Characteristics Of Exploratory Eye-Movement In Subjects At Clinical-High Risk For Psychosis

Objective To explore the eye-movement feature in subjects at clinical high risk(CHR)for psychosis.To further investigate which eye-movement characteristics can best discriminate CHR subjects from healthy controls and which eye-movement characteristics can predict their outcome of 1-year follow-up.Methods 100 CHR subjects and 80 healthy controls were recruited and assessed with the Structured Interview for Prodromal Syndromes(SIPS)/ Scale of Prodromal Symptoms(SOPS),controls were matched with CHR subjects in gender,age and education level.CHR subjects finished the clinical follow-up of 1 year and then they were divided into two groups: converters(to psychosis)and non-converters.All of the subjects finished the fixation stability,smooth pursuit,prosaccade/ antisaccade tasks and Matrics Consensus Cognitive Battery test.Analysis included the comparision of the eye-movement performance among the 3 groups,the predictive validity of the 3 groups(by using the Logistics regression and receiver operating characteristic analysis)and the relationship between eye-movement performance and cognitive function.Results A total of 90 CHR subjects completed the clinical follow-up of 1 year and16 of them transitioned to a psychotic disorder over the course of follow-up.CHR subjects showed lower fixation intensity,lower accuracy of pursuit and longer antisaccade latency compared with healthy controls,which displayed statistical differences.And there were significant correlations between the above eye-movement characteristics and cognitive function in the CHR subjects.Logistic regression was performed to discriminate CHR subjects from controls.Using the predicted probabilities as the ROC curve predictor variable,the Area Under Curve(AUC)was 0.772.On the other hand,converters showed lower fixation intensity,lower fixation continuity and lower velocity of pursuit initial phase compared with non-converters,which displayed statistical differences.Logistic regression was performed to discriminate converters from non-converters.Using the predicted Probabilities as the ROC curve predictor variable,the AUC was 0.723.Conclusions CHR subjects show deficits in eye-movement.The eye-movement characteristics which reflect the cognitive function can discriminate CHR subjects from controls,and the eye-movement characteristics which reflect the sensorimotor transformational function can predict their clinical outcome.