Abstract
Advances in neurobiology suggest that neuronal response of the primary visual cortex to natural stimuli may be attributed to sparse approximation of images, encoding stimuli to activate specific neurons although the underlying mechanisms are still unclear. The responses of retinal ganglion cells (RGCs) to natural and random checkerboard stimuli were simulated using fast independent component analysis. The neuronal response to stimuli was measured using kurtosis and Treves–Rolls sparseness, and the kurtosis, lifetime and population sparseness were analyzed. RGCs exhibited significant lifetime sparseness in response to natural stimuli and random checkerboard stimuli. About 65 and 72% of RGCs do not fire all the time in response to natural and random checkerboard stimuli, respectively. Both kurtosis of single neurons and lifetime response of single neurons values were larger in the case of natural than in random checkerboard stimuli. The population of RGCs fire much less in response to random checkerboard stimuli than natural stimuli. However, kurtosis of population sparseness and population response of the entire neurons were larger with natural than random checkerboard stimuli. RGCs fire more sparsely in response to natural stimuli. Individual neurons fire at a low rate, while the occasional “burst” of neuronal population transmits information efficiently.
Similar content being viewed by others
References
Atick JJ (1992) Could information theory provide an ecological theory of sensory processing? Network 22(1–4):4–44
Bakouie F, Pishnamazi M, Zeraati R, Gharibzadeh S (2017) Scale-freeness of dominant and piecemeal perceptions during binocular rivalry. Cogn Neurodyn 11(4):319–326
Barlow HB (1961) Possible principles underlying the transformation of sensory messages. MIT Press, Cambridge, pp 217–234
Barranca VJ, Kovacic G, Zhou D et al (2014) Sparsity and compressed coding in sensory systems. PLoS Comput Biol 10(8):e1003793
Bartsch U, Oriyakhel W, Kenna PF, Linke S, Richard G, Petrowitz B, Humphries P, Farrar GJ, Ader M (2008) Retinal cells integrate into the outer nuclear layer and differentiate into mature photoreceptors after subretinal transplantation into adult mice. Exp Eye Res 86(4):691–700
Felsen G, Dan Y (2005) A natural approach to studying vision. Nat Neurosci 8(12):1643
Felsen G, Touryan J, Han F et al (2005) Cortical sensitivity to visual features in natural scenes. PLoS Biol 3(10):e342
Field DJ (1994) What is the goal of sensory coding? Neural Comput 6(4):559–601
Gravier A, Quek C, Duch W, Wahab A, Gravier-Rymaszewska J (2016) Neural network modelling of the influence of channelopathies on reflex visual attention. Cogn Neurodyn 10(1):49–72
Gross CG (1994) How inferior temporal cortex became a visual area. Cereb Cortex 4(5):455
Hadjinicolaou AE, Cloherty SL, Kameneva T et al (2016) Frequency responses of rat RGCs. PLoS ONE 11(6):e0157676
Hasenstaub A, Otte S, Callaway E et al (2010) Metabolic cost as a unifying principle governing neuronal biophysics. Proc Natl Acad Sci USA 107(27):12329
Hoyer PO, Hyvarinen A (2000) Independent component analysis applied to feature extraction from colour and stereo images. Netw Comput Neural Syst 11(3):191–210
Hubel DH, Wiesel TN (1997) Functional architecture of macaque monkey visual cortex. Proc R Soc Lond B 198:1–59
Huberman AD, Feller MB, Chapman B (2008) Mechanisms underlying development of visual maps and receptive fields. Annu Rev Neurosci 31:479–509
Hyvarinen A (1999) Survey on independent component analysis. Neural Comput Surv 2(4):94–128
Hyvärinen A (1999) Fast independent component analysis with noisy data using Gaussian moments. Proc Int Symp Circuits Syst 5:V57–V61
Hyvarinen A, Hoyer PO (2002) A two-layer sparse coding model learn simple and complex cell receptive fields and topography from natural images. Vis Res 41(18):2413–2423
