5:45 PM SALIENCE MATTERS: BRAIN POTENTIALS DISTINGUISH PREMORBID ATTENTION PROBLEMS AMONG CHILDREN AT-RISK FOR SCHIZOPHRENIA

Background: A robust marker of brain dysfunction in schizophrenia (Cohen’s d=0.89) is reduction in amplitude of the P3 (or P300) event-related potential (ERP) that indexes attention and working memory processes. It is elicited typically using a two-tone auditory oddball paradigm that presents infrequent task-relevant target stimuli (which require a behavioural response) within a train of frequent task-irrelevant non-target (or standard) stimuli. The P3 amplitude reduction to target stimuli is present at the chronic and first-episode phases of schizophrenia, and may predict transition to psychosis among at-risk adolescents/young adults. We sought to characterise potential premorbid brain function abnormalities among at-risk children aged 9-12 years in detail, using an auditory novelty oddball task variant that dissociates an earlier, automatic frontocentral P3a subcomponent elicited by infrequent task-irrelevant salient novel stimuli from a later parietal P3b sub-component elicited by the infrequent task-relevant target stimuli. Methods: We examined brain function in two at-risk groups, including a group with family history of schizophrenia (FHx; n=20) and a group of children presenting a triad of replicated antecedents of schizophrenia (ASz; n=20), namely (i) psychotic-like experiences; (ii) a social, emotional, or behavioural problem; and (iii) a speech-and/or-motor developmental delay or abnormality. At-risk groups were compared with a group of typicallydeveloping (TD; n=28) peers who had no family history or antecedents of schizophrenia. Groups did not differ significantly on age, sex, ethnicity, socio-economic status, or handedness. Three group (FHx, ASz, TD) x three stimulus (targets, novels, non-targets) x five site (Fz, FCz, Cz, CPz, Pz) repeated-measures ANOVAs on peak P3 amplitude and latency data were conducted, and were repeated as ANCOVAs to control for poorer intellectual function (IQ) among the at-risk groups relative to TD children. Results: A significant group-by-condition interaction indicated reduced novelty P3a amplitude, but not target P3b or non-target P3 amplitudes, in both at-risk groups relative to TD children. This result remained after correcting for IQ differences between groups. No latency differences were observed. Discussion: At-risk children aged 9-12 years, both those with family history and those presenting multiple childhood antecedents, display disturbances in frontal mechanisms supporting involuntary attention orienting to salient stimuli, while more conscious parietal mechanisms supporting stimulus evaluation and context updating are, as yet, unaffected in these children. The latter may emerge more proximally to transition to psychosis; 48month follow-up assessments are underway to ascertain the evolution of these components as participants mature through adolescence.