Domain-general and stimulus-specific functional connectivity networks supporting working memory for faces, body parts, tools, and places

Working Memory (WM) load is a central WM characteristic that influences different task-dependent activities. Building on previous research, demonstrating that both WM load and visual stimulus category (faces, body parts, tools, and places) influence brain activity. This study focuses on differences in functional connectivity networks, using the network-based statistics (NBS) approach. For this purpose, we retrieved the WM functional MRI dataset from the Human Connectome Project dataset for 100 subjects. The results indicated that functional connectivity was predominantly negative under working memory load (0-back > 2-back) across all categories, with stronger connectivity observed within the Visual, Frontoparietal, and Dorsal Attention networks. Interestingly, the Default Mode Network also showed increased connectivity under greater working memory load, aligning with recent findings that suggest its involvement under higher working memory demands. These results demonstrate that there are both domain-general and stimulus-specific functional networks supporting working memory. This micropublication was created as part of the Neuromatch Impact Scholars Program 2025. Project Website: https://impact-scholars.github.io/yilmaz-2026-connectivity-memory-load Repository: https://github.com/impact-scholars/yilmaz-2026-connectivity-memory-load