Abstract
An important function of the brain is to predict which stimulus is likely to occur based on the perceived cues. The present research studied the branching behavior of a computational network model of populations of excitatory and inhibitory neurons, both analytically and through simulations. Results show how synaptic efficacy, retroactive inhibition and short-term synaptic depression determine the dynamics of selection between different branches predicting sequences of stimuli of different probabilities. Further results show that changes in the probability of the different predictions depend on variations of neuronal gain. Such variations allow the network to optimize the probability of its predictions to changing probabilities of the sequences without changing synaptic efficacy.
Similar content being viewed by others
Availability of data and material
Not applicable.
Code availability
Available at https://github.com/elifkoksal/dynamicBranching
Notes
This is the simplest way to form graphs of overlapping patterns. Preliminary results seem to indicate that the model can catch latching behavior in more general situations (overlaps with variable size). This is forthcoming work.
References
Aguilar, C., Chossat, P., Krupa, M., et al. (2017). Latching dynamics in neural networks with synaptic depression. PLoS ONE, 12(8), e0183710.
Albrengues, C., Lavigne, F., Aguilar, C., et al. (2019). Linguistic processes do not beat visuo-motor constraints, but they modulate where the eyes move regardless of word boundaries: Evidence against top-down word-based eye-movement control during reading. PLoS ONE, 14(7), e0219666.
Amari, S. (1972). Characteristics of random nets of analog neuron-like elements. IEEE Transactions on Systems, Man, and Cybernetics, 5, 643–57.
Amit, D. J., & Brunel, N. (1997). Model of global spontaneous activity and local structured activity during delay periods in the cerebral cortex. Cerebral Cortex, 7(3), 237–52.
Amit, D. J., Brunel, N., & Tsodyks, M. V. (1994). Correlations of cortical hebbian reverberations: Theory versus experiment. Journal of Neuroscience, 14, 6435–45.
Amit, D. J., Bernacchia, A., & Yakovlev, V. (2003). Multiple-object working memory-a model for behavioral performance. Cerebral Cortex, 13(5), 435–43.
Bak, P., & Chialvo, D. R. (2001). Adaptive learning by extrernal dynamics and negative feedback. Physical Review E, 63(3), 031912.
Bak, P., & Paczuski, M. (1995). Complexity, contingency and criticality. Proceedings of the National Academy of Sciences, 92(15), 6669–96.
Bastos, A. M., Lundqvist, M., Waite, A. S., et al. (2020). Layer and rhythm specificity for predictive routing. Proceedings of the National Academy of Sciences, 117(49), 31459–69.
Bell, A. H., Summerfield, C., Morin, E. L., et al. (2016). Encoding of stimulus probability in macaque inferior temporal cortex. Current Biology, 26(17), 2280–90.
Bienenstock, E., & Lehmann, D. (1998). Regulated criticality in the brain? Advances in Complex Systems, 1(4), 361–84.
Bliss, T. V., & Collingridge, G. L. (1993). A synaptic model of memory: long-term potentiation in the hippocampus. Nature, 361, 31–39.
Bliss, T. V., & Lomo, T. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. Journal of Physiology, 232, 331–56.
Brunel, N. (1996). Hebbian learning of context in recurrent neural networks. Neural Computation, 15(8), 1677–710.
Brunel, N., & Lavigne, F. (2009). Semantic priming in a cortical network model. Journal of Cognitive Neuroscience, 21(2300-19).
Brunel, N., & Wang, X. J. (2001). Effects of neuromodulation in a cortical network model of object working memory dominated by recurrent inhibition. Journal of Computational Neuroscience, 11(1), 63–85.
Bunge, S. A., Kahn, I., Wallis, J. D., et al. (2003). Neural circuits subserving the retrieval and maintenance of abstract rules. Journal of Neuophysiology, 90(5), 3419–28.
Busemeyer, J. R., & Townsend, J. T. (1993). Decision field theory: a dynamic-cognitive approach to decision making in an uncertain environment. Psychological Review, 100(3), 432.
