Systematic reduction of the dimensionality of natural scenes allows accurate predictions of retinal ganglion cell spike outputs.
Systematic reduction of the dimensionality of natural scenes allows accurate predictions of retinal ganglion cell spike outputs.
复制标题
DOI：
10.1073/pnas.2121744119
复制
发表时间：
2022-11-16
影响因子：
11.1
通讯作者：
中科院分区：
文献类型：
--
作者：
在过去的60年里，大量的研究已经描述了包括视网膜神经节细胞在内的视觉神经元的空间感受野。推广这些模型来描述对自然图像复杂空间结构的反应仍然具有挑战性。在这里，我们确定了如何简化自然电影中的空间特征，同时最大限度地改变灵长类伞状神经节细胞的尖峰反应。这种方法识别了自然刺激的低维版本，这些版本保留了阳伞尖峰反应的∼80%的结构。这些低维刺激只需要了解细胞经典的中心环绕感受野。识别调节遮阳伞棘波反应的刺激特征为确定视网膜编码如何影响后续视觉加工提供了有用的工具。哺乳动物的视网膜通过一系列的线性和非线性电路机制将自然场景转换为视网膜神经节细胞(RGC)的尖峰输出。虽然许多个体整合机制已经被很好地理解，但我们对多种机制如何相互作用来编码自然输入中存在的复杂空间特征知之甚少。在这里，我们确定了自然场景中的关键空间特征，这些特征通过灵长类遮阳伞RGC进行形状编码。我们的方法确定了自然场景空间结构的简化，最小限度地改变了RGC的尖峰反应。我们观察到，将自然电影分成16个线性整合的区域描述了∼80%的遮阳伞RGC尖峰反应的结构；这种表现取决于区域的数量，而不是它们的精确空间位置。我们使用简化的刺激来设计高维的异构体，它概括了人们对自然主义电影的反应。最后，我们对视网膜计算进行了建模，这些计算将闪现的自然图像转换为一维的尖峰计数。
Numerous studies over the past 60 years have characterized spatial receptive fields of visual neurons, including retinal ganglion cells. Generalizing these models to describe responses to the complex spatial structure of natural images remains challenging. Here, we determine how spatial features in natural movies can be simplified while minimally altering spike responses of primate parasol ganglion cells. This approach identifies low-dimensional versions of natural stimuli that preserve ∼80% of the structure of parasol spike responses. These low-dimensional stimuli only require knowledge of a cell’s classical center-surround receptive field. Identifying the stimulus features that mediate parasol spike responses provides a useful tool for determining how retinal coding impacts subsequent visual processing. The mammalian retina engages a broad array of linear and nonlinear circuit mechanisms to convert natural scenes into retinal ganglion cell (RGC) spike outputs. Although many individual integration mechanisms are well understood, we know less about how multiple mechanisms interact to encode the complex spatial features present in natural inputs. Here, we identified key spatial features in natural scenes that shape encoding by primate parasol RGCs. Our approach identified simplifications in the spatial structure of natural scenes that minimally altered RGC spike responses. We observed that reducing natural movies into 16 linearly integrated regions described ∼80% of the structure of parasol RGC spike responses; this performance depended on the number of regions but not their precise spatial locations. We used simplified stimuli to design high-dimensional metamers that recapitulated responses to naturalistic movies. Finally, we modeled the retinal computations that convert flashed natural images into one-dimensional spike counts.
登录
查看更多内容
影响因子：
16.2
作者：
Schreyer HM;Gollisch T
通讯作者：
Gollisch T
影响因子：
5.5
作者：
BARLOW, HB
通讯作者：
BARLOW, HB
影响因子：
5.5
作者：
ENROTHCUGELL, C;FREEMAN, AW
通讯作者：
FREEMAN, AW
影响因子：
8.6
作者：
RUDERMAN, DL;BIALEK, W
通讯作者：
BIALEK, W
影响因子：
7.7
作者：
Shah, Nishal P.;Brackbill, Nora;Chichilnisky, E. J.
通讯作者：
Chichilnisky, E. J.