SPEF: Semantic Prior-Selected Expert Fusion for Working-Memory Prediction from Multimodal Brain Connectomes

Predicting cognition from functional and structural connectomes requires learning from thousands of correlated edges from comparatively modest neuroimaging cohorts. In order to stabilize the prediction framework, we introduce Semantic Prior-selected Expert Fusion (SPEF), a transparent framework that converts cognitive task descriptions and atlas region names into a frozen semantic atlas prior, selects compact functional-connectivity (FC) and structural-connectivity (SC) expert subspaces, and hierarchically fuses the experts in a strong data-driven connectome backbone. The prior is generated independently of participant imaging and behavioral data, validated against the canonical AAL116 ordering, normalized and frozen, and then transferred to connectivity space to define compact FC and SC subspaces. On 510 Human Connectome Project participants, five-seed, five-fold repeated nested cross-validation gives Pearson $r=0.2439$, RMSE $11.293$, and MAE $9.130$ for SPEF-ridge on a working memory task. Among the evaluated multimodal comparators and SPEF variants, the best Pearson correlation is $0.2439$, attained by both matched SPEF-ridge and cross-task SPEF-NC-ridge, while the latter achieves the lowest RMSE ($11.288$) and MAE ($9.121$). The working-memory semantic prior also exceeds shuffled and random prior controls by $+0.0127$ and $+0.0148$ mean seed-wise correlation, respectively, while the matched and cross-task semantic priors perform similarly. Prediction-linked coefficient maps show reproducible FC and SC structure. Relative to an architecture-matched no-prior full-connectome ridge expert, SPEF increases FC edge-rank stability ($0.723$ vs. $0.544$) and yields a substantially less adverse top-ROI perturbation contrast ($-0.0017$ vs. $-0.0864$), while other stability measures remain mixed. SPEF provides an auditable framework for introducing language-derived cognitive knowledge as an anatomically grounded inductive bias for multimodal connectome based prediction and candidate biomarker analyses.