Chen, B., & Miller, P. (2020). Attractor-state itinerancy in neural circuits with synaptic depression. Journal of Mathematical Neuroscience, 10(1), 1–19.
Chialvo, D. R. (2010). Emergent complex neural dynamics. Nature Physics, 6(10), 744–50.
Cocchi, L., Gollo, L. L., Zalesky, A., et al. (2017). Criticality in the brain: A synthesis of neurobiology, models and cognition. Progress in Neurobiology, 158, 132–52.
Dasgupta, I., Schulz, E., Tenenbaum, J. B., et al. (2020). A theory of learning to infer. Psychological Review, 127(3), 412.
Dehghani, N., Peyrache, A., Telenczuk, B., et al. (2016). Dynamic balance of excitation and inhibition in human and monkey neocortex. Scientific Reports, 6(1), 23176.
Delaney-Busch, N., Morgan, E., Lau, E. F., et al. (2017). Comprehenders rationally adapt semantic predictions to the statistics of the local environment: A Bayesian model of trial-by-trial N400 amplitudes. In 39th Annual Conference of the Cognitive Science Society. London, England.
DeLong, K. A., Urbach, T. P., & Kutas, M. (2005). Probabilistic word pre-activation during language comprehension inferred from electrical brain activity. Nature Neuroscience, 8(8), 1117–21.
Ding, N., Melloni, L., Zhang, H., et al. (2015). ortical tracking of hierarchical linguistic structures in connected speech. Nature Neuroscience, 19(1), 158–64.
Erickson, C. A., & Desimone, R. (1999). Responses of macaque perirhinal neurons during and after visual stimulus association learning. Journal of Neuroscience, 19(10404-16).
FitzGerald, T., Dolan, R. J., & Friston, K. J. (2015). Dopamine, reward learning, and active inference. Frontiers in Computational Neuroscience, 9, 136.
Friston, K. J., Shiner, T., FitzGerald, T., et al. (2012). Dopamine, affordance and active inference. PLoS Computational Biology, 8(1), e1002327.
Fujimichi, R., Naya, Y., Koyano, K. W., et al. (2010). Unitized representation of paired objects in area 35 of the macaque perirhinal cortex. Europen Journal of Neuroscience, 32(4), 659–67.
Fuster, J. M., & Alexander, G. E. (1971). Neuron activity related to short-term memory. Science, 173(3997), 652–4.
Gershman, S. J. (2019). How to never be wrong. Psychonomic Bulletin and Review, 26(1), 13–28.
Gershman, S. J., & Uchida, N. (2019). Believing in dopamine. Nature Review. Neuroscience, 20(11), 703–14.
Gochin, P. M., Colombo, M., Dorfman, G. A., et al. (1994). Neural ensemble coding in inferior temporal cortex. Journal of Neuophysiology, 71, 2325–37.
Hahnloser, R. H., Kozhevnikov, A. A., & Fee, M. S. (2002). An ultra-sparse code underliesthe generation of neural sequences in a songbird. Nature, 419(6902), 65–70.
Harnal, H., & Giraud, A. L. (2012). Cortical oscillations and sensory predictions. Trends in Cognitive Science, 16, 390–8.
Harvey, C. D., Coen, P., & Tank, D. W. (2012). Choice-specific sequences in parietal cortex during a virtual-navigation decision task. Nature, 484(7392), 62–68.
Hebb, D. (1949). The Organization of Behavior: A Neuropsychological Theory. New York, NY: Wiley and Sons.
Hopfield, J. J. (1982). Neural networks and physical systems with emergent collective computational abilities. Proceedings of the National Academy of Sciences, 79(8), 2554–58.
Hung, C., Kreiman, G., Poggio, J., & DiCarlo, T. (2005). Fast read-out of object information in inferior temporal cortex. Science, 310, 863–6.
Hutchison, K., Heap, S., Neely, J., et al. (2014). Attentional control and asymmetric associative priming. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40(3), 844–56.
Ison, M. J., Quian, Q. R., & Fried, I. (2015). Rapid encoding of new memories by individual neurons in the human brain. Neuron, 87(1), 220–30.
