Functional Connectivity and Fluid Intelligence - A Multiverse Analysis

The current project aims to investigate previously reported associations between graph-theoretical measures (see sections 3.2 and 4.1) and fluid intelligence using data from the CoScience Personality Project. We will evaluate the robustness of potential findings across different connectivity measures and thresholding methods within a multiverse framework. We focus on indirect measures of functional connectivity in sensor space that are estimated in the frequency spectrum and are less sensitive to volume conduction. Following network construction, individual fluid intelligence scores will be regressed on graph-theoretical measures using multi-level models to account for potential site effects in the multi-site data set. Statistical significance will be assessed using nonparametric permutation testing. The robustness of findings across multiverse specifications will be evaluated by examining the consistency in size and direction of effect sizes.