Abstract
Neural codes are believed to have adapted to the statistical properties of the natural environment. However, the principles that govern the organization of ensemble activity in the visual cortex during natural visual input are unknown. We recorded populations of up to 500 neurons in the mouse primary visual cortex and characterized the structure of their activity, comparing responses to natural movies with those to control stimuli. We found that higher order correlations in natural scenes induced a sparser code, in which information is encoded by reliable activation of a smaller set of neurons and can be read out more easily. This computationally advantageous encoding for natural scenes was state-dependent and apparent only in anesthetized and active awake animals, but not during quiet wakefulness. Our results argue for a functional benefit of sparsification that could be a general principle governing the structure of the population activity throughout cortical microcircuits.
This is a preview of subscription content, access via your institution
Access options
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Rent or buy this article
Prices vary by article type
from$1.95
to$39.95
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
References
Attneave, F. Some informational aspects of visual perception. Psychol. Rev. 61, 183–193 (1954).
Barlow, H.B. The coding of sensory messages. Curr. Probl. Anim. Behav. 331–360 (1961).
Simoncelli, E.P. & Olshausen, B.A. Natural image statistics and neural representation. Annu. Rev. Neurosci. 24, 1193–1216 (2001).
Olshausen, B.A. & Field, D.J. Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381, 607–609 (1996).
Bell, A.J. & Sejnowski, T.J. The 'independent components' of natural scenes are edge filters. Vision Res. 37, 3327–3338 (1997).
Field, D.J. What is the goal of sensory coding? Neural Comput. 6, 559–601 (1994).
Gerhard, H.E., Wichmann, F.A. & Bethge, M. How sensitive is the human visual system to the local statistics of natural images? PLoS Comput. Biol. 9, e1002873 (2013).
Vinje, W.E. & Gallant, J.L. Sparse coding and decorrelation in primary visual cortex during natural vision. Science 287, 1273–1276 (2000).
Yen, S.-C., Baker, J. & Gray, C.M. Heterogeneity in the responses of adjacent neurons to natural stimuli in cat striate cortex. J. Neurophysiol. 97, 1326–1341 (2007).
Willmore, B.D.B., Mazer, J.A. & Gallant, J.L. Sparse coding in striate and extrastriate visual cortex. J. Neurophysiol. 105, 2907–2919 (2011).
Weliky, M., Fiser, J., Hunt, R.H. & Wagner, D.N. Coding of natural scenes in primary visual cortex. Neuron 37, 703–718 (2003).
Yao, H., Shi, L., Han, F., Gao, H. & Dan, Y. Rapid learning in cortical coding of visual scenes. Nat. Neurosci. 10, 772–778 (2007).
Tolhurst, D.J., Smyth, D. & Thompson, I.D. The sparseness of neuronal responses in ferret primary visual cortex. J. Neurosci. 29, 2355–2370 (2009).
Felsen, G., Touryan, J., Han, F. & Dan, Y. Cortical sensitivity to visual features in natural scenes. PLoS Biol. 3, e342 (2005).
Kampa, B.M., Roth, M.M., Gobel, W. & Helmchen, F. Representation of visual scenes by local neuronal populations in layer 2/3 of mouse visual cortex. Front. Neural Circuits 5, 18 (2011).
Cotton, R.J., Froudarakis, E., Storer, P., Saggau, P. & Tolias, A.S. Three-dimensional mapping of microcircuit correlation structure. Front. Neural Circuits 7, 151 (2013).
Freeman, J., Ziemba, C.M., Heeger, D.J., Simoncelli, E.P. & Movshon, J.A. A functional and perceptual signature of the second visual area in primates. Nat. Neurosci. 16, 974–981 (2013).
Garaschuk, O., Milos, R.-I. & Konnerth, A. Targeted bulk-loading of fluorescent indicators for two-photon brain imaging in vivo. Nat. Protoc. 1, 380–386 (2006).
Schüz, A. & Palm, G. Density of neurons and synapses in the cerebral cortex of the mouse. J. Comp. Neurol. 286, 442–455 (1989).
Baudot, P. et al. Animation of natural scene by virtual eye-movements evokes high precision and low noise in V1 neurons. Front. Neural Circuits 7, 206 (2013).
Wallace, D.J. et al. Rats maintain an overhead binocular field at the expense of constant fusion. Nature 498, 65–69 (2013).
Roweis, S.T. & Saul, L.K. Nonlinear dimensionality reduction by locally linear embedding. Science 290, 2323–2326 (2000).
Eichhorn, J., Sinz, F. & Bethge, M. Natural image coding in V1: how much use is orientation selectivity? PLoS Comput. Biol. 5, e1000336 (2009).
Hosoya, T., Baccus, S.A. & Meister, M. Dynamic predictive coding by the retina. Nature 436, 71–77 (2005).
Carandini, M. & Heeger, D.J. Normalization as a canonical neural computation. Nat. Rev. Neurosci. 13, 51–62 (2012).
