Studying Healthy Psychosislike Experiences to Improve Illness Prediction
Philip R. Corlett, Sonia Bansal, James M. Gold
JAMA Psychiatry March 8, 2023 DOI: 10.1001/jamapsychiatry.2023.0059 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Commentary Peer reviewed |
|---|---|
| Citations | 1 |
| Key points | Argues that healthy people with unusual beliefs and experiences who are not ill can serve as "adversarial examples" that cause false alarms in predictive models of psychosis, and proposes that explicitly training such models with adversarial examples should focus them on features most relevant to casehood, benefiting clinical research, diagnosis, and treatment. |
Abstract
Importance: Distinguishing delusions and hallucinations from unusual beliefs and experiences has proven challenging. Observations: The advent of neural network and generative modeling approaches to big data offers a challenge and an opportunity; healthy individuals with unusual beliefs and experiences who are not ill may raise false alarms and serve as adversarial examples to such networks. Conclusions and Relevance: Explicitly training predictive models with adversarial examples should provide clearer focus on the features most relevant to casehood, which will empower clinical research and ultimately diagnosis and treatment.
Citations in the library
Cites 3
- Auditory verbal hallucinations and continuum models of psychosis: A systematic review of the healthy voice-hearer literature Clinical Psychology Review November 1, 2016
- Beyond Trauma: A Multiple Pathways Approach to Auditory Hallucinations in Clinical and Nonclinical Populations Schizophrenia Bulletin August 8, 2018
- Varieties of Voice-Hearing: Psychics and the Psychosis Continuum. Schizophrenia Bulletin January 2017