Proceedings of the Annual Meeting of the Cognitive Science Society

Transcranial direct current stimulation (tDCS) is a promising non-invasive intervention for Alzheimer's disease (AD). However, current clinical practices rely on generic stimulation targets, which fail to account for individual differences among AD patients, leading to heterogeneous outcomes. To enable personalized neuromodulation, we propose a normative model and connectome-driven target prioritization decision (NCTPD) model. Using a functional connectivity (FC) normative model, we identify each patient's abnormal FC patterns and abnormal regions. We further quantify the coupling strength between candidate targets and abnormal regions via individual structural connectivity (SC), simulating the propagation mechanism of SC-guided current effects to multiple abnormal brain regions, thereby establishing target prioritization. Finally, we perform virtual stimulation on the prioritized targets in the digital twin brain models of 21 AD patients. Higher-ranked targets show more significant regression of abnormal FC toward the normative reference, indicating superior modulation effects. These results validate the effectiveness of the NCTPD model.