrifice and essentialist-beliefs).Conclusion. AP may be neither uniformly beneficial nor harmful. Rather, it can foster parental fulfilment which increases parental well-being, but also instil dysfunctional beliefs about sacrifice and unavoidable strain that increase parental-stress. To improve evidence-based personalized parenting-recommendations, longitudinal research is needed to disentangle when AP fosters positive parental identity and when it reinforces parental vulnerabilities.
- 16 September 2026
- Published by Center for Open Science in PsyArXiv
Open AccessHealthy attachment for the parents? An investigation on the association between attachment parenting and parental mental health.
Introduction. Attachment parenting (AP) became popular for it promotes responsive caregiving, but has been criticized for over-extensively translating attachment research into challenging parenting standards, mirroring detrimental intensive-parenting beliefs that adversely affect parental health. Meanwhile, evidence on AP remains limited and assessments of AP-beliefs and AP-practice are lacking. This study aims to develop and validate an AP-questionnaire and test the associations between AP and parental mental health and explore the mediating role of intensive-parenting beliefs.Methods. In a preregistered cross-sectional online-survey, 1664 parents (98.5% mothers) of children aged up to six years answered an extensive item-list assessing AP-beliefs and AP implementation into practice, an intensive parenting questionnaire, and mental health measures. Confirmatory factor analyses were used for AP-questionnaire validation and correlation- and mediation-analyses for exploring the association between AP and mental health.Results. A 26-item AP-questionnaire was extracted that assesses AP-beliefs and AP- practice, both for the global AP-approach and seven individual core-principles. AP-indicators consistently showed small/minimal correlations with better mental health, whereas mediation analyses revealed parallel, opposing pathways: a positive, direct association from AP to mental health (possibly explained via parental fulfilment) and adverse indirect pathways through intensive-parenting beliefs (challenge-/self-sacrifice and essentialist-beliefs).Conclusion. AP may be neither uniformly beneficial nor harmful. Rather, it can foster parental fulfilment which increases parental well-being, but also instil dysfunctional beliefs about sacrifice and unavoidable strain that increase parental-stress. To improve evidence-based personalized parenting-recommendations, longitudinal research is needed to disentangle when AP fosters positive parental identity and when it reinforces parental vulnerabilities.
- Open Access
- 29 May 2026
- Published by Leibniz Institute for Psychology (ZPID) in Clinical Psychology in Europe
- Vol. 8 (2)
Are you really paranoid, just because they are after you? Exploring the underlying sensitization processes of intersectional discrimination on everyday-life paranoia
Background: Elevated paranoia levels have been found in discriminated, minoritized groups. Social-cognitive models of paranoia posit that experiences of discrimination strengthen negative core beliefs, which in turn foster clinical paranoia. In contrast, the healthy cultural mistrust hypothesis proposes that elevated paranoia in minoritized groups reflects an adaptive response to increased exploitation/discrimination. To explore whether elevated paranoia in discriminated groups can be fully understood within clinical models, we tested whether a history of discrimination amplifies the association between everyday-life stressors and subsequent state-paranoia and whether this moderation effect remains when controlling for established cognitive risk factors for paranoia (i.e., dysfunctional beliefs). Method: A general population sample ( n = 108) answered a baseline self-report questionnaire of lifetime discrimination experiences (LDE) and core beliefs about oneself and others. Next, they reported state-paranoia, negative affect, and exclusion experiences in 2.5h intervals for seven days of ambulatory assessment. LDE and core-beliefs were tested as concurrent moderators for the associations between state-paranoia and putative triggers (negative-affect, exclusion) using multilevel regression. Results: More LDE amplified the association between paranoia, negative affect, and exclusion experiences. When controlling for the moderation effects of negative beliefs, effect sizes for the LDE moderation were lower but largely remained significant. Conclusions: Increased state-paranoia in the daily lives of people with discrimination experiences can be partially, but not fully, explained by clinical, cognitive risk factors. Consequently, healthy cultural mistrust can be considered an additional contributing factor to higher mistrust/paranoia in minoritized groups. Future in-depth research needs to disentangle mechanisms of emerging paranoia and adaptive mistrust in minoritized groups.
