Usuda, N. | bioRxiv Science

bioRxiv2026

Large-scale dimensional behavioral profiling dissociates fear memory from locomotor confounds in mice: The necessity of baseline-normalized metrics

Fear conditioning is widely used to assess associative memory in mice, yet percent freezing conflates memory with baseline locomotor and anxiety-related traits. A systematic survey of recent studies (2020-2025) found that fewer than 1% statistically integrate locomotor activity into freezing analyses. Here, we address this gap using a large-scale dataset of >10,000 mice across >160 comparisons, including genetic mutations, pharmacological interventions and aging, tested in 15 standardized behavioral paradigms. Conventional freezing scores covaried strongly with general locomotor activity, obscuring memory-related phenotypes. Multiple factor analysis identified two principal behavioral dimensions, locomotor activity and learning/memory: conventional freezing aligned with the locomotor dimension, whereas freezing subtraction and the activity suppression ratio mapped onto the memory dimension and improved detection of synaptic plasticity phenotypes. These analyses show that baseline locomotor normalization is essential for interpreting fear conditioning as a memory assay and provide an open framework for selecting and reporting locomotor-normalized metrics.

neuroscience↗

bioRxiv2024

Accurate localization of cortical and subcortical sources of M/EEG signals by a convolutional neural network with a realistic head conductivity model: Validation with M/EEG simulation, evoked potentials, and invasive recordings

While electroencephalography (EEG) and magnetoencephalography (MEG) are well-established non-invasive methods in neuroscience and clinical medicine, they suffer from low spatial resolution. Particularly challenging is the accurate localization of subcortical sources of M/EEG, which remains a subject of debate. To address this issue, we propose a four-layered convolutional neural network (4LCNN) designed to precisely locate both cortical and subcortical source activity underlying M/EEG signals. The 4LCNN was trained using a vast dataset generated by forward M/EEG simulations based on a realistic head volume conductor model. The 4LCNN implicitly learns the characteristics of M/EEG and their sources from the training data without need for explicitly formulating and fine-tuning optimal priors, a common challenge in conventional M/EEG source imaging techniques. We evaluated the efficacy of the 4LCNN model on a validation dataset comprising forward M/EEG simulations and two types of real experimental data from humans: 1) somatosensory evoked potentials recorded by EEG, and 2) simultaneous recordings from invasive electrodes implanted in the brain and MEG signals. Our results demonstrate that the 4LCNN provides robust and superior estimation accuracy compared to conventional M/EEG source imaging methods, aligning well with established neuroscience knowledge. Notably, the accuracy of the subcortical regions was as accurate as that of the cortical regions. The 4LCNN method, as a data-driven approach, enables accurate source localization of M/EEG signals, including in subcortical regions, suggesting future contributions to various research endeavors such as contributions to the clinical diagnosis, understanding of the pathophysiology of various neuronal diseases and basic brain functions.

neuroscience↗