Predictive coding of natural images by V1 firing rates and rhythmic synchronization

被引:25
|
作者
Uran, Cem [1 ,5 ]
Peter, Alina [1 ]
Lazar, Andreea [1 ]
Barnes, William [1 ,2 ]
Klon-Lipok, Johanna [1 ,2 ]
Shapcott, Katharine A. [1 ,3 ]
Roese, Rasmus [1 ]
Fries, Pascal [1 ,4 ]
Singer, Wolf [1 ,2 ,3 ]
Vinck, Martin [1 ,5 ]
机构
[1] Max Planck Gesell, Ernst Strungmann Inst ESI Neurosci Cooperat, D-60528 Frankfurt, Germany
[2] Max Planck Inst Brain Res, D-60438 Frankfurt, Germany
[3] Frankfurt Inst Adv Studies, D-60438 Frankfurt, Germany
[4] Radboud Univ Nijmegen, Donders Inst Brain Cognit & Behav, Dept Biophys, NL-6525 AJ Nijmegen, Netherlands
[5] Radboud Univ Nijmegen, Donders Ctr Neurosci, Dept Neuroinformat, NL-6525 AJ Nijmegen, Netherlands
关键词
PRIMARY VISUAL-CORTEX; GAMMA-OSCILLATIONS; MACAQUE V1; NEURONAL SYNCHRONY; SURROUND SUPPRESSION; LUMINANCE-CONTRAST; SPATIAL SUMMATION; SALIENCY MAP; FREQUENCY; MODULATION;
D O I
10.1016/j.neuron.2022.01.002
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
Predictive coding is an important candidate theory of self-supervised learning in the brain. Its central idea is that sensory responses result from comparisons between bottom-up inputs and contextual predictions, a process in which rates and synchronization may play distinct roles. We recorded from awake macaque V1 and developed a technique to quantify stimulus predictability for natural images based on self-supervised, generative neural networks. We find that neuronal firing rates were mainly modulated by the contextual predictability of higher-order image features, which correlated strongly with human perceptual similarity judgments. By contrast, V1 gamma (gamma)-synchronization increased monotonically with the contextual predictability of low-level image features and emerged exclusively for larger stimuli. Consequently, gamma-synchronization was induced by natural images that are highly compressible and low-dimensional. Natural stimuli with low predictability induced prominent, late-onset beta (beta)-synchronization, likely reflecting cortical feedback. Our findings reveal distinct roles of synchronization and firing rates in the predictive coding of natural images.
引用
收藏
页码:1240 / +
页数:27
相关论文
共 50 条
  • [41] Predictive feedback to V1 dynamically updates with sensory input
    Edwards, Grace
    Vetter, Petra
    McGruer, Fiona
    Petro, Lucy S.
    Muckli, Lars
    SCIENTIFIC REPORTS, 2017, 7
  • [42] Colour coding in LGN and V1 revealed by fMRI pattern classification
    Wucrger, S. M.
    Oxley, D.
    Marsman, J-B
    Parkes, L.
    PERCEPTION, 2007, 36 : 167 - 167
  • [43] Practising orientation identification improves orientation coding in V1 neurons
    Aniek Schoups
    Rufin Vogels
    Ning Qian
    Guy Orban
    Nature, 2001, 412 : 549 - 553
  • [44] Practising orientation identification improves orientation coding in V1 neurons
    Schoups, A
    Vogels, R
    Qian, N
    Orban, G
    NATURE, 2001, 412 (6846) : 549 - 553
  • [45] The simultaneous coding of orientation and contrast in the responses of V1 complex cells
    Gawne, TJ
    EXPERIMENTAL BRAIN RESEARCH, 2000, 133 (03) : 293 - 302
  • [46] Sparse coding with an overcomplete basis set: A strategy employed by V1?
    Olshausen, BA
    Field, DJ
    VISION RESEARCH, 1997, 37 (23) : 3311 - 3325
  • [47] Optimal coding through divisive normalization models of V1 neurons
    Valerio, R
    Navarro, R
    NETWORK-COMPUTATION IN NEURAL SYSTEMS, 2003, 14 (03) : 579 - 593
  • [48] The simultaneous coding of orientation and contrast in the responses of V1 complex cells
    Timothy J. Gawne
    Experimental Brain Research, 2000, 133 : 293 - 302
  • [49] Spike-frequency adaptation as a mechanism for dynamic coding in V1
    Schwabe, L
    Adorján, P
    Obermayer, K
    NEUROCOMPUTING, 2001, 38 : 351 - 358
  • [50] The Orientation Selectivity of Spike-LFP Synchronization in Macaque V1 and V4
    Li, Zhaohui
    Gao, Mengyu
    Wang, Yongtian
    FRONTIERS IN COMPUTATIONAL NEUROSCIENCE, 2019, 13