Do Cognitive Subtypes Exist in People at Clinical High Risk for Psychosis? Results From the EU-GEI Study

Area of Science:

• Neuroscience
• Psychiatry
• Cognitive Science

Background:

• Cognitive deficits are linked to socio-occupational functioning in individuals at Clinical High Risk for Psychosis (CHR-P).
• Previous research suggests potential cognitive subtypes within CHR-P populations.

Purpose of the Study:

• To investigate if clustering cognitive data from CHR-P individuals reveals clinically meaningful subtypes.
• To validate any identified cognitive subtypes by examining functioning, psychosis symptoms, and neuroimaging data.

Main Methods:

• Employed k-means and density-based spatial clustering on cognitive data from 291 CHR-P subjects.
• Validated clusters by comparing functioning, psychosis symptoms, transition outcomes, and grey matter volume.
• Utilized network analysis to explore relationships between cognitive and clinical measures.

Main Results:

• No inherent cluster structure was identified in the cognitive data.
• K-means clustering yielded 'spared' and 'impaired' groups, but these lacked significant associations with functioning, symptoms, or grey matter volume.
• Network analysis revealed separate subnetworks for cognition and clinical/functioning measures.

Conclusions:

• Evidence for distinct cognitive subtypes in this CHR-P sample was not found.
• Caution is advised when inferring cognitive subtypes from unsupervised learning in CHR-P.
• Future research should explore a broader range of cognitive domains for potential subtypes in CHR-P populations.