Abstract
The human brain is not static. It is a living, dynamic web of connections, constantly adapting to experience, injury, and intervention. This adaptability—what we call neuroplasticity—is at the heart of recovery, resilience, and innovation in modern neuroscience. In this issue of Brain Connectivity, we are proud to present the scientific articles that push the boundaries of how we understand and measure these networks, not just in the lab, but with implications for diagnosis, prognosis, and therapy in real-world clinical settings.
Our community continues to grow in both reach and impact. Increasingly, neurologists, psychiatrists, neuroradiologists, psychologists, rehabilitation specialists, neuroscientists, biomedical engineers, and data scientists are speaking a shared language—connectivity. At Brain Connectivity, our mission is to bring these voices together to foster translational progress: from theory to technology, from algorithm to bedside.
This issue brings together four exceptional studies that exemplify this vision. From Alzheimer’s disease and cerebral palsy to cognitive performance and mood modulation, these articles remind us that brain networks are more than maps—they are the mechanisms of change.
Adaptive Brain Graphs for Early Alzheimer’s Detection
In one of the most ambitious studies in this issue, Limei Song and colleagues from Yantai University and Shandong Second Medical University, China, introduce a novel artificial intelligence-based model designed to enhance the early diagnosis of Alzheimer’s disease. Their article entitled “Task-radMBNet: An improved Task-driven dynamic graph sparsity pattern radiomics-based morphological brain network for Alzheimer’s disease characterization” proposes a model combining structural brain imaging with advanced network analysis to highlight subtle but meaningful changes in brain morphology. But what sets this study apart is not just its technical sophistication—it is its scale and clinical relevance.
The model was trained and validated on 1351 magnetic resonance imaging (MRI) scans from participants spanning the cognitive spectrum—from healthy controls to individuals with mild cognitive impairment and dementia—drawn from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) and the European DTI Study on Dementia (EDSD). This transcontinental validation lends strong credibility to the model’s generalizability across diverse populations. Unlike traditional approaches that rely on fixed patterns, their new model dynamically adapts the network structure to the diagnostic task at hand. This flexibility allows it to focus more precisely on connections and brain regions relevant to Alzheimer’s pathology.
Using MRI of brain structure, specifically T1-weighted images, the authors extracted detailed radiomics features—quantitative markers of shape, intensity, and texture that are not visible to the eye of neuroradiologists. These features were then analyzed through a dual-channel graph model: one channel learning how brain regions are connected, the other analyzing their internal features. This fusion of structure and topology yielded impressive diagnostic accuracy for early-stage disease, even before overt cognitive symptoms arise.
The translational potential of this work is significant. It leverages widely available clinical imaging, meaning it could be implemented without the need for costly new scanning protocols. Its adaptability suggests promise for other neurological conditions, and its interpretability allows clinicians to visualize the connections that matter most for a given diagnosis. In short, it is a step toward explainable, accessible artificial intelligence for real-world neurodegenerative care.
Task Versus Rest: EEG Connectomes Predict Memory
Working memory—the mental ability to hold and manipulate information—is fundamental to daily life. Yet, predicting how well someone performs in this domain from brain data remains a major challenge. In this study, Anton Pashkov and colleagues, from the Federal Neurosurgical Center and South Ural State University, Russia, used high-density electroencephalography (EEG), a noninvasive and cost-effective method, to compare brain connectivity during rest and during a mental task involving memory.
Their article entitled “Direct comparison of EEG resting state and task functional connectivity patterns for predicting working memory performance using connectome-based predictive modeling” is one of the first of its kind to apply predictive modeling to both resting and active brain states using EEG. Their goal was to see which type of data—rest or task—would better forecast working memory performance. They found that task-related brain activity had a slight edge, particularly in the alpha and beta frequency bands, which are known to support attention and cognitive control.
Crucially, this study also tested how different technical choices—such as the number of brain regions analyzed or how brain connectivity is calculated—can influence results. By carefully benchmarking these variables, the researchers provide valuable guidance for future experiments. Their findings move us closer to real-time, individualized brain-based assessments of cognitive capacity, which could be especially useful in clinical settings like stroke rehabilitation, dementia screening, or attention deficit evaluation.
Motor Connectivity Reveals Functional Limits in Cerebral Palsy
Children with cerebral palsy often experience motor challenges that affect both arms, yet the brain mechanisms behind these difficulties remain only partially understood—especially during movement itself. This study by Sang Wook Lee and colleagues from the Catholic University of America and National Institutes of Health, United States, entitled “Dynamic task-related changes in EEG brain connectivity during a button-press task in children with and without bilateral cerebral palsy” used a 64-channel electroencephalography system to explore how brain connectivity changes during an actual motor task: pressing a button with one hand.
The researchers compared children with bilateral cerebral palsy to typically developing peers while they performed this task. What they found was striking. During movement execution, children with cerebral palsy showed significantly higher communication between motor and sensory areas, especially between the two sides of the brain. The more impaired the child’s movement, the stronger these connections became.
While this might sound paradoxical, it reflects the brain’s attempt to compensate for damage by recruiting additional pathways. But this increased effort may also signal inefficiency—too many neurons working to perform a simple action. These findings open the door to using brain connectivity as both a marker of impairment and a target for therapy. Perhaps more importantly, this study demonstrates that electroencephalography offers a child-friendly, movement-compatible way to evaluate brain function—without the constraints of traditional imaging.
Rewiring Emotional Circuits with Magnetic Stimulation
In this exploratory but thought-provoking study, entitled “Effects of transcranial magnetic stimulation on cognitive-affective task-based functional connectivity,” Merideth Addicott and colleagues, Wake Forest University, Durham Veterans Affairs Medical Center, and University of California San Diego, United States, investigated whether brain stimulation could influence the way the brain connects during emotional distress. Using a technique called repetitive transcranial magnetic stimulation, they targeted a specific area of the brain known to be functionally linked to emotional processing. Participants received stimulation over five days and completed a mental task designed to elicit cognitive stress, all while undergoing functional magnetic resonance imaging.
What the researchers found was that low-frequency stimulation modulated connectivity between the auditory cortex and the posterior insula—a deep brain structure involved in the experience of negative emotions and bodily awareness. In essence, the stimulation appeared to dampen the brain’s usual pattern of reaction to emotional challenge.
Although preliminary, these results suggest that noninvasive stimulation could one day help reshape dysfunctional brain circuits in mood disorders, stress-related conditions, or cognitive fatigue. The study also highlights the value of assessing connectivity during task engagement, rather than only at rest—an important direction for future clinical applications.
Closing Reflections
The brain is a system of systems. And as these studies show, it is a system that changes—with time, with challenge, and with care. Whether through adaptive modeling in dementia, EEG-based memory profiling, child-focused rehabilitation, or modulation of emotion circuits, this issue celebrates a unifying message: connectivity is not just what we observe—it is what we can influence.
As we continue to grow as a journal, we remain committed to supporting research that bridges laboratory insights with patient lives. Upcoming issues will feature special themes on pediatric neurodevelopment, brain-computer interface ethics, and imaging-based biomarkers in psychiatry and neurorehabilitation.
Thank you for being part of this shared endeavor. Stay curious, stay connected.