Age of gray matters: Neuroprediction of recidivism

Kent A. Kiehl, Nathaniel E. Anderson, Eyal Aharoni, J. Michael Maurer, Keith A. Harenski, Vikram Rao, Eric D. Claus, Carla Harenski, Mike Koenigs, Jean Decety, David Kosson, Tor D. Wager, Vince Daniel Calhoun, Vaughn R. Steele

Research output: Contribution to journalArticle

Abstract

Age is one of the best predictors of antisocial behavior. Risk models of recidivism often combine chronological age with demographic, social and psychological features to aid in judicial decision-making. Here we use independent component analyses (ICA) and machine learning techniques to demonstrate the utility of using brain-based measures of cerebral aging to predict recidivism. First, we developed a brain-age model that predicts chronological age based on structural MRI data from incarcerated males (n = 1332). We then test the model's ability to predict recidivism in a new sample of offenders with longitudinal outcome data (n = 93). Consistent with hypotheses, inclusion of brain-age measures of the inferior frontal cortex and anterior-medial temporal lobes (i.e., amygdala) improved prediction models when compared with models using chronological age; and models that combined psychological, behavioral, and neuroimaging measures provided the most robust prediction of recidivism. These results verify the utility of brain measures in predicting future behavior, and suggest that brain-based data may more precisely account for important variation when compared with traditional proxy measures such as chronological age. This work also identifies new brain systems that contribute to recidivism which has clinical implications for treatment development.

Original languageEnglish (US)
Pages (from-to)813-823
Number of pages11
JournalNeuroImage: Clinical
Volume19
DOIs
StatePublished - Jan 1 2018
Externally publishedYes

Fingerprint

Brain
Psychological Models
Aptitude
Frontal Lobe
Proxy
Temporal Lobe
Amygdala
Neuroimaging
Gray Matter
Decision Making
Demography
Psychology
Therapeutics

Keywords

  • Age
  • Antisocial
  • MRI
  • Neuroprediction
  • Recidivism

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Neurology
  • Clinical Neurology
  • Cognitive Neuroscience

Cite this

Kiehl, K. A., Anderson, N. E., Aharoni, E., Maurer, J. M., Harenski, K. A., Rao, V., ... Steele, V. R. (2018). Age of gray matters: Neuroprediction of recidivism. NeuroImage: Clinical, 19, 813-823. https://doi.org/10.1016/j.nicl.2018.05.036

Age of gray matters : Neuroprediction of recidivism. / Kiehl, Kent A.; Anderson, Nathaniel E.; Aharoni, Eyal; Maurer, J. Michael; Harenski, Keith A.; Rao, Vikram; Claus, Eric D.; Harenski, Carla; Koenigs, Mike; Decety, Jean; Kosson, David; Wager, Tor D.; Calhoun, Vince Daniel; Steele, Vaughn R.

In: NeuroImage: Clinical, Vol. 19, 01.01.2018, p. 813-823.

Research output: Contribution to journalArticle

Kiehl, KA, Anderson, NE, Aharoni, E, Maurer, JM, Harenski, KA, Rao, V, Claus, ED, Harenski, C, Koenigs, M, Decety, J, Kosson, D, Wager, TD, Calhoun, VD & Steele, VR 2018, 'Age of gray matters: Neuroprediction of recidivism', NeuroImage: Clinical, vol. 19, pp. 813-823. https://doi.org/10.1016/j.nicl.2018.05.036
Kiehl KA, Anderson NE, Aharoni E, Maurer JM, Harenski KA, Rao V et al. Age of gray matters: Neuroprediction of recidivism. NeuroImage: Clinical. 2018 Jan 1;19:813-823. https://doi.org/10.1016/j.nicl.2018.05.036
Kiehl, Kent A. ; Anderson, Nathaniel E. ; Aharoni, Eyal ; Maurer, J. Michael ; Harenski, Keith A. ; Rao, Vikram ; Claus, Eric D. ; Harenski, Carla ; Koenigs, Mike ; Decety, Jean ; Kosson, David ; Wager, Tor D. ; Calhoun, Vince Daniel ; Steele, Vaughn R. / Age of gray matters : Neuroprediction of recidivism. In: NeuroImage: Clinical. 2018 ; Vol. 19. pp. 813-823.
@article{4652e9f891874ec0a43df54e00e28442,
title = "Age of gray matters: Neuroprediction of recidivism",
abstract = "Age is one of the best predictors of antisocial behavior. Risk models of recidivism often combine chronological age with demographic, social and psychological features to aid in judicial decision-making. Here we use independent component analyses (ICA) and machine learning techniques to demonstrate the utility of using brain-based measures of cerebral aging to predict recidivism. First, we developed a brain-age model that predicts chronological age based on structural MRI data from incarcerated males (n = 1332). We then test the model's ability to predict recidivism in a new sample of offenders with longitudinal outcome data (n = 93). Consistent with hypotheses, inclusion of brain-age measures of the inferior frontal cortex and anterior-medial temporal lobes (i.e., amygdala) improved prediction models when compared with models using chronological age; and models that combined psychological, behavioral, and neuroimaging measures provided the most robust prediction of recidivism. These results verify the utility of brain measures in predicting future behavior, and suggest that brain-based data may more precisely account for important variation when compared with traditional proxy measures such as chronological age. This work also identifies new brain systems that contribute to recidivism which has clinical implications for treatment development.",
keywords = "Age, Antisocial, MRI, Neuroprediction, Recidivism",
author = "Kiehl, {Kent A.} and Anderson, {Nathaniel E.} and Eyal Aharoni and Maurer, {J. Michael} and Harenski, {Keith A.} and Vikram Rao and Claus, {Eric D.} and Carla Harenski and Mike Koenigs and Jean Decety and David Kosson and Wager, {Tor D.} and Calhoun, {Vince Daniel} and Steele, {Vaughn R.}",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.nicl.2018.05.036",
language = "English (US)",
volume = "19",
pages = "813--823",
journal = "NeuroImage: Clinical",
issn = "2213-1582",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Age of gray matters

