Abstract
Pattern separation is a fundamental brain computation that converts small differences in input patterns into large differences in output patterns. Several synaptic mechanisms of pattern separation have been proposed, including code expansion, inhibition and plasticity; however, which of these mechanisms play a role in the entorhinal cortex (EC)–dentate gyrus (DG)–CA3 circuit, a classical pattern separation circuit, remains unclear. Here we show that a biologically realistic, full-scale EC–DG–CA3 circuit model, including granule cells (GCs) and parvalbumin-positive inhibitory interneurons (PV+-INs) in the DG, is an efficient pattern separator. Both external gamma-modulated inhibition and internal lateral inhibition mediated by PV+-INs substantially contributed to pattern separation. Both local connectivity and fast signaling at GC–PV+-IN synapses were important for maximum effectiveness. Similarly, mossy fiber synapses with conditional detonator properties contributed to pattern separation. By contrast, perforant path synapses with Hebbian synaptic plasticity and direct EC–CA3 connection shifted the network towards pattern completion. Our results demonstrate that the specific properties of cells and synapses optimize higher-order computations in biological networks and might be useful to improve the deep learning capabilities of technical networks.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$99.00 per year
only $8.25 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
Output datasets can be regenerated from the code76. As the full output dataset generated in this work is huge (>10 Tb), deposit in a publicly available repository is not practical at the current time point. Specific data will be provided by the corresponding author on request . Source data are provided with this paper.
Code availability
References
Yassa, M. A. & Stark, C. E. Pattern separation in the hippocampus. Trends Neurosci. 34, 515–525 (2011).
Rolls, E. T. Pattern separation, completion, and categorisation in the hippocampus and neocortex. Neurobiol. Learn. Mem. 129, 4–28 (2016).
Chavlis, S. & Poirazi, P. Pattern separation in the hippocampus through the eyes of computational modeling. Synapse 71, e21972 (2017).
Cayco-Gajic, N. A. & Silver, R. A. Re-evaluating circuit mechanisms underlying pattern separation. Neuron 101, 584–602 (2019).
Leutgeb, J. K., Leutgeb, S., Moser, M. B. & Moser, E. I. Pattern separation in the dentate gyrus and CA3 of the hippocampus. Science 315, 961–966 (2007).
Scharfman, H. E. The dentate gyrus: A comprehensive guide to structure, function, and clinical implications. Progress Brain Res. 163, 627–637 (2007).
Bischofberger, J., Engel, D., Frotscher, M. & Jonas, P. Timing and efficacy of transmitter release at mossy fiber synapses in the hippocampal network. Pflügers Arch. 453, 361–372 (2006).
Guzman, S. J., Schlögl, A., Frotscher, M. & Jonas, P. Synaptic mechanisms of pattern completion in the hippocampal CA3 network. Science 353, 1117–1123 (2016).
Marr, D. A theory of cerebellar cortex. J. Physiol. 202, 437–470 (1969).
Albus, J. S. A theory of cerebellar function. Math. Biosci. 10, 25–61 (1971).
Amaral, D. G., Ishizuka, N. & Claiborne, B. Neurons, numbers and the hippocampal network. Prog. Brain Res. 83, 1–11 (1990).
Boss, B. D., Turlejski, K., Stanfield, B. B. & Cowan, W. M. On the numbers of neurons in fields CA1 and CA3 of the hippocampus of Sprague-Dawley and Wistar rats. Brain Res. 406, 280–287 (1987).
Amrein, I., Slomianka, L. & Lipp, H. P. Granule cell number, cell death and cell proliferation in the dentate gyrus of wild-living rodents. European J. Neurosci. 20, 3342–3350 (2004).
Coultrip, R., Granger, R. & Lynch, G. A cortical model of winner-take-all competition via lateral inhibition. Neural Netw. 5, 47–54 (1992).
Wiechert, M. T., Judkewitz, B., Riecke, H. & Friedrich, R. W. Mechanisms of pattern decorrelation by recurrent neuronal circuits. Nat. Neurosci. 13, 1003–1010 (2010).
Papadopoulou, M., Cassenaer, S., Nowotny, T. & Laurent, G. Normalization for sparse encoding of odors by a wide-field interneuron. Science 332, 721–725 (2011).
