Mathilde Antoniades, James Gilleen, Gemma Modinos

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


Neuroanatomical studies have provided some evidence for a neurobiological continuity between psychotic symptoms in patients with schizophrenia and subclinical psychotic-like experiences as identified via self-report questionnaires (or schizotypy) in otherwise healthy individuals. Despite a number of structural brain imaging studies published in subjects with schizotypy, these have often been limited by relatively small sample sizes. This study represents the first meta-analysis of structural MRI scans in schizotypy assessed with standardized methods at 17 centres worldwide.
The total sample included in the case-control meta-analysis involved 496 subjects with high schizotypy and 588 comparison subjects with low schizotypy as identified with validated self-report schizotypy questionnaires (i.e., the Community Assessment of Psychic Experiences, the Oxford-Liverpool Inventory of Feelings and Experiences, or the Schizotypal Personality Questionnaire). Left and right lateral ventricle, thalamus, caudate, putamen, pallidum, accumbens, hippocampus and amygdala volumes, as well as intracranial volumes (ICV), were obtained with FreeSurfer from high-resolution T1-weighted structural brain scans locally by each site. Analyses of individual subject data were also performed by the site that contributed the sample, using code created within the ENIGMA collaboration. Group differences for each ROI within each sample were examined using univariate linear regression analysis in R predicting the left, right and mean subcortical volumes with group (low or high schizotypy) controlling for age, sex, site and ICV. For each sample, Cohen’s d effect sizes for each ROI were computed based on the group contrast t-statistics. Random-effects meta-analyses of Cohen’s d effect sizes for each ROI were performed using R’s metafor package.
For the case-control meta-analysis, mean (range) age across samples was 29.2 (19.2–44.9) years for subjects with high schizotypy and 28.5 (19.4–42.5) years for subjects with low schizotypy. High and low schizotypy samples were on average 48.6% (25–100) and 49.0% (8–100) male. Compared to low schizotypy, people with high schizotypy had smaller left nucleus accumbens volumes (Cohen’s d = -0.13, p = 0.04). No other significant meta-analytical effects were identified.
Worldwide cooperative analyses of brain imaging data support a profile of subcortical abnormalities involving the ventral striatum in healthy individuals with subclinical psychotic-like experiences. Though the effect size is small, the result is consistent with findings in schizophrenia patients and suggesting that these effects are not secondary to potential influences of disease chronicity or antipsychotic medication. The lack of significant effects in other reported regions in schizophrenia research may indicate that further subcortical abnormalities may not be present or detectable in schizotypy samples and could reflect mechanisms of resilience in this population.
Original languageEnglish
Title of host publicationSchizophrenia Bulletin
PublisherOxford University Press
Publication statusPublished - 9 Apr 2019
EventSchizophrenia International Research Society - USA
Duration: 4 Apr 20197 Apr 2019


ConferenceSchizophrenia International Research Society

Cite this