Sadeghzadeh, S. | Search bioRxiv

bioRxiv2025

Experience-Dependent Plasticity of Thalamoprefrontal Circuitry Characterizes Learning Across Species

Learning is an adaptive process which is thought to require precise coordination among multiple brain regions over time. Thalamoprefrontal circuitry is central to multiple domains of cognitive control and executive function, yet its evolution during learning is not characterized. In this study, we explored thalamoprefrontal circuit signaling in mice and connectivity dynamics in humans over the course of learning a homologous psychomotor task. In mice, we targeted a genetically encoded Ca2+indicator to the mediodorsal-prefrontal (MdT-dmPFC) projection and used fiber photometry to record synaptic dorsomedial PFC (dmPFC) activity. As learning sessions progressed, all mice demonstrated a decrease in reaction time (RT) performance on successful trials (p<0.001). In concomitant synaptic MdT-dmPFC circuit activity across sessions, we observed a significant increase in activity during the anticipatory (p<0.01) and reward retrieval (p<0.01) periods, and a non-significant trend towards an increase in preparatory activity (p<0.15). However, following the learning period, during task re-exposure, we observed a significant shift in circuit activity, away from anticipatory (p<0.001) and towards the preparatory period (p<0.001) over the course of re-exposure sessions. Furthermore, we observed an emergence of a learner (decrease in RT) and a non-learner group (increase in RT) during the task re-exposure period. Over the course of a single analogous task session in humans, we also observed a learner and a non-learner group. When analyzing the thalamoprefrontal regional connectivity dynamics of early and late trials for learners, we observed a significant increase in low frequency and decrease in high frequency synchrony (connectivity tilt) in thalamoprefrontal pathways during preparatory and anticipatory periods. Interestingly, pairwise interactions specifically between the anterior corona radiata (ACR) and superior frontal gyrus (SFG), the human homolog to the genetically targeted MdT-dmPFC circuitry in mice, in fact demonstrated a robustly opposite connectivity tilt effect distinguishing learners from non-learners (Cohens d > 2). Overall, these findings may provide cross-species evidence of novel, conserved thalamoprefrontal circuit mechanisms of adaptive learning.

neuroscience↗

bioRxiv2025

ConvNeXt-Driven Detection of Alzheimers Disease: A Benchmark Study on Expert-Annotated AlzaSet MRI Dataset Across Anatomical Planes

Alzheimers disease (AD) is a leading worldwide cause of cognitive impairment, necessitating accurate, inexpensive diagnostic tools to enable early recognition. In this study, we present a robust deep learning approach for AD classification based on structural MRI scans, ConvNeXt, an emergent convolutional architecture inspired by vision transformers. We further present AlzaSet, a novel, expertly labeled clinical dataset with axial, coronal, and sagittal perspectives from AD and cognitively normal control subjects. Three ConvNeXt sizes (Tiny, Small, Base) were compared and benchmarked against existing state-of-the-art CNN models (VGG16, VGG19, InceptionV3, DenseNet1 21). ConvNeXt-Base consistently outperformed the other models on coronal slices with accuracy of 98.37% and an AUC of 0.992. Coronal views were determined to be most diagnostically informative, with emphasis on visualization of the medial temporal lobe. Moreover, comparison with recent ensemble-based techniques showed that superior performance with comparable computational efficiency. These results indicate that ConvNeXt-capable models applied to clinically curated data sets have strong potential to provide scalable, real-time AD screening in diverse settings, including both high-resource and resource-constrained settings.

neuroscience↗