We aimed to explore dynamic functional network connectivity (dFNC) differences and to elucidate their association with neurocognitive changes in cirrhotic patients. Sliding window correlation approach was employed to calculate dFNC. FNC states were determined by k-means clustering method, and then state analysis was conducted. The patients showed decreased FNC in State 2 and state 3. The patients spent significantly longer time in State 4 that was with weakest FNC. We observed a significant correlation between Psychometric Hepatic Encephalopathy Score and fraction time/mean dwell time in State 4. Aberrant dFNC may be the underlying mechanism of neurocognitive impairments in cirrhosis.