Background: Age-related changes in the brain-parenchyma may alter the brains functional connectivity. Studies have consistently reported significant age-effects in the default mode network (DMN). Less is known about sex differences and the interaction effect with older age. Both independent component analysis and seed-based analysis have been widely used in the literature.

Method: We investigated 29 young (age=25.0±3.4) and 31 older (n=31, age=68.4±2.7) healthy adults (♀50%) with a resting-state fMRI protocol at 3T. We evaluated voxel-wise differences in brain functional connectivity in relation to age and sex of the participants by taking advantage of a recently developed quantitative data-driven analysis (QDA) method, where the connection counters (CCI) and strength (CSI) indies for each voxel in the brain are measured.

Results: Compared with the CSI result, more extensive brain regions showed significantly reduction in CCI in older age. Besides the brain regions involved in the DMN, we found that older age induced also functional connectivity decline in thalamus and insula cortex. On average, females had lower CCI and CSI in the right middle frontal gyrus. The anterior cingulate, right middle frontal gyrus, and left lentiform nucleus showed age and sex interaction effect. The finding of reduced functional connectivity in older adults was also confirmed by the result from linear regression analysis.

Conclusion: Age-related effect as detected by the QDA method is largely consistently with previously reported findings. Furthermore, our study added interesting new findings in age-related differences. This indicates that QDA provides a sensitive model-free approach for voxel-wise analysis of functional connectivity.