Haider, B. et al. Synaptic and network mechanisms of sparse and reliable visual cortical activity during nonclassical receptive field stimulation. Neuron 65, 107–121 (2010).
Yeh, C.-I., Xing, D. & Shapley, R.M. 'Black' responses dominate macaque primary visual cortex V1. J. Neurosci. 29, 11753–11760 (2009).
Niell, C.M. & Stryker, M.P. Modulation of visual responses by behavioral state in mouse visual cortex. Neuron 65, 472–479 (2010).
Polack, P.-O., Friedman, J. & Golshani, P. Cellular mechanisms of brain state-dependent gain modulation in visual cortex. Nat. Neurosci. 16, 1331–1339 (2013).
Bennett, C., Arroyo, S. & Hestrin, S. Subthreshold mechanisms underlying state-dependent modulation of visual responses. Neuron 80, 350–357 (2013).
Sinz, F. & Bethge, M. Temporal adaptation enhances efficient contrast gain control on natural images. PLoS Comput. Biol. 9, e1002889 (2013).
Schwartz, O. & Simoncelli, E.P. Natural signal statistics and sensory gain control. Nat. Neurosci. 4, 819–825 (2001).
Zylberberg, J., Murphy, J.T. & DeWeese, M.R. A sparse coding model with synaptically local plasticity and spiking neurons can account for the diverse shapes of v1 simple cell receptive fields. PLoS Comput. Biol. 7, e1002250 (2011).
Bonin, V., Histed, M.H., Yurgenson, S. & Reid, R.C. Local diversity and fine-scale organization of receptive fields in mouse visual cortex. J. Neurosci. 31, 18506–18521 (2011).
Smith, S.L. & Hausser, M. Parallel processing of visual space by neighboring neurons in mouse visual cortex. Nat. Neurosci. 13, 1144–1149 (2010).
Barak, O., Rigotti, M. & Fusi, S. The sparseness of mixed selectivity neurons controls the generalization–discrimination trade-off. J. Neurosci. 33, 3844–3856 (2013).
Rigamonti, R., Brown, M.A. & Lepetit, V. Are sparse representations really relevant for image classification?. in Comput. Vis. Pattern Recognit. CVPR 2011 IEEE Conf. on 1545–1552, http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5995313 (IEEE, 2011).
Olshausen, B.A. & Field, D.J. Sparse coding of sensory inputs. Curr. Opin. Neurobiol. 14, 481–487 (2004).
Kalatsky, V.A. & Stryker, M.P. New paradigm for optical imaging: temporally encoded maps of intrinsic signal. Neuron 38, 529–545 (2003).
Wang, Q. & Burkhalter, A. Area map of mouse visual cortex. J. Comp. Neurol. 502, 339–357 (2007).
Olsen, S.R., Bortone, D.S., Adesnik, H. & Scanziani, M. Gain control by layer six in cortical circuits of vision. Nature 483, 47–52 (2012).
Komai, S., Denk, W., Osten, P., Brecht, M. & Margrie, T.W. Two-photon targeted patching (TPTP) in vivo. Nat. Protoc. 1, 647–652 (2006).
Pelli, D.G. The VideoToolbox software for visual psychophysics: transforming numbers into movies. Spat. Vis. 10, 437–442 (1997).
Vogelstein, J.T. et al. Fast nonnegative deconvolution for spike train inference from population calcium imaging. J. Neurophysiol. 104, 3691–3704 (2010).
Duda, R.O., Hart, P.E. & Stork, D.G. Pattern Classification Part 1 (Wiley, 2000).
Berens, P. et al. A fast and simple population code for orientation in primate V1. J. Neurosci. 32, 10618–10626 (2012).
van Hateren, J.H. & van der Schaaf, A. Independent component filters of natural images compared with simple cells in primary visual cortex. Proc. Biol. Sci. 265, 359–366 (1998).
Acknowledgements
We thank W.J. Ma for discussions, P. Storer for help setting up the two-photon system, D. Murray, A. Laudano and A.M. Morgan for organizational help and V. Kalatsky for help with the intrinsic-imaging system. This work was supported by US National Institutes of Health grants NIH-Pioneer award DP1-OD008301, NEI DP1-EY023176 and NIDA RO1-DA028525 to A.S.T.; a McKnight Scholar Award and Beckman Foundation Young Investigator Award to A.S.T.; NEI P30-EY002520; the German National Academic Foundation (P.B.); the Bernstein Center for Computational Neuroscience (FKZ 01GQ1002); and the German Excellency Initiative through the Centre for Integrative Neuroscience Tübingen (EXC307).