- 29 May 2026
- Published by Leibniz Institute for Psychology (ZPID) in Clinical Psychology in Europe
- Vol. 8 (2)
Open AccessAre you really paranoid, just because they are after you? Exploring the underlying sensitization processes of intersectional discrimination on everyday-life paranoia
Background: Elevated paranoia levels have been found in discriminated, minoritized groups. Social-cognitive models of paranoia posit that experiences of discrimination strengthen negative core beliefs, which in turn foster clinical paranoia. In contrast, the healthy cultural mistrust hypothesis proposes that elevated paranoia in minoritized groups reflects an adaptive response to increased exploitation/discrimination. To explore whether elevated paranoia in discriminated groups can be fully understood within clinical models, we tested whether a history of discrimination amplifies the association between everyday-life stressors and subsequent state-paranoia and whether this moderation effect remains when controlling for established cognitive risk factors for paranoia (i.e., dysfunctional beliefs). Method: A general population sample ( n = 108) answered a baseline self-report questionnaire of lifetime discrimination experiences (LDE) and core beliefs about oneself and others. Next, they reported state-paranoia, negative affect, and exclusion experiences in 2.5h intervals for seven days of ambulatory assessment. LDE and core-beliefs were tested as concurrent moderators for the associations between state-paranoia and putative triggers (negative-affect, exclusion) using multilevel regression. Results: More LDE amplified the association between paranoia, negative affect, and exclusion experiences. When controlling for the moderation effects of negative beliefs, effect sizes for the LDE moderation were lower but largely remained significant. Conclusions: Increased state-paranoia in the daily lives of people with discrimination experiences can be partially, but not fully, explained by clinical, cognitive risk factors. Consequently, healthy cultural mistrust can be considered an additional contributing factor to higher mistrust/paranoia in minoritized groups. Future in-depth research needs to disentangle mechanisms of emerging paranoia and adaptive mistrust in minoritized groups.
- Open Access
- 5 May 2026
- Published by Wiley in Suicide and Life-Threatening Behavior
- Vol. 56 (3) , e70104
The Impact of Family Factors on Adolescent Intensive Outpatient Psychotherapy Outcomes for Suicidal Thoughts, Behaviors, and Depression
Background: Although they constitute well‐established risk factors for suicidal thoughts and behaviors (STBs), few studies have investigated how family factors such as family cohesion and family conflict influence the trajectory of treatment response among depressed and suicidal adolescents. In this study, we examined the association between baseline family factors and response to an intensive outpatient program (IOP) for STBs in adolescents. Methods: Participants ( n = 637) either reported their satisfaction with family cohesion or family conflict levels at baseline and provided weekly self‐reports of depression and STBs throughout IOP treatment. We calculated multi‐level regression models to test for the interaction effects of treatment‐day × family‐factor‐levels and explored further demographic and clinical moderators. Results: Higher levels of family cohesion correlated with more reduction of STBs over time ( β = −0.11). Moreover, the effects of family cohesion and family conflict on treatment were moderated by patient age, ethnic minority status, and symptom severity. Discussion: STBs and depression improved with IOP, with family cohesion partially moderating treatment response. Differences due to family factors were more prominent in older and ethnically minoritized adolescents, and those with more severe symptoms. Future studies should elucidate how family factors co‐vary with changes in STBs during treatment to better understand these effects.
- 5 May 2026
- Published by Wiley in Suicide and Life-Threatening Behavior
- Vol. 56 (3) , e70104
Open AccessThe Impact of Family Factors on Adolescent Intensive Outpatient Psychotherapy Outcomes for Suicidal Thoughts, Behaviors, and Depression
Background: Although they constitute well‐established risk factors for suicidal thoughts and behaviors (STBs), few studies have investigated how family factors such as family cohesion and family conflict influence the trajectory of treatment response among depressed and suicidal adolescents. In this study, we examined the association between baseline family factors and response to an intensive outpatient program (IOP) for STBs in adolescents. Methods: Participants ( n = 637) either reported their satisfaction with family cohesion or family conflict levels at baseline and provided weekly self‐reports of depression and STBs throughout IOP treatment. We calculated multi‐level regression models to test for the interaction effects of treatment‐day × family‐factor‐levels and explored further demographic and clinical moderators. Results: Higher levels of family cohesion correlated with more reduction of STBs over time ( β = −0.11). Moreover, the effects of family cohesion and family conflict on treatment were moderated by patient age, ethnic minority status, and symptom severity. Discussion: STBs and depression improved with IOP, with family cohesion partially moderating treatment response. Differences due to family factors were more prominent in older and ethnically minoritized adolescents, and those with more severe symptoms. Future studies should elucidate how family factors co‐vary with changes in STBs during treatment to better understand these effects.