Kamiński, J., Sullivan, S., Chung, J. M., et al. (2017). Persistently active neurons in human medial frontal and medial temporal lobe support working memory. Nature neuroscience, 20(4), 590–601.
Kang, C. J., & Treves, A. (2019). The challenge of taming a latching network near criticality. The Functional Role of Critical Dynamics in Neural Systems (vol. 11, p. 81–94). Springer Series on Bio- and Neurosystems.
Kirkwood, A., & Bear, M. F. (1994). Homosynaptic long-term depression in the visual cortex. Neuroscience, 14, 3404–12.
Köksal Ersöz, E., Aguilar, C., Chossat, P., et al. (2020). Neuronal mechanisms for sequential activation of memory items: Dynamics and reliability. PLoS ONE, 15(4), e0231165.
Körding, K. P., & Wolpert, D. M. (2004). Bayesian integration in sensorimotor learning. Nature, 427(6971), 244–7.
Kornblith, S., Quian Quiroga, R., Koch, C., et al. (2017). Persistent single-neuron activity during working memory in the human medial temporal lobe. Current Biology, 27(7), 1026–1032. https://doi.org/10.1016/j.cub.2017.02.013
Kreiman, G., Hung, C. P., Kraskov, A., et al. (2006). Object selectivity of local field potentials and spikes in the macaque inferior temporal cortex. Neuron, 49, 433–45.
Kutas, M., DeLong, K. A., & Smith, N. J. (2011). A look around at what lies ahead: Prediction and predictability in language processing. In Predictions in the Brain: Using Our Past to Generate a Future (pp. 190–207). Oxford University Press.
Lam, N., Borduqui, T., Hallak, J., et al. (2021). Effects of altered excitation-inhibition balance on decision making in a cortical circuit model. Journal of Neuroscience. https://doi.org/10.1523/JNEUROSCI.1371-20.2021
Lau, E. F., Holcomb, P. J., & Kuperberg, G. R. (2013). Dissociating N400 effects of prediction from association in single-word contexts. Journal of Cognitive Neuroscience, 25(3), 484–502.
Lavigne, F. (2004). Aim networks: autoincursive memory networks for anticipation toward learned goals. International Journal of Computing Anticipatory Systems, 14, 196–214.
Lavigne, F., & Darmon, N. (2008). Dopaminergic neuromodulation of semantic priming in a cortical network model. Neuropsychologia, 46, 3074–87.
Lavigne, F., & Denis, S. (2002). Neural network modeling of learning of contextual constraints on adaptive anticipations. International Journal of Computing Anticipatory Systems, 12, 253–68.
Lavigne, F., Vitu, F., & d’Ydewalle, G. (2000). The influence of semantic context on initial eye landing sites in words. Acta Psychologica, 104(2), 191–214.
Lavigne, F., Dumercy, L., & Darmon, N. (2011). Determinants of multiple semantic priming: A meta-analysis and spike frequency adaptive model of a cortical network. Journal of Cognitive Neuroscience, 23(6), 1447–74.
Lavigne, F., Dumercy, L., Chanquoy, L., et al. (2012). Dynamics of the semantic priming shift: Behavioral experiments and cortical network model. Cogitive Neurodynamics, 6(6), 467–83.
Lavigne, F., Chanquoy, L., Dumercy, L., et al. (2013). Early dynamics of the semantic priming shift. Advances in Cognitive Psychology, 9(1), 1–14.
Lavigne, F., Avnaïm, F., & Dumercy, L. (2014). Inter-synaptic learning of combination rules in a cortical network model. Frontiers in Psychology, 5, 842.
Lazartigues, L., Mathy, F., & Lavigne, F. (2021). Statistical learning of unbalanced exclusive-or temporal sequences in humans. PLoS ONE, 16(2), e0246826.
Lerner, I., & Shriki, O. (2014). Internally and externally driven network transitions as a basis for automatic and strategic processes in semantic priming: theory and experimental validation. Frontiers in Psychology, 5(314), 00314.