Author information
Authors and Affiliations
Contributions
E.F., P.B., M.B. and A.S.T. designed the study; E.F. and P.B. created the stimuli; R.J.C., E.F. and P.S. built the two-photon microscope; E.F., R.J.C. and D.Y. wrote preprocessing code; E.F. performed the experiments; E.F., P.B. and A.S.E. analyzed the data; E.F., A.S.E., F.H.S. and M.B. performed the modeling; and P.B. and E.F. wrote the paper with input from A.S.E., F.H.S., M.B. and A.S.T.; P.B. and A.S.T. supervised all stages of the project.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Integrated supplementary information
Supplementary Figure 1 Examples of raw data for 3D and 2D scanning
(a) 3D scanning: ΔF/F calcium signals from the 20 cells shown in Fig. 1d out of the 139 total recorded in this site. The 4 different colors indicate presentations of multiple repetitions of the 4 different movies. (b) 2D scanning: Raw ΔF/F calcium signals from the 19 cells recorded in this site.
Supplementary Figure 2 Histograms of all measures of population activity
Histograms of all the measures of population activity shown in Fig. 3 for main dataset (L2/3; anesthetized animals; n = 315 sites). Columns from left to right: histograms for natural movies, phase scrambled movies and the percentage difference between the two conditions relative to the natural condition. Black line and number indicate the median of each distribution.
Supplementary Figure 3 Dependence of sparseness on population and bin size
Population sparseness for 100 ms, 500 ms and 1000 ms bins (indicated by the line style) and natural (blue) and phase scrambled (red) movies. Population sparseness does not depend on population size but larger bin sizes reduce the apparent sparseness. Shaded area around the mean represents ± 1SEM.
Supplementary Figure 4 Correlation of lifetime and population sparseness
(a) Scatter plot of population and lifetime sparseness for the main dataset (N=315). The two measures are highly correlated (linear correlation, r = 0.87; p < 0.001). Gray line indicates best fitting linear regression. (b) Scatter plot of difference in population and lifetime sparseness between the responses to natural and phase scrambled stimuli. The two measures are strongly correlated (linear correlation, r = 0.61, p < 0.001).
Supplementary Figure 5 Control analysis for the effect of eye movements
Median difference in measures of population activity between unprocessed and phase scrambled version of the virtual reality environment movies generated by Wallace et al.20 (n = 13 sites; blue), a natural image sequence with motion vectors extracted from the virtual reality movie and their phase scrambled version (n = 13 sites; red), and between stimulation with natural and phase scrambled movies from the mouse cam lacking eye movements (blue). Positive values indicate that the measure is higher under stimulation with natural stimuli. Error bars encompass the 95%-confidence intervals of the median.
Supplementary Figure 6 Measures of population activity for the V1 models
(a) Median of the measures of population activity for the population responses of model 1 (LNP). Error bars encompass the 95%-confidence intervals of the median. (b) As in (a) but for model 2 (adaptive non-linearity). (c) As in (a) but for model 3 (divisive normalization). (d) Median difference in measures of population activity for the population responses between natural and phase scrambled sets separately for Model 1, Model 2, Model 3 and in vivo data (purple, green, yellow and gray, respectively). Positive values indicate that the measure is higher under stimulation with natural movies. Error bars encompass the 95%-confidence intervals of the median.
Supplementary Figure 7 Example behavioral traces during the awake experiments.
Linear pupil displacement from the median position across the experiment in millimeters, the whisker velocity and the ball velocity in mm/s are shown from top to bottom (red, yellow and blue, respectively). Periods with large pupil displacements, or running and whisking activity between the upper and lower boundaries (dashed lines), were discarded (gray background). Periods during which the animal was not whisking or not running were considered as quiet (green background, below green dashed line; see Methods). Periods during which the animal was either whisking, running or both were considered as active (purple trace, above purple dashed line; see Methods).
Supplementary information
Supplementary Text and Figures
Supplementary Figures 1–7 and Supplementary Tables 1 and 2 (PDF 1118 kb)
Example natural movie stimulus
Example of a natural stimulus used in this study acquired with a camera mounted on the head of a mouse (MP4 5019 kb)
Example phase-scrambled movie generated from Supplementary Movie 1.
Example of a phase scrambled stimulus generated from Supplementary Movie 1 (MP4 6144 kb)
Rights and permissions
About this article
Cite this article
Froudarakis, E., Berens, P., Ecker, A. et al. Population code in mouse V1 facilitates readout of natural scenes through increased sparseness. Nat Neurosci 17, 851–857 (2014). https://doi.org/10.1038/nn.3707
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/nn.3707
This article is cited by
-
Uncovering hidden network architecture from spiking activities using an exact statistical input-output relation of neurons
Communications Biology (2023)
-
SmaRT2P: a software for generating and processing smart line recording trajectories for population two-photon calcium imaging
Brain Informatics (2022)
-
Reduced neural activity but improved coding in rodent higher-order visual cortex during locomotion
Nature Communications (2022)
-
Reduced excitatory neuron activity and interneuron-type-specific deficits in a mouse model of Alzheimer’s disease
Communications Biology (2022)
-
State-dependent pupil dilation rapidly shifts visual feature selectivity
Nature (2022)