An inefficient integration between bottom-up visual input and higher-order visual processing regions is implicated in the manifestation of visual hallucinations (VH) in Parkinson's disease (PD). Using graph theory, the current study aimed to investigate white matter contributions to this perceptual imbalance hypothesis. Twenty-nine PD patients reported their hallucinatory behaviour on a questionnaire and performed a behavioural test that has been shown to elicit misperceptions. A composite score derived from these measures was used as a proxy for hallucinations severity and was correlated to connectivity strength of the network using the Network Based Statistic approach. The results showed that the severity of VH was associated with reduced connectivity within a large sub-network. This network included the majority of the diverse club and showed overall greater between- and within-module scores, compared to nodes not associated with hallucination severity. Furthermore, a reduction in between-module connectivity in the lateral occipital cortex, insula and pars orbitalis, as well as decreased within-module connectivity in the prefrontal, somatosensory and primary visual cortices were associated with VH severity. In contrast, the severity of VH was associated with an increase in between- and within-module connectivity in the orbitofrontal and temporal cortex, as well as regions comprising the dorsal attentional and default mode network. These results suggest that the severity of VHs is associated with marked alterations in structural network topology, highlighted by a reduction in connectivity strength across a large sub-network, as well as changes in participation across top-down visual processing centres, visual and attentional networks. Therefore, impaired integration across the perceptual hierarchy may result in the inefficient transfer of information that gives rise to VHs in PD.

Changes in structural network topology correlate with severity of hallucinatory behaviour in Parkinson's disease