Fig 1.
Depiction of a previously published association network based on self-reported data from autistic adults.
This association network depicts the interrelations between ASD characteristics, well-being domains, and aspects of daily functioning as found in self-reported data presented in Deserno et al., 2017. The nodes represent the same variables as depicted in Fig 3, while the edges represent regularized partial correlations between those variables.
Fig 2.
Diagram of the three analytic steps of the current study.
Fig 3.
This PCR network is a merged visualisation of three subnetworks based on ratings by three clinical expert groups and depicts the cause-effect ratings between ASD characteristics (blue), well-being domains (brown), and aspects of daily functioning (rose) as perceived by the clinicians.
The nodes each represent a unique variable and the edges represent the averaged directed cause-effect rating. For reasons of visual comprehensibility, we only included edges with an edge weight higher than 0.5 and used curved arrows in case of bidirectional relations.
Fig 4.
Centrality indices for the PCR network depicted in Fig 3.
For the meaning of each node abbreviation see previous Figures and S1 Table.
Fig 5.
This scatterplot depicts the relationship between the values of the PCR network and the association network.
The values on the line parallel to the x-axis represent absent edges (value of 0) in the association network. The values in the lower left section represent edges that are negative in both graphs, while those values in the upper right section represent edges that are positive in both graphs.
Table 1.
Contingency table for concordant and disconcordant positive (1), negative (-1) and absent (0) links in the PCR and the empirical network.
Fig 6.
In the first row the three subnetworks are estimated with constraints based on the network structure found in the clinician network (PCR-constrained).
In the second row, the graphical lasso networks are depicted individually, for comparison (Graphical lasso).