Lerner, I., Bentin, S., & Shriki, O. (2012). Spreading activation in an attractor network with latching dynamics: automatic semantic priming revisited. Cognitive Science, 36, 1339–82.
Levina, A., & Herrmann, M. (2006). Dynamical synapses give rise to a power-law distribution of neuronal avalanches. In Advances in Neural Information Processing Systems (pp 771–78. 18). MIT Press, Cambridge, MA, USA.
Luka, B. J., & Van Petten, C. (2014). Prospective and retrospective semantic processing: Prediction, time, and relationship strength in event-related potentials. Brain and Language, 135, 115–29.
Magnasco, M. O., Piro, O., & Cecchi, G. A. (2009). Self-tuned critical anti-Hebbian networks. Physical Review Letters, 102(25), 258102.
Messinger, A., Squire, L., Zola, S. M., et al (2001). Neuronal representations of stimulus associations develop in the temporal lobe during learning. Proceedings of the National Academy of Sciences, 98(12239-44).
Miller, E. K. (1999). The prefrontal cortex: complex neural properties for complex behavior. Neuron, 22, 15–17.
Miller, K. D., & Fumarola, F. (2012). Mathematical equivalence of two common forms of firing rate models of neural networks. Neural Computation, 24(1), 25–31.
Minier, L., Fagot, J., & Rey, A. (2016). The temporal dynamics of regularity extraction in non-human primates. Cognitive Science, 40(4), 1019–30.
Miyashita, Y. (1988). Neuronal correlate of visual associative long-term memory in the primate temporal cortex. Nature, 335.
Miyashita, Y., & Chang, H. S. (1988). Neuronal correlate of pictorial short-term memory in the primate temporal cortex. Nature, 331, 68–70.
Mongillo, G., Amit, D. J., & Brunel, N. (2003). Retrospective and prospective persistent activity induced by hebbian learning in a recurrent cortical network. Europen Journal of Neuroscience, 18(7), 2011–24.
Mongillo, G., Rumpel, S., & Loewenstein, Y. (2018). Inhibitory connectivity defines the realm of excitatory plasticity. Nature Neuroscience, 21(10), 1463–70.
Muhammad, R., Wallis, J. D., & Miller, E. K. (2006). A comparison of abstract rules in the prefrontal cortex, premotor cortex, inferior temporal cortex and striatum. Journal of Cognitive Neuroscience, 18(6), 974–89.
Naya, Y., Yoshida, M., & Miyashita, Y. (2001). Backward spreading of memory-retrieval signal in the primate temporal cortex. Science, 291(5504), 661–64.
Naya, Y., Yoshida, M., Takeda, M., et al. (2003). Delay-period activities in two subdivisions of monkey inferotemporal cortex during pair association memory task. The European Journal of Neuroscience, 18, 2915–8.
Neely, J. H. (1991). Semantic priming effects in visual word recognition: A selective review of current findings and theories. In Basic Processes in Reading: Visual Word Recognition. Lawrence Erlbaum Associates, Inc. (pp. 264–336).
Pereira, U., & Brunel, N. (2020). Unsupervised learning of persistent and sequential activity. Frontiers in computational neuroscience, 13, 97.
Quian, Q. R. (2012). Concept cells: the building blocks of declarative memory functions. Nature Review Neuroscience, 13, 587–97.
Quian, Q. R. (2016). Neuronal codes for visual perception and memory. Neuropsychologia, 83, 227–41.
Quian, Q. R., & Kreiman, G. (2010). Measuring sparseness in the brain: comment on Bowers. Psychological Review, 117, 291–99.
Rainer, G., Rao, S. C., & Miller, E. K. (1999). Prospective coding for objects in primate prefrontal cortex. Journal of Neuroscience, 19, 5493–5505.
Reddy, L., Poncet, M., Self, M. W., et al. (2015). Learning of anticipatory responses in single neurons of the human medial temporal lobe. Nature communications, 6(1), 1–8.
Rolls, E. T., & Tovee, M. J. (1995). Sparseness of the neuronal representation of stimuli in the primate temporal visual cortex. Journal of Neuophysiology, 73(2), 713–26.