Negative BOLD Responses Surpass Positive Responses in Task Specificity, Reflecting Neural Reconfigurations Better Than Functional Connectivity

ObjectiveTo investigate whether the Negative BOLD Response (NBR) is more task-specific than the Positive BOLD Response (PBR) during cognitive tasks and to determine whether task-evoked activity reflects brain reconfigurations during different tasks better than functional connectivity. MethodsFunctional Magnetic Resonance Imaging (fMRI) data were collected from 214 participants under 50 years old (152 in Dataset 1 and 62 in Dataset 2) performing twelve cognitive tasks spanning vocabulary, speed of processing, fluid reasoning, and memory domains. Data analysis included subject-level and group-level analyses, focusing on comparing the spatial patterns and task specificity of NBR and PBR through similarity measures using Dice coefficients. Additionally, functional connectivity was assessed using the Multi-session Hierarchical Bayesian Model (MS-HBM) to evaluate its sensitivity to task-induced brain reconfigurations compared to task-evoked activity. ResultsNBR demonstrated significantly greater task specificity compared to PBR across all cognitive tasks, with lower mean Dice coefficients for NBR maps (mean: 0.44, SD: 0.13) than for PBR maps (mean: 0.67, SD: 0.09; t(65) = 18.38, p < 0.001). Functional connectivity analyses indicated that the default mode network (DMN) remained stable across tasks, suggesting that task-evoked activity reflects task-specific brain reconfigurations better than functional connectivity. ConclusionThe findings confirm that NBR is inherently more task-specific than PBR and that task-evoked activity provides a more sensitive measure of task-specific neural reconfigurations than functional connectivity. This enhances our understanding of the neural mechanisms underlying cognitive processes and highlights the importance of considering NBR in cognitive neuroscience research.