Investigating cognitive neuroscience theories of human intelligence: A connectome-based predictive modeling approach.

This model (200 ROI, p < .01) is the only model of MD network connectivity that produces statistically significant predictions, suggesting that only coarse‐grained representations of highly significant MD functional edges are able to make generalizable predictions of g using CPM. 3.7 Process Overlap Theory (POT) For positive (right‐tailed) global connectivity profiles specified by POT (i.e., connections that represent functional overlap), we find evidence that whole‐brain functional edges do a relatively poor job at predicting g compared with other connectivity profiles, with the best‐performing model (Figure 7a ) generating predictions of r = .11 and p 1000 = .01 based on 200 vertices and 19,900 unique functional edges (200 ROI, p < .01).

also in 132,142 other papers