Schizophrenia is a psychiatric disorder with heterogeneous clinical manifestations and complex aetiology. Notably, the triple-network model proposes an interesting framework for investigating abnormal neurocircuit activity at rest in schizophrenia. The present study on 30 chronic schizophrenia individuals and 30 controls aimed to explore the differences in EEG resting state effective connectivity within a triple-network model using source-localization-based Directed Transfer Function. Our findings revealed multiband effective connectivity disturbances within default mode (DMN), central executive (CEN), and salience (SN) networks in schizophrenia. The most significant difference was manifested in a global DMN hyperconnectivity, accompanied by low-band hyperconnectivity and high-band hypoconnectivity in CEN, along with the aberrant information flows in SN. In conclusion, our study presents novel insights into schizophrenia neuropathology, with a particular emphasis on the reversed directionality in information flows between hubs of SN, DMN, and CEN. This may be suggested as a promising biomarker of schizophrenia.