Psychometric intelligence is closely related to working memory capacity. Here we aim to determine the associations of neural activation patterns during the N-back working memory paradigm with psychometric intelligence and working memory performance. We solved the statistical problems of previous studies using (1) a large cohort of 1235 young adults and (2) robust voxel-by-voxel permutation-based statistics at the whole-brain level. Many of the significant correlations were weak, and our findings were not consistent with those of previous studies. We observed that many of the significant correlations involved brain areas in the periphery or boundaries between the task-positive network (TPN) and task-negative network (TNN), suggesting that the expansion of the TPN or TNN is associated with greater cognitive ability. Lower activity in TPN and less task-induced deactivation (TID) in TNN were associated with greater cognitive ability. These findings indicate that subjects with greater cognitive ability have a lower brain response to task demand, consistent with the notion that TID in TNN reflects cognitive demand but partly inconsistent with the prevailing neural efficiency theory. One exception was the pre-supplementary motor area, which plays a key role in cognitive control and sequential processing. In this area, intelligent subjects demonstrated greater activity related to working memory, suggesting that the pre-supplementary motor area plays a unique role in the execution of working memory tasks in intelligent subjects.
Keywords: Brain activity; General intelligence; N-back task; Non-verbal reasoning; Working memory.