Lag-structure in fMRI across Three Psychiatric Groups: State-dependency and Clinical-behavioral Correlates

Le résumé fourni par la source

fMRI measures beyond zero-lag functional connectivity could serve as useful tools for understanding the distinct spatio-temporal dynamics characterizing psychiatric conditions. Therefore, the primary objective was to investigate whether and how state-dependence influences lag-structure in healthy controls (n = 95). Moreover, the study aimed to explore clinical-behavioral correlates of state-dependent lag-structure in three groups of psychiatric patients (35 ADHD, 38 Bipolar Disorder and 23 Schizophrenia patients, diagnosed according to DSM-IV-TR). Lag-structure was computed from cross-correlation coefficients in resting-state and stop-signal scans. Between and within-group differences were compared through non-parametric tests. Correlations with clinical-behavioral parameters were evaluated using linear regressions (Brief Psychiatric Rating Scale - BPRS, task reaction time). Compared to healthy controls, lag-structure within default-mode, executive-control and salience networks was generally increased in ADHD (min Z-score - 3.983), generally decreased in Schizophrenia (min Z-score - 3.716) and mixed increased/decreased in Bipolar patients (min Z-score - 3.912, max 4.739). Widespread state-dependent reductions of lag-structure were observed across all groups from rest to task (max Q-statistics: healthy controls 58; ADHD 22; Bipolar 26; Schizophrenia 17). Correlations with clinical-behavioral features (BPRS, reaction time) were positive in the executive-control network and negative in the bilateral thalamus for ADHD; negative in the cerebellum for Schizophrenia; positive in the right temporal gyri, amygdala, hippocampus, cerebellum for Bipolar Disorder (p = 0.05). In summary, according to these preliminary results, differences in lag-structure in comparison to healthy controls may be described as progressively increased in magnitude from ADHD to Bipolar Disorder and Schizophrenia, with specific clinical and behavioral correlates according to each diagnostic group.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.