Functional Neuroimaging Predictors of Self-Reported Psychotic Symptoms in Adolescents

Josiane Bourque, Philip A Spechler, Stéphane Potvin, Robert Whelan, Tobias Banaschewski, Arun L W Bokde, Uli Bromberg, Christian Büchel, Erin Burke Quinlan, Sylvane Desrivières, Herta Flor, Vincent Frouin, Penny Gowland, Andreas Heinz, Bernd Ittermann, Jean-Luc Martinot, Marie-Laure Paillère-Martinot, Sarah C McEwen, Frauke Nees, Dimitri Papadopoulos Orfanos, Tomáš Paus, Luise Poustka, Michael N Smolka, Nora C Vetter, Henrik Walter, Gunter Schumann, Hugh Garavan, Patricia J Conrod

This study investigated the neural correlates of psychotic-like experiences in youths during tasks involving inhibitory control, reward anticipation, and emotion processing. A secondary aim was to test whether these neurofunctional correlates of risk were predictive of psychotic symptoms 2 years later. Functional imaging responses to three paradigms-the stop-signal, monetary incentive delay, and faces tasks-were collected in youths at age 14, as part of the IMAGEN study. At baseline, youths from London and Dublin sites were assessed on psychotic-like experiences, and those reporting significant experiences were compared with matched control subjects. Significant brain activity differences between the groups were used to predict, with cross-validation, the presence of psychotic symptoms in the context of mood fluctuation at age 16, assessed in the full sample. These prediction analyses were conducted with the London-Dublin subsample (N=246) and the full sample (N=1,196). Relative to control subjects, youths reporting psychotic-like experiences showed increased hippocampus/amygdala activity during processing of neutral faces and reduced dorsolateral prefrontal activity during failed inhibition. The most prominent regional difference for classifying 16-year-olds with mood fluctuation and psychotic symptoms relative to the control groups (those with mood fluctuations but no psychotic symptoms and those with no mood symptoms) was hyperactivation of the hippocampus/amygdala, when controlling for baseline psychotic-like experiences and cannabis use. The results stress the importance of the limbic network's increased response to neutral facial stimuli as a marker of the extended psychosis phenotype. These findings might help to guide early intervention strategies for at-risk youths.