- Open Access
- 11 April 2025
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 55, e114
Predicting experiences of paranoia and auditory verbal hallucinations in daily life with ambulatory sensor data – A feasibility study
Background: Prediction models that can detect the onset of psychotic experiences are a key component of developing Just-In-Time Adaptive Interventions (JITAI). Building these models on passively collectable data could substantially reduce user burden. In this study, we developed prediction models to detect experiences of auditory verbal hallucinations (AVH) and paranoia using ambulatory sensor data and assessed their stability over 12 weeks. Methods: Fourteen individuals diagnosed with a schizophrenia-spectrum disorder participated in a 12-day Ecological Momentary Assessment (EMA) study. They wore ambulatory sensors measuring autonomic arousal (i.e., electrodermal activity, heart rate variability) and completed questionnaires assessing the intensity/distress of AVHs and paranoia once every hour. After 12 weeks, participants repeated the EMA for four days for a follow-up assessment. We calculated prediction models to detect AVHs, paranoia, and AVH-/paranoia-related distress using random forests within nested cross-validation. Calculated prediction models were applied to the follow-up data to assess the stability of prediction models. Results: Prediction models calculated with physiological data achieved high accuracy both for AVH (81%) and paranoia (69%–75%). Accuracy increased by providing models with baseline information about psychotic symptom levels (AVH: 86%; paranoia: 80%–85%). During the follow-up EMA accuracy dropped slightly throughout all models but remained high (73%–84%). Conclusions: Relying solely on physiological data to detect psychotic symptoms achieved substantial accuracy that remained sufficiently stable over 12 weeks. Experiences of AVHs can be predicted with higher accuracy and long-term stability than paranoia. The findings tentatively suggest that psychophysiology-based prediction models could be used to develop and enhance JITAIs for psychosis.
- 11 April 2025
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 55, e114
Open AccessPredicting experiences of paranoia and auditory verbal hallucinations in daily life with ambulatory sensor data – A feasibility study
Background: Prediction models that can detect the onset of psychotic experiences are a key component of developing Just-In-Time Adaptive Interventions (JITAI). Building these models on passively collectable data could substantially reduce user burden. In this study, we developed prediction models to detect experiences of auditory verbal hallucinations (AVH) and paranoia using ambulatory sensor data and assessed their stability over 12 weeks. Methods: Fourteen individuals diagnosed with a schizophrenia-spectrum disorder participated in a 12-day Ecological Momentary Assessment (EMA) study. They wore ambulatory sensors measuring autonomic arousal (i.e., electrodermal activity, heart rate variability) and completed questionnaires assessing the intensity/distress of AVHs and paranoia once every hour. After 12 weeks, participants repeated the EMA for four days for a follow-up assessment. We calculated prediction models to detect AVHs, paranoia, and AVH-/paranoia-related distress using random forests within nested cross-validation. Calculated prediction models were applied to the follow-up data to assess the stability of prediction models. Results: Prediction models calculated with physiological data achieved high accuracy both for AVH (81%) and paranoia (69%–75%). Accuracy increased by providing models with baseline information about psychotic symptom levels (AVH: 86%; paranoia: 80%–85%). During the follow-up EMA accuracy dropped slightly throughout all models but remained high (73%–84%). Conclusions: Relying solely on physiological data to detect psychotic symptoms achieved substantial accuracy that remained sufficiently stable over 12 weeks. Experiences of AVHs can be predicted with higher accuracy and long-term stability than paranoia. The findings tentatively suggest that psychophysiology-based prediction models could be used to develop and enhance JITAIs for psychosis.