The human ability to adaptively implement a wide variety of tasks is thought to emerge from the dynamic transformation of cognitive information. This study hypothesized that these transformations are implemented via conjunctive activations in "conjunction hubs"—brain regions that selectively integrate sensory, cognitive, and motor activations. Using fMRI data during a cognitive control task and advances in functional connectivity mapping, a task-performing neural network model was constructed. Simulations over this empirically-estimated functional connectivity model verified the importance of conjunction hubs in cognitive computations, producing above-chance task performance by integrating sensory and task rule activations within these hubs. The findings reveal the role of conjunction hubs in flexible cognitive computations and demonstrate the feasibility of using empirically-estimated neural network models to understand cognitive computations in the human brain.

Takuya Ito, Guangyu Robert Yang, Patryk Laurent, Douglas H. Schultz, Michael W. Cole