Neurocognitive, behavioural and neurodevelopmental correlates of schizotypy clusters in adolescents from the general population

Introduction: Studies on the neurocognitive correlates of schizotypy dimensions have found inconsistent results. This might stem from the fact that correlational methods, in contrast to cluster analysis, do not account for the possibility that a subject presents high scores on more than one dimension simultaneously. We aimed to establish clusters of normal adolescents based on schizotypy dimensions and compare them on neurocognitive, behavioural, and neurodevelopmental markers. Methods: Two hundred seventy normal adolescents from the general population (mean age 13.4, SD=0.72) attending obligatory education were evaluated. Results: A K-means iterative cluster analysis was performed with the Perceptual Aberration, Revised Social Anhedonia and Physical Anhedonia Scales. A forced four-cluster model yielded the following clusters: 'negative schizotypy', 'high or mixed schizotypy', 'positive schizotypy', and 'normal scorers'. Comparisons with ANOVAs showed that 'high schizotypes' performed poorly on neurocognition (Wechsler Intelligence Scales for Children-Revised (WISC-R) and Verbal Fluency (FAS)) and obtained the highest teacher ratings (TRF) of behavioural problems. 'Negative schizotypes' had the worst WCST results and more dermatoglyphic abnormalities. Both clusters had more neurological soft signs than 'normal scorers' and 'positive schizotypes'. Conclusions: Our results with community adolescents found the same cluster structure than the previous cluster analytic studies conducted in adult college subjects. Furthermore, we showed differences among them on neurocognitive and malneurodevelopment markers consistent with the adult literature on schizotypy. © 2002 Elsevier Science B.V. All rights reserved.