Multimodal Prediction of Distressing and Persistent Psychotic-Like Experiences in Adolescence across Environment, Genetics, and Neuroimaging
- 1University of California San Diego
Abstract
Adolescent psychotic-like experiences, particularly those that are distressing and persistent (dpPLEs), may elevate risk for a range of psychiatric disorders, yet the relative predictive contributions of brain structure, genetics, and environment to dpPLEs remain unknown. Using four timepoints from 7,399 children in the Adolescent Brain Cognitive Development (ABCD) Study, we developed stacked ensemble machine-learning models integrating cortical morphometry and whole-brain microstructural properties, polygenic risk scores, family psychiatric history, and childhood environmental factors to predict adolescent dpPLEs. Environmental predictors alone, driven by area deprivation index, neighborhood drug-related crime, and cumulative early adversity, were the strongest predictive domain (ROC-AUC = 0.687 {+/-} 0.003), outperforming genetic (0.619 {+/-} 0.020) and all imaging models (0.588-0.650). Adding genetic risk to environmental features produced the best-performing model overall (ROC-AUC = 0.697 {+/-} 0.007), whereas adding neuroimaging yielded no further incremental gain (ROC-AUC=0.683-0.696). Beyond environmental exposures, family history of psychosis, cross-disorder polygenic risk, and both globally distributed and regionally specific imaging components contributed to prediction. Childhood environmental factors thus capture a core predictive signal preceding dpPLEs, while genetic and neuroimaging features added comparatively modest discriminative prediction value, but may offer complementary biological insights.
Related
- Multimodal neuroimaging signature of sleep problems captures preadolescent mental health trajectories
- Multimodal neuroimaging and electrophysiological signatures in amnestic mild cognitive impairment with different amyloid-β deposition status: correlations with cognitive function and diagnostic value
- Multimodal neuroimaging of cue-induced craving and cortical alterations in etomidate use disorder
- From brain networks to the patient-environment system in clinical neuroimaging
- Environmental predominance on brain structure and connectivity in language-related regions: a multimodal twin study in adolescence
BibTeX
@misc{lal2026multimodal,
title = {{Multimodal Prediction of Distressing and Persistent Psychotic-Like Experiences in Adolescence across Environment, Genetics, and Neuroimaging}},
author = {A. Lal and N. R. Karcher and J. Ahern and A. Rigby and S. Rai and K. Cadenhead and J. Linkersdorfer and D. J. Hagler and D. Pecheva and D. Barch and T. L. Jernigan and P. Parekh and A. M. Dale and C. Makowski},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.22.26365776},
url = {https://www.medrxiv.org/content/10.64898/2026.10.22.26365776v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Multimodal Prediction of Distressing and Persistent Psychotic-Like Experiences in Adolescence across Environment, Genetics, and Neuroimaging
AU - A. Lal
AU - N. R. Karcher
AU - J. Ahern
AU - A. Rigby
AU - S. Rai
AU - K. Cadenhead
AU - J. Linkersdorfer
AU - D. J. Hagler
AU - D. Pecheva
AU - D. Barch
AU - T. L. Jernigan
AU - P. Parekh
AU - A. M. Dale
AU - C. Makowski
PY - 2026
DA - 2026/10/22
PB - medRxiv
DO - 10.64898/2026.10.22.26365776
UR - https://www.medrxiv.org/content/10.64898/2026.10.22.26365776v1
AB - Adolescent psychotic-like experiences, particularly those that are distressing and persistent (dpPLEs), may elevate risk for a range of psychiatric disorders, yet the relative predictive contributions of brain structure, genetics, and environment to dpPLEs remain unknown. Using four timepoints from 7,399 children in the Adolescent Brain Cognitive Development (ABCD) Study, we developed stacked ensemble machine-learning models integrating cortical morphometry and whole-brain microstructural properties, polygenic risk scores, family psychiatric history, and childhood environmental factors to predict adolescent dpPLEs. Environmental predictors alone, driven by area deprivation index, neighborhood drug-related crime, and cumulative early adversity, were the strongest predictive domain (ROC-AUC = 0.687 {+/-} 0.003), outperforming genetic (0.619 {+/-} 0.020) and all imaging models (0.588-0.650). Adding genetic risk to environmental features produced the best-performing model overall (ROC-AUC = 0.697 {+/-} 0.007), whereas adding neuroimaging yielded no further incremental gain (ROC-AUC=0.683-0.696). Beyond environmental exposures, family history of psychosis, cross-disorder polygenic risk, and both globally distributed and regionally specific imaging components contributed to prediction. Childhood environmental factors thus capture a core predictive signal preceding dpPLEs, while genetic and neuroimaging features added comparatively modest discriminative prediction value, but may offer complementary biological insights.
ER -