Hyvarinen A, Hoyer PO, Mika OI (2001) Topographic independent component analysis. Neural Comput 13(7):1527–1558
Jessell Thomas M, Kandel Eric R, Schwartz JH (2000) Principles of neural science, 5th edn. McGraw-Hill, New York, pp 533–540
Kameneva T, Maturana MI, Hadjinicolaou AE et al (2016) RGCs: mechanisms underlying depolarization block and differential responses to high frequency electrical stimulation of ON and OFF cells. J Neural Eng 13(1):016017
Kandel E, Schwartz J (2013) Principles of neural science, 5th edn. McGraw-Hill, New York
Khoshbin-e-Khoshnazar MR (2014) Quantum superposition in the retina: evidences and proposals. NeuroQuantology 12(1):97–101
Laughlin SB (2001) Energy as a constraint on the coding and processing of sensory information. Curr Opin Neurobiol 11(4):475–480
Levy WB, Baxter RA (1999) Energy efficient neural codes Neural codes and distributed representations. MIT Press, Cambridge, pp 531–543
Lewick M (2002) Efficient coding of natural sounds. Nat Neurosci 5:356–363
Maturana MI, Kameneva T, Burkitt AN et al (2014) The effect of morphology upon electrophysiological responses of RGCs: simulation results. J Comput Neurosci 36(2):157–175
Maturana MI, Apollo NV, Hadjinicolaou AE et al (2016) A simple and accurate model to predict responses to multi-electrode stimulation in the retina. PLoS Comput Biol 12(4):e1004849
Mizraji E, Lin J (2017) The feeling of understanding: an exploration with neural models. Cogn Neurodyn 11:135–146
Momtaz HZ, Daliri MR (2016) Predicting the eye fixation locations in the gray scale images in the visual scenes with different semantic contents. Cogn Neurodyn 10(1):31–47
Olshausen BA, Field DJ (1996) Emergence of simple cell receptive properties by learning a sparse code for natural images. Nature 381:607–609
Olshausen BA, Field DJ (1997) Sparse coding with an over complete basis set: a strategy employed by V1. Vision Res 37:3313–3325
Olshausen BA, Field DJ (2004) Sparse coding of sensory inputs. Curr Opin Neurobiol 14:481–487
Peters JF, Tozzi A, Ramanna S et al (2017) The human brain from above: an increase in complexity from environmental stimuli to abstractions. Cogn Neurodyn 11(1):1–4
Pillow JW, Shlens J, Paninski L, Sher A, Litke AM (2008) Spatio-temporal correlations and visual signaling in a complete neuronal population. Nature 454:995–999
Protopapa F, Siettos CI, Myatchin I, Lagae L (2016) Children with well controlled epilepsy possess different spatio-temporal patterns of causal network connectivity during a visual working memory task. Cogn Neurodyn 10(2):99–111
Qiu XW, Gong HQ, Zhang PM et al (2016) The oscillation-like activity in bullfrog ON-OFF retinal ganglion cell. Cogn Neurodyn 10(6):481
Qureshi TA, Hunter A, Al-Diri B (2014) A Bayesian framework for the local configuration of retinal junctions. IEEE Comput Vis Pattern Recogn 167:3105–3110
Reich LN, Bedell HE (2000) Relative legibility and confusions of letter acuity targets in the peripheral and central retina. Optom Vis Sci Off Publ Am Acad Optom 77(5):270–275
Rieke F, Warland D, van Steveninck RR et al (1997) Spikes: exploring the neural code. MIT, Cambridge
Schiller PH (1986) The central visual system. Vision Res 26(9):1351
Simoncelli EP (2003) Vision and the statistics of the visual environment. Curr Opin Neurobiol 1(13):144–149
Simoncelli EP, Olshausen BA (2001) Natural image statistics and neural representation. Annu Rev Neurosci 24:1193–1216
Theunissen FE, David SV, Singh NC et al (2001) Estimating spatio-temporal receptive fields of auditory and visual neurons from their responses to natural stimuli. Network 12(3):289
Touryan J, Felsen G, Dan Y (2005) Spatial structure of complex cell receptive fields measured with natural images. Neuron 45(5):781
