five

A Role for Ocular Dominance in Binocular Integration - Dataset and code.

收藏
Mendeley Data2024-05-10 更新2024-06-29 收录
下载链接:
https://zenodo.org/records/8161772
下载链接
链接失效反馈
官方服务:
资源简介:
Dataset and code used for data analysis featured in "A Role for Ocular Dominance in Binocular Integration". Abstract: Neurons in primate primary visual cortex (V1) combine left and right-eye information to form a binocular output. Controversy surrounds whether ocular dominance, the preference of these neurons for one eye over the other, is functionally relevant. Here, we demonstrate that ocular dominance impacts gain-control during binocular combination. We recorded V1 spiking activity while monkeys passively viewed grating stimuli. Gratings were either presented to one eye (monocular), both eyes with the same contrasts (binocular balanced), or both eyes with different contrasts (binocular imbalanced). We found that contrast placed in a neurons’ dominant eye was weighted more strongly than contrast placed in a neurons’ non-dominant eye. This asymmetry covaried with neurons’ ocular dominance. We then tested whether accounting for ocular dominance within divisive normalization improves the fit to neural data. We found that ocular dominance significantly improved model performance, with interocular normalization providing the best fits. These findings suggest that V1 ocular dominance is relevant for response normalization during binocular stimulation.

用于《双眼整合中优势眼的作用》("A Role for Ocular Dominance in Binocular Integration")一文数据分析的数据集与代码。摘要:灵长类初级视觉皮层(primary visual cortex, V1)的神经元会整合左右眼的视觉信息以生成双眼整合输出信号。目前学界对于优势眼(ocular dominance,即神经元对单侧眼睛的偏好性)是否具备功能相关性仍存在争议。本研究证实,在双眼信息整合过程中,优势眼会对增益调控产生影响。实验中,我们在猕猴被动观看光栅刺激时记录了其V1区的神经元放电活动(spiking activity)。光栅刺激分为三种呈现模式:单眼刺激(仅作用于单侧眼)、双眼平衡刺激(双侧眼对比度一致)以及双眼失衡刺激(双侧眼对比度存在差异)。我们发现,神经元对优势眼输入的对比度权重显著高于非优势眼输入的对比度权重,且这种权重不对称性与神经元的优势眼特性呈共变关系。随后,我们测试了在除法归一化(divisive normalization)模型中纳入优势眼因素是否能够提升对神经实验数据的拟合效果,结果显示,优势眼因素可显著改善模型性能,其中双眼间归一化模型的拟合效果最佳。上述研究结果表明,V1区的优势眼特性在双眼刺激下的神经元反应归一化过程中具备功能相关性。
创建时间:
2023-07-26
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作