Distress, impairment and the network of psychotic like experiences

It has been proposed that subclinical psychotic-like experiences (PLEs) may causally impact each other over time and engage with one another in patterns of mutual reinforcement and feedback. This subclinical network of... [ view full abstract ]

It has been proposed that subclinical psychotic-like experiences (PLEs) may causally impact each other over time and engage with one another in patterns of mutual reinforcement and feedback. This subclinical network of experiences in turn may facilitate the onset of psychotic disorder. PLEs however are not inherently distressing, nor do they inevitably lead to impairment. The question arises therefore whether non-distressing PLEs, distressing PLEs, or both, meaningfully inform an extended psychosis phenotype. The current study first aimed to exploit valuable ordinal data that captured the absence, occurrence and associated impairment of PLEs in the general population in order to: (i) construct a general population based severity network of PLEs; and (ii) to determine which PLEs may be ‘central’ to the psychosis phenotype at subclinical levels. The study then aimed to partition the available ordinal data into two sets of binary data to test whether an occurrence network comprised of PLE data denoting absence (coded 0) and occurrence/impairment (coded 1) mirrored an impairment network comprised of binary PLE data denoting absence/occurrence (coded 0) and impairment (coded 1). Networks were constructed using state-of-the-art regularized pairwise Markov Random Fields (PMRF) for both ordinal and binary data. The ordinal PMRF revealed a strongly connected network of 16 PLEs which differed substantially in their centrality values. Nodes denoting paranoia were among the most central in the network. The binary PMRF impairment network structure was similar to the occurrence network, however the impairment network suggested stronger interconnectivity between PLEs.