Format

Send to

Choose Destination
PLoS Comput Biol. 2019 Apr 23;15(4):e1006897. doi: 10.1371/journal.pcbi.1006897. eCollection 2019 Apr.

Deep convolutional models improve predictions of macaque V1 responses to natural images.

Cadena SA1,2,3, Denfield GH3,4, Walker EY3,4, Gatys LA1,2, Tolias AS2,3,4,5, Bethge M1,2,3,6, Ecker AS1,2,3.

Author information

1
Centre for Integrative Neuroscience and Institute for Theoretical Physics, University of Tübingen, Tübingen, Germany.
2
Bernstein Center for Computational Neuroscience, Tübingen, Germany.
3
Center for Neuroscience and Artificial Intelligence, Baylor College of Medicine, Houston, Texas, United States of America.
4
Department of Neuroscience, Baylor College of Medicine, Houston, Houston, Texas, United States of America.
5
Department of Electrical and Computer Engineering, Rice University, Houston, Houston, Texas, United States of America.
6
Max Planck Institute for Biological Cybernetics, Tübingen, Germany.

Abstract

Despite great efforts over several decades, our best models of primary visual cortex (V1) still predict spiking activity quite poorly when probed with natural stimuli, highlighting our limited understanding of the nonlinear computations in V1. Recently, two approaches based on deep learning have emerged for modeling these nonlinear computations: transfer learning from artificial neural networks trained on object recognition and data-driven convolutional neural network models trained end-to-end on large populations of neurons. Here, we test the ability of both approaches to predict spiking activity in response to natural images in V1 of awake monkeys. We found that the transfer learning approach performed similarly well to the data-driven approach and both outperformed classical linear-nonlinear and wavelet-based feature representations that build on existing theories of V1. Notably, transfer learning using a pre-trained feature space required substantially less experimental time to achieve the same performance. In conclusion, multi-layer convolutional neural networks (CNNs) set the new state of the art for predicting neural responses to natural images in primate V1 and deep features learned for object recognition are better explanations for V1 computation than all previous filter bank theories. This finding strengthens the necessity of V1 models that are multiple nonlinearities away from the image domain and it supports the idea of explaining early visual cortex based on high-level functional goals.

PMID:
31013278
PMCID:
PMC6499433
DOI:
10.1371/journal.pcbi.1006897
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Public Library of Science Icon for PubMed Central
Loading ...
Support Center