Lin, A. C., Bygrave, A. M., de Calignon, A., Lee, T. & Miesenböck, G. Sparse, decorrelated odor coding in the mushroom body enhances learned odor discrimination. Nat. Neurosci. 17, 559–568 (2014).
Maass, W. On the computational power of winner-take-all. Neural Comput. 12, 2519–2535 (2000).
de Almeida, L., Idiart, M. & Lisman, J. E. A second function of gamma frequency oscillations: an E%-max winner-take-all mechanism selects which cells fire. J. Neurosci. 29, 7497–7503 (2009).
Tetzlaff, T., Helias, M., Einevoll, G. T. & Diesmann, M. Decorrelation of neural-network activity by inhibitory feedback. PLoS Comput. Biol. 8, e1002596 (2012).
Geiger, J. R. P., Lübke, J., Roth, A., Frotscher, M. & Jonas, P. Submillisecond AMPA receptor-mediated signaling at a principal neuron-interneuron synapse. Neuron 18, 1009–1023 (1997).
Espinoza, C., Guzman, S. J., Zhang, X. & Jonas, P. Parvalbumin+ interneurons obey unique connectivity rules and establish a powerful lateral-inhibition microcircuit in dentate gyrus. Nat. Commun. 9, 4605 (2018).
O'Reilly, R. C. & McClelland, J. L. Hippocampal conjunctive encoding, storage, and recall: avoiding a trade-off. Hippocampus 4, 661–682 (1994).
Neunuebel, J. P. & Knierim, J. J. CA3 retrieves coherent representations from degraded input: direct evidence for CA3 pattern completion and dentate gyrus pattern separation. Neuron 81, 416–427 (2014).
Vyleta, N. P., Borges-Merjane, C. & Jonas, P. Plasticity-dependent, full detonation at hippocampal mossy fiber–CA3 pyramidal neuron synapses. eLife 5, e17977 (2016).
Cayco-Gajic, N. A., Clopath, C. & Silver, R. A. Sparse synaptic connectivity is required for decorrelation and pattern separation in feedforward networks. Nat. Commun. 8, 1116 (2017).
Witter, M. P. The perforant path: projections from the entorhinal cortex to the dentate gyrus. Prog. Brain Res. 163, 43–61 (2007).
Bliss, T. V. P. & Lømo, T. Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J. Physiol. 232, 331–356 (1973).
McNaughton, B. L., Douglas, R. M. & Goddard, G. V. Synaptic enhancement in fascia dentata: cooperativity among coactive afferents. Brain Res. 157, 277–293 (1978).
McHugh, T. J. et al. Dentate gyrus NMDA receptors mediate rapid pattern separation in the hippocampal network. Science 317, 94–99 (2007).
McNaughton, B. L. & Morris, R. G. M. Hippocampal synaptic enhancement and information storage within a distributed memory system. Trends Neurosci. 10, 408–415 (1987).
Steward, O. Topographic organization of the projections from the entorhinal area to the hippocampal formation of the rat. J. Comp. Neurol. 167, 285–314 (1976).
Zhang, X., Schlögl, A. & Jonas, P. Selective routing of spatial information flow from input to output in hippocampal granule cells. Neuron 107, 1212–1225 (2020).
Valiant, L. G. The hippocampus as a stable memory allocator for cortex. Neural Comput. 24, 2873–2899 (2012).
Dasgupta, S., Stevens, C. F. & Navlakha, S. A neural algorithm for a fundamental computing problem. Science 358, 793–796 (2017).
Sharma J. & Navlakha, S. Improving similarity search with high-dimensional locality-sensitive hashing. Preprint at https://arxiv.org/abs/1812.01844 (2018).
Bartos, M. et al. Fast synaptic inhibition promotes synchronized gamma oscillations in hippocampal interneuron networks. Proc. Natl Acad. Sci. USA 99, 13222–13227 (2002).
Claiborne, B. J., Amaral, D. G. & Cowan, W. M. A light and electron microscopic analysis of the mossy fibers of the rat dentate gyrus. J. Comp. Neurol. 246, 435–458 (1986).
Henze, D. A., Wittner, L. & Buzsáki, G. Single granule cells reliably discharge targets in the hippocampal CA3 network in vivo. Nat. Neurosci. 5, 790–795 (2002).
Vandael, D., Borges-Merjane, C., Zhang, X. & Jonas, P. Short-term plasticity at hippocampal mossy fiber synapses is induced by natural activity patterns and associated with vesicle pool engram formation. Neuron 107, 509–521 (2020).
Bragin, A. et al. Gamma (40–100 Hz) oscillation in the hippocampus of the behaving rat. J. Neurosci. 15, 47–60 (1995).
Pernía-Andrade, A. J. & Jonas, P. Theta-gamma-modulated synaptic currents in hippocampal granule cells in vivo define a mechanism for network oscillations. Neuron 81, 140–152 (2014).
Majani, E., Erlanson, R. & Abu-Mostafa, Y. On the k-winners takes-all network. Adv. Neural Inf. Process. Syst. 1, 634–642 (1989).
Ellias, S. A. & Grossberg, S. Pattern formation, contrast control, and oscillations in the short term memory of shunting on-center off-surround networks. Biol. Cybern. 20, 69–98 (1975).
Tamamaki, N. & Nojyo, Y. Projection of the entorhinal layer II neurons in the rat as revealed by intracellular pressure-injection of neurobiotin. Hippocampus 3, 471–480 (1993).
Hu, H., Gan, J. & Jonas, P. Fast-spiking, parvalbumin+ GABAergic interneurons: from cellular design to microcircuit function. Science 345, 1255263 (2014).
Nörenberg, A., Hu, H., Vida, I., Bartos, M. & Jonas, P. Distinct nonuniform cable properties optimize rapid and efficient activation of fast-spiking GABAergic interneurons. Proc. Natl Acad. Sci. USA 107, 894–899 (2010).
Kraushaar, U. & Jonas, P. Efficacy and stability of quantal GABA release at a hippocampal interneuron-principal neuron synapse. J. Neurosci. 20, 5594–5607 (2000).
Chamberland, S., Timofeeva, Y., Evstratova, A., Volynski, K. & Tóth, K. Action potential counting at giant mossy fiber terminals gates information transfer in the hippocampus. Proc. Natl Acad. Sci. USA 115, 7434–7439 (2018).
Toth, K., Suares, G., Lawrence, J. J., Philips-Tansey, E. & McBain, C. J. Differential mechanisms of transmission at three types of mossy fiber synapse. J. Neurosci. 20, 8279–8289 (2000).
LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436–444 (2015).
Babadi, B. & Sompolinsky, H. Sparseness and expansion in sensory representations. Neuron 83, 1213–1226 (2014).
de la Rocha, J., Doiron, B., Shea-Brown, E., Josić, K. & Reyes, A. Correlation between neural spike trains increases with firing rate. Nature 448, 802–806 (2007).
Hoeffding, W. Masstabinvariante Korrelationsstheorie. Schriften Math. Instituts Angew. Math. Univ. Berlin 5, 179–233 (1940).
Kowalski, J., Gan, J., Jonas, P. & Pernía-Andrade, A. J. Intrinsic membrane properties determine hippocampal differential firing pattern in vivo in anesthetized rats. Hippocampus 26, 668–682 (2016).
Engin, E. et al. Tonic inhibitory control of dentate gyrus granule cells by α5-containing GABAA receptors reduces memory interference. J. Neurosci. 35, 13698–13712 (2015).
Espinoza Martinez, C. M. Parvalbumin+ Interneurons Enable Efficient Pattern Separation in Hippocampal Microcircuits (IST Austria, 2019); https://doi.org/10.15479/AT:ISTA:6363
Braganza, O., Mueller-Komorowska, D., Kelly, T. & Beck, H. Quantitative properties of a feedback circuit predict frequency-dependent pattern separation. eLife 9, e53148 (2020).
Bartos, M., Vida, I., Frotscher, M., Geiger, J. R. P. & Jonas, P. Rapid signaling at inhibitory synapses in a dentate gyrus interneuron network. J. Neurosci. 21, 2687–2698 (2001).
Hu, H. & Jonas, P. A supercritical density of Na+ channels ensures fast signaling in GABAergic interneuron axons. Nat. Neurosci. 17, 686–693 (2014).
Bucurenciu, I., Kulik, A., Schwaller, B., Frotscher, M. & Jonas, P. Nanodomain coupling between Ca2+ channels and Ca2+ sensors promotes fast and efficient transmitter release at a cortical GABAergic synapse. Neuron 57, 536–545 (2008).
Jones, B. W. et al. Targeted deletion of AKAP7 in dentate granule cells impairs spatial discrimination. eLife 5, e20695 (2016).
Pehlevan, C., Sengupta, A. M. & Chklovskii, D. B. Why do similarity matching objectives lead to Hebbian/Anti-Hebbian networks? Neural Comput. 30, 84–124 (2018).
Myers, C. E. & Scharfman, H. E. A role for hilar cells in pattern separation in the dentate gyrus: a computational approach. Hippocampus 19, 321–337 (2009).
Johnston, S. T., Shtrahman, M., Parylak, S., Gonçalves, J. T. & Gage, F. H. Paradox of pattern separation and adult neurogenesis: A dual role for new neurons balancing memory resolution and robustness. Neurobiol. Learn. Mem. 129, 60–68 (2016).
Schneider, C. J., Bezaire, M. & Soltesz, I. Toward a full-scale computational model of the rat dentate gyrus. Front. Neural Circuits 6, 83 (2012).
Wang, X. J. & Buzsáki, G. Gamma oscillation by synaptic inhibition in a hippocampal interneuronal network model. J. Neurosci.16, 6402–6413 (1996).
Ermentrout, B. Type I membranes, phase resetting curves, and synchrony. Neural Comput. 8, 979–1001 (1996).
Carnevale, N. T. & Hines, M. L. The Neuron Book (Cambridge Univ. Press, 2006).
Schmidt-Hieber, C., Jonas, P. & Bischofberger, J. Subthreshold dendritic signal processing and coincidence detection in dentate gyrus granule cells. J. Neurosci. 27, 8430–8441 (2007).
Paxinos, G. & Franklin, K. The Mouse Brain in Stereotaxic Coordinates 4th edn (Academic, 2012).
Han, Z. S., Buhl, E. H., Lörinczi, Z. & Somogyi, P. A high degree of spatial selectivity in the axonal and dendritic domains of physiologically identified local-circuit neurons in the dentate gyrus of the rat hippocampus. European J. Neurosci. 5, 395–410 (1993).
Hefft, S. & Jonas, P. Asynchronous GABA release generates long-lasting inhibition at a hippocampal interneuron-principal neuron synapse. Nat. Neurosci. 8, 1319–1328 (2005).
Hosp, J. A. et al. Morpho-physiological criteria divide dentate gyrus interneurons into classes. Hippocampus 24, 189–203 (2014).
Armstrong, C. & Soltesz, I. Basket cell dichotomy in microcircuit function. J. Physiol. 590, 683–694 (2012).
Guzman, S. J. et al. Pattern Separation Network (IST Austria, 2021); https://doi.org/10.15479/AT:ISTA:10110
Acknowledgements
We thank A. Aertsen, N. Kopell, W. Maass, A. Roth, F. Stella and T. Vogels for critically reading earlier versions of the manuscript. We are grateful to F. Marr and C. Altmutter for excellent technical assistance, E. Kralli-Beller for manuscript editing, and the Scientific Service Units of IST Austria for efficient support. Finally, we thank T. Carnevale, L. Erdös, M. Hines, D. Nykamp and D. Schröder for useful discussions, and R. Friedrich and S. Wiechert for sharing unpublished data. This project received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 692692, P.J.) and the Fond zur Förderung der Wissenschaftlichen Forschung (Z 312-B27, Wittgenstein award to P.J. and P 31815 to S.J.G.).
Author information
Authors and Affiliations
Contributions
P.J. and S.J.G. designed the model and the layout of the simulations. P.J. and A.S. performed large-scale simulations on computer clusters. C.E., X.Z. and B.A.S. provided experimental data. P.J. and S.J.G. analyzed the data. P.J. wrote the paper and all authors jointly revised it.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information Nature Computational Science thanks Ad Aertsen, Alessandro Treves and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Handling editor: Ananya Rastogi, in collaboration with the Nature Computational Science team.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Quantitative analysis of pattern separation in neuronal networks.
a, b, Schematic illustration of pattern separation. (a) Neuronal activity at the input (top) and the output level (bottom) during two similar contexts (top). Red, cells active in pattern A; green, cells active in pattern B; yellow, cells active in both patterns. (b) Overlay of neuronal activity at the input (top) and the output level (bottom). Highly overlapping input patterns (A, B; top) are converted into weakly overlapping output patterns (A′, B′; bottom). Modified from Johnston et al., 2016 (ref. 65). c, d, Analysis of pattern separation and pattern completion in input-output correlation plots (Rout–Rin graphs). Rin and Rout represent pairwise correlations in input and output patterns. Red dashed line indicates pattern identity. Area below identity line (red and green stripes, c) represents a regime in which Rout < Rin, that is, pattern separation. Area above identity line (yellow area, d) corresponds to a regime where Rout > Rin, that is, pattern completion. Insets, Venn diagrams of two patterns before and after pattern separation (c) and pattern completion (d). e, f, Quantitative analysis of Rout–Rin graphs. Data points (black points) represent output and input correlations for all pairs of patterns; 4950 data points total. An integral-based metric, ψ, provides a robust assessment of the average pattern separation behavior (e, main panel). ψ was computed as the area between identity line (IL, red dashed line) and the interpolated Rout–Rin curve (light gray area), normalized to the maximum area (0.5). A slope-based measure, γ, provides a selective analysis of pattern separation in a region of interest in which differences between input patterns are small (e, inset). γ was computed as the slope of the Rout–Rin curve for Rin → 1. A rank correlation-based measure, ρ, provides an analysis of the ability of the network to preserve rank order similarity (f). ρ was computed as the Pearson’s correlation coefficient of the ranks of all Rout versus the ranks of all Rin data points. Rout–Rin plot and rank correlation plots are shown for standard model parameters (same data as in Fig. 1c, f; see Supplementary Table 1).
Supplementary information
Supplementary Information
Supplementary Figs. 1–9 and Table 1.
Supplementary Software 1
Zipped files for example simulations.
Source data
Source Data Fig. 1
Original values Rout versus Rin plot to obtain Psi and Gamma, rank correlation plot to obtain Rho.
Source Data Fig. 2
Original values. Fig. 2b: Activity, Psi, Gamma, and Rho as a function of Iμ. Figs. 2c,d: Psi as a function of Iμ and Jgamma with LI and without LI. Fig. 2e–g: Psi for different cEI, cIE, sigmaEI, sigmaIE, JEI and JIE.
Source Data Fig. 3
Original values. Fig. 3c: Psi as a function of nEC and nGC. Fig. 3d: Psi as a function of nEC:nGC ratio. Fig. 3f,g: Psi for different nEC, alphaEC, cEC-GC and sigmaEC-GC.
Source Data Fig. 4
Original values. Fig. 4b: Psi for different sigmaEI and sigmaIE. Fig. 4c: Psi as a function of sigmaEI and sigmaIE. Fig. 4d: Distribution of delay E-I and delay I-E. Fig. 4e,f: Psi for different deltasynE, deltasynI, tauE and taum.
Source Data Fig. 5
Original values. Fig. 5d: Psi as a function of number of MFBs. Fig. 5f: Psi as a function of MFB synaptic strength.
Source Data Fig. 6
Original values. Fig. 6c: Psi as a function of LTP at EC–GC synapses. Fig. 6f: Psi as a function of Imu EC–CA3.
Source Data Extended Data Fig. 1
Original values for Rout versus Rin plot to obtain Psi and Gamma, rank correlation plot to obtain Rho.
Rights and permissions
About this article
Cite this article
Guzman, S.J., Schlögl, A., Espinoza, C. et al. How connectivity rules and synaptic properties shape the efficacy of pattern separation in the entorhinal cortex–dentate gyrus–CA3 network. Nat Comput Sci 1, 830–842 (2021). https://doi.org/10.1038/s43588-021-00157-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s43588-021-00157-1
This article is cited by
-
Assessments of dentate gyrus function: discoveries and debates
Nature Reviews Neuroscience (2023)
-
Phase information is conserved in sparse, synchronous population-rate-codes via phase-to-rate recoding
Nature Communications (2023)
-
Insights into hippocampal network function
Nature Computational Science (2021)