T2 - Neuroprediction of recidivism

AU - Kiehl, Kent A.

AU - Anderson, Nathaniel E.

AU - Aharoni, Eyal

AU - Maurer, J. Michael

AU - Harenski, Keith A.

AU - Rao, Vikram

AU - Claus, Eric D.

AU - Harenski, Carla

AU - Koenigs, Mike

AU - Decety, Jean

AU - Kosson, David

AU - Wager, Tor D.

AU - Calhoun, Vince Daniel

AU - Steele, Vaughn R.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Age is one of the best predictors of antisocial behavior. Risk models of recidivism often combine chronological age with demographic, social and psychological features to aid in judicial decision-making. Here we use independent component analyses (ICA) and machine learning techniques to demonstrate the utility of using brain-based measures of cerebral aging to predict recidivism. First, we developed a brain-age model that predicts chronological age based on structural MRI data from incarcerated males (n = 1332). We then test the model's ability to predict recidivism in a new sample of offenders with longitudinal outcome data (n = 93). Consistent with hypotheses, inclusion of brain-age measures of the inferior frontal cortex and anterior-medial temporal lobes (i.e., amygdala) improved prediction models when compared with models using chronological age; and models that combined psychological, behavioral, and neuroimaging measures provided the most robust prediction of recidivism. These results verify the utility of brain measures in predicting future behavior, and suggest that brain-based data may more precisely account for important variation when compared with traditional proxy measures such as chronological age. This work also identifies new brain systems that contribute to recidivism which has clinical implications for treatment development.

AB - Age is one of the best predictors of antisocial behavior. Risk models of recidivism often combine chronological age with demographic, social and psychological features to aid in judicial decision-making. Here we use independent component analyses (ICA) and machine learning techniques to demonstrate the utility of using brain-based measures of cerebral aging to predict recidivism. First, we developed a brain-age model that predicts chronological age based on structural MRI data from incarcerated males (n = 1332). We then test the model's ability to predict recidivism in a new sample of offenders with longitudinal outcome data (n = 93). Consistent with hypotheses, inclusion of brain-age measures of the inferior frontal cortex and anterior-medial temporal lobes (i.e., amygdala) improved prediction models when compared with models using chronological age; and models that combined psychological, behavioral, and neuroimaging measures provided the most robust prediction of recidivism. These results verify the utility of brain measures in predicting future behavior, and suggest that brain-based data may more precisely account for important variation when compared with traditional proxy measures such as chronological age. This work also identifies new brain systems that contribute to recidivism which has clinical implications for treatment development.

KW - Age

KW - Antisocial

KW - MRI

KW - Neuroprediction

KW - Recidivism

UR - http://www.scopus.com/inward/record.url?scp=85048265178&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048265178&partnerID=8YFLogxK

U2 - 10.1016/j.nicl.2018.05.036

DO - 10.1016/j.nicl.2018.05.036

M3 - Article

C2 - 30013925

AN - SCOPUS:85048265178

VL - 19

SP - 813

EP - 823

JO - NeuroImage: Clinical

JF - NeuroImage: Clinical

SN - 2213-1582

ER -