Using machine learning to determine a functional classifier of reward responsiveness and its association with adolescent psychiatric symptomatology
2024

Machine Learning Classifier for Reward Responsiveness in Adolescents

Sample size: 299 publication 10 minutes Evidence: moderate

Author Information

Author(s): Robert James Richard Blair, Johannah Bashford-Largo, Ahria Dominguez, Matthew Dobbertin, Karina S. Blair, Sahil Bajaj

Primary Institution: Child and Adolescent Mental Health Center, Copenhagen University Hospital

Hypothesis

Can a machine learning classifier distinguish reward responsiveness in adolescents and its association with psychiatric symptoms?

Conclusion

The study found that adolescents with conduct disorder and ADHD show significant disruptions in reward responsiveness compared to typically developing peers.

Supporting Evidence

  • Machine learning analysis identified a hyperplane with 80.77% accuracy.
  • Adolescents with externalizing diagnoses showed significantly less normative reward responses.
  • Deficient reward response was linked to higher symptoms of conduct disorder, major depressive disorder, and ADHD.

Takeaway

This study used a computer program to see how well teenagers respond to rewards compared to punishments, finding that some kids with mental health issues don't respond as well to rewards.

Methodology

The study used a Support Vector Machine (SVM) algorithm to analyze brain responses to rewards and punishments in adolescents.

Potential Biases

Potential biases due to the reliance on self-reported measures and the lack of structured diagnostic interviews.

Limitations

The study did not use structured diagnostic interviews, which may affect the reliability of the psychiatric diagnoses.

Participant Demographics

299 adolescents aged 10-19 years, with a mean age of 15.07 years, including 117 females.

Statistical Information

P-Value

<0.001

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1017/S003329172400240X

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication