Related to clinical symptoms a decreased microstate class

Related to clinical symptoms, a decreased microstate class D duration was observed while schizophrenia patients were self-reporting auditory hallucinations (Kindler et al., 2011), and another study reports negative correlations of microstate class D duration with positive, paranoid symptoms (Koenig et al., 1999). These findings led to the proposal that microstate class D might not reflect specific symptoms per se but that it rather subserves a protective function when present for a sufficient duration as precocious termination of this state was observed while patients were experiencing positive symptoms (Kindler et al., 2011).
We also found increased class C microstate occurrence, mean duration, time anti fungal coverage and GEV in patients with schizophrenia, which has also been reported in some of the previous papers on microstates in schizophrenia (Kikuchi et al., 2007; Lehmann et al., 2005; Nishida et al., 2013). The increased class C duration in both schizophrenia patients and 22q11DS adolescents with respect to age-matched controls could explain the increased risk of 22q11DS adolescents to develop schizophrenia. The class C microstate was related to the salience RSN, a network comprised of core activations in the bilateral insula and the anterior cingulate cortex (Britz et al., 2010). Auditory task activations of the anterior cingulate followed by a downstream reduction of auditory activity in 22q11DS seem to suggest a deficit of salience processing in this population (Rihs et al., 2013). Functionally, the salience RSN was shown to be involved in the detection and orientation towards relevant internal and external stimuli (Menon, 2011). Numerous functional and structural studies reported salience dysfunctions in schizophrenia and a recent model is proposing the important role of the salience RSN in recruiting networks crucial for optimal information processing (Menon, 2011; Palaniyappan and Liddle, 2012). Hence, the salience network dysfunction would be responsible for the emergence of psychotic symptoms, such as active auditory hallucinations, which were associated with increased activation in nodes of the SN (Jardri et al., 2011; Palaniyappan and Liddle, 2012). The results of our study are in line with this model since we observe deviances in resting activity of two microstate anti fungal associated with the salience and the CEN RSNs both in the schizophrenia as well in the ultra-high-risk 22q11DS group.
Our fractal analysis investigated the long-range temporal dependencies of EEG microstates. The index of fractality, the Hurst exponent, is not linked to a specific microstate class, but depends on the interactions between all four of them. A previous study on healthy controls showed that the Hurst exponent does not significantly differ across healthy individuals (Van de Ville et al., 2010). Here, we replicated the same result for our controls, but found significant differences of the Hurst exponents within the patient groups. These differences suggest that disturbances in the temporal dynamics of microstates are reflected by significant differences of Hurst exponent in both the 22q11DS and schizophrenia patients. Importantly, this result suggests that the efficiency of the global dynamics of EEG microstates was preserved and could point to a compensatory mechanism between the class C and D microstates. Thus, the deviant temporal dynamics of EEG microstates in clinical populations could explain differences in information processing strategies, while a disruption of the long range dependency of the resting state dynamics would only occur in more pronounced alterations of the functional activity such as during unconscious states.
Moreover, the results of the syntax analysis suggest that in schizophrenia patients the increased class C duration is probably explained by the significant transitions from all other classes towards the class C microstate, while in healthy adults there is a significant transition pattern from all other classes towards class D (see Fig. 2.3 and 2.4). Another study reported more transitions in individuals with schizophrenia from class D to class C and in reverse for the adult controls (Nishida et al., 2013). Interestingly, in the 22q11DS, we also find increased transition probabilities from class B to class C compared to adolescent controls. We could speculate that while the salience RSN is optimally recruiting the target networks in healthy adults, in schizophrenia a disconnection could be expressed by recurrent transitions towards the salience RSN in an effort to recruit the required networks. The fact that the same dynamics are observed in our ultra-high-risk population with 22q11DS could be indicative of a key mechanism in temporal network dynamics shared by 22q11DS and schizophrenia.