Tozzi A, Peters JF (2017) From abstract topology to real thermodynamic brain activity. Cogn Neurodyn 11(3):283
Treichler DG (1967) Are you missing the boat in training aids? Film Audio-Visual Commun 1:14–16
Treves A, Rolls ET (1991) What determines the capacity of auto associative memories in the brain? Network 2:371–397
Urakawa T, Bunya M, Araki O (2017) Involvement of the visual change detection process in facilitating perceptual alternation in the bistable image. Cogn Neurodyn 11(9):1–12
Vinje W, Gallant J (2002) Natural stimulation of the non-classical receptive field increases information transmission efficiency in V1. J Neurosci 22:2904–2915
Wang G, Wang R (2017) Sparse coding network model based on fast independent component analysis. Neural Comput Appl 13:1–7
Wang RB, Zhu YT (2016) Can the activities of the large scale cortical network be expressed by neural energy? Cogn Neurodyn 10(1):1–5
Wang YH, Wang RB, Zhu YT (2017) Optimal path-finding through mental exploration based on neural energy field gradients. Cogn Neurodyn 11(1):99–111
Willmore B, Tolhurst DJ (2001) Characterizing the sparseness of neural codes. Network. 12(3):255–270
Wohrer A, Kornprobst P (2009) Virtual Retina: a biological retina model and simulator, with contrast gain control. J Comput Neurosci 26(2):219–249
Yan RJ, Gong HQ, Zhang PM, He SG, Liang PJ (2016) Temporal properties of dual-peak responses of mouse RGCs and effects of inhibitory pathways. Cogn Neurodyn 10(3):211–223
Zhang YY, Jin X, Gong HQ, Liang PJ (2010) Temporal and spatial patterns of retinal ganglion cells in response to natural stimuli. Prog Biochem Biophys 37(4):389–396
Zheng HW, Wang RB, Qu JY (2016) Effect of different glucose supply conditions on neuronal energy metabolism. Cogn Neurodyn 10(6):1–9
Acknowledgements
This study was funded by the National Natural Science Foundation of China (Nos. 11232005, 11472104).
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
Sparse coding
Sparse coding includes a class of unsupervised methods for learning sets of complete bases for efficient data representation. The aim of sparse coding is to develop a set of basic vectors that represent an input vector as a linear combination of the basic vectors:
where \(x = (x_{1} ,x_{2} , \ldots ,x_{n} )^{T}\) represents input data, \(A = (a_{1} ,a_{2} , \ldots ,a_{m} )^{T}\) is base matrix, \(a_{i}\) is column i in \(A\), which represents the basic functions. \(S = (s_{1} ,s_{2} , \ldots ,s_{m} )^{T}\) denotes coefficient matrix. With a complete basis, \(S\) is no longer uniquely determined by the input vector \(x\). Therefore, we introduced the additional criterion of sparsity in sparse coding. We define sparsity in terms of few non-zero components or few components not close to zero. The choice of sparsity as a desired characteristic in our representation of the input data is motivated by the observation that most sensory data such as natural images may be described as the superposition of a small number of atomic elements such as surfaces or edges. Other justifications such as comparisons of the properties of the primary visual cortex have also been advanced.
We define the sparse coding cost function using a set of n input vectors as follows:
where \(a_{j}\) represents basic function, \(s_{j}\) is coefficient, \(x_{i}\) is input data, \(\lambda\) is a constant, \(H(s_{j} )\) denotes a sparsity cost function, which penalizes \(s_{j}\) for being far from zero. Usually a common choice for the sparsity cost is the L1 penalty \(H(s_{j} ) = \left| {s_{j} } \right|_{1}\), but it is non-differentiable when basic function equals 0, therefore, we selected sparsity cost \(H(s_{j} ) = \sqrt {s_{j}^{2} + \varepsilon }\), wherein \(\varepsilon\) is a constant.
We interpret the first term of the sparse coding objective as a reconstruction term, which uses the algorithm to provide a good representation of x and the second term as a sparsity penalty, which is a sparce representation of \(x\). The constant \(\lambda\) is a scaling constant determining the relative importance of these two contributions.
In addition, it is possible to make the sparsity penalty arbitrarily small by scaling down \(s_{j}\) and scaling \(a_{j}\) up using a large constant. To prevent this event, we constrain \(\left\| {a_{j} } \right\|^{2} \le C,\quad \forall j = 1,2, \ldots m\) to be less than the constant C.
The full sparse coding cost function including our constraint is as follows:
However, the constraint of \(\left\| {a_{j} } \right\|^{2} \le C,\quad \forall j = 1,2, \ldots m\) cannot be enforced using simple gradient-based methods. This constraint is weakened to a “weight decay” term designed to keep the entries of \(A\) small. Therefore, we added the constraints to the objective function to provide a new objective function:
where \(X\) is input data,\(A = (a_{1} ,a_{2} , \ldots a_{m} )^{T}\) is base matrix, \(S = (s_{1} ,s_{2} , \ldots s_{m} )^{T}\) is coefficient matrix, \(\lambda\) and \(\gamma\) are constants.
The objective function is non-convex, and hence impossible to optimize well using gradient-based methods. However, given \(A\), the problem of finding \(S\) that minimizes \(F(A,S)\) is convex. Similarly, given \(S\), the problem of finding \(A\) that minimizes \(F(A,S)\) is also convex suggesting an alternative to optimize \(A\) for a fixed \(S\), and then optimizing \(S\) with a fixed \(A\).
The analytic solution of \(A\) is obtained as follows:
The analytic solution of \({\text{S}}\) is provided by:
Therefore, the learning equation of basic function \(a_{i}\) is represented by:
The learning equation of coefficient \(s_{i}\) is as follows:
Using the simple iterative algorithm on a large dataset (including 10,000 patches) results in prolonged iterations and convergence of the algorithm. To increase the rate of convergence by accelerating the iteration, the algorithm may be run on mini-patch selecting a mini-patch random subset of 1000 patches from the 10,000 patches.
A faster and better convergence may be obtained via initialization of the feature matrix \(S\) before using gradient descent (or other methods) to optimize the objective function for \(S\) given \(A\). In practice, initializing \(S\) randomly at each iteration results in poor convergence unless a good optimum is found for \(S\) before optimizing for \(A\). A better way to initialize \(S\) involves the following steps:
-
1.
Random initialization of \(A\)
-
2.
Repetition until convergence
-
1.
Selection of a mini-patch random subset.
-
2.
Initialization of \(S\) with \(S = A^{T} X\), dividing the feature by the the corresponding basic vector in \(A\).
-
3.
Finding \(S\) that minimizes \(F(A,S)\) for the \(A\) in the previous step.
-
4.
Determination of \(A\) that minimizes \(F(A,S)\) for the \(S\) found in the previous step. Using this method, good local optima can be reached relatively quickly.
-
1.
We obtained the base matrix \(A\) trained by FASTICA. Using the FASTICA method, we derived the coefficient matrix and the objective function values through traditional sparse coding. We selected a convex function \(H(s_{j} ) = \sqrt {s_{j}^{2} + \varepsilon }\) for the sparsity cost in objective function, where \(\varepsilon\) = 0.01 and \(\lambda\) = 0.3. Since the base matrix \(\left\| A \right\|^{2}\) in objective function was obtained and normalized by FASTICA, \(\left\| A \right\|^{2}\) equals 1. \(\left\| A \right\|^{2}\) does not affect the optimization of the objective function, and therefore, \(\gamma\) = 0.
Rights and permissions
About this article
Cite this article
Wang, G., Wang, R., Kong, W. et al. Simulation of retinal ganglion cell response using fast independent component analysis. Cogn Neurodyn 12, 615–624 (2018). https://doi.org/10.1007/s11571-018-9490-4
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11571-018-9490-4