<p>Due to hardware limitations, the spatial resolution of hyperspectral images (HSIs) captured by satellites and various sensors is typically low. To obtain high-resolution HSIs (HR-HSIs), an effective approach is to fuse and reconstruct low-resolution HSIs (LR-HSIs) with high-resolution multispectral images (HR-MSIs). However, existing methods often process spatial and spectral features separately or sequentially, limiting cross-modal information exchange and causing feature degradation during fusion. In this study, we propose a multi-information interactive feature aggregation network for HR-HSI reconstruction. Specifically, our method employs two paths: one for spatial reconstruction (HR-MSI) and the other for spectral resolution recovery (LR-HSI). These paths aggregate multi-level features in a parallel and hierarchical manner, effectively reducing feature loss. We utilize a spatial attention module at the beginning of the HR-MSI path to reconstruct spatial information and an improved self-attention module at the end to model spatial context associations and recover high-frequency spatial details. In the LR-HSI path, we integrate spatial features from the HR-MSI path multiple times, compensating for spatial resolution loss during upsampling. Additionally, a spatial-spectral attention module adaptively refines the fused feature maps. We conduct experiments on three public datasets and show that our method achieves competitive or superior performance compared to five state-of-the-art approaches in terms of quantitative metrics for HR-HSI reconstruction.</p>