Effective connectivity and gamma oscillations in a group at risk of psychosis

22q11.2 Deletion Syndrome (22q11.2DS) has been shown to be associated with a markedly increased risk for schizophrenia. Therefore, 22q11.2DS is a homogeneous genetic liability model which enables studies intending to identify functional abnormalities that may precede disease onset of schizophrenia. Being able to define these functional abnormalities could potentially assist in the search of biomarkers for schizophrenia.

These are highly desired since early notification as well as early treatment have shown positive effects on everyday functioning in schizophrenia patients. This thesis aimed at looking for functional abnormalities, known to be found in schizophrenia, in a cohort of 22q11.2 deletion carriers. The search for functional abnormalities in the 22q11.2 deletion syndrome cohort, were carried out measuring EEG while subjects engaged in a roving mismatch negativity (MMN) paradigm as well as an auditory steady state paradigm.

Both of these paradigms are known to involve processes that are impaired in schizophrenia. This thesis ties together the three main contributions which are divided into three studies. In the first study, the responses to a roving MMN paradigm were assessed in 22q11.2 deletion carriers and healthy controls.

Both conventional analysis of the MMN responses as well as a more sophisticated approach by means of Dynamic Causal Modelling (DCM) were carried out. DCM is a technique to extract effective connectivity between pre-specified bran areas. With this technique we investigated the underlying network of change detection in the two groups.

While we found no indication of a reduced MMN response at the scalp level in the 22q11.2 deletion carriers, the underlying network of change detection differed as compared to healthy controls. Second study extended study number one by employing a parametric DCM to study the underlying network of repetition suppression in 22q11.2 deletion carriers and healthy controls.

While results for the control group indicated a processing in accordance with the predictive coding theory, this was not the case for the 22q11.2 deletion carriers. Finally, in the third study the ability to generate 40 Hz cortical oscillations were assessed in 22q11.2 deletion carriers as well as healthy controls using an auditory steady state paradigm.

Here, it was found that both phase and power of the 40 Hz oscillatory activity were reduced in 22q11.2 deletion carriers as compared to healthy controls. In the three studies, results both similar and dissimilar to what is observed in the schizophrenia literature were found. The studies contribute in understanding the underlying pathology of 22q11.2 deletion syndrome and if results are confirmed by longitudinal follow up studies, the results might contribute in the search of biomarkers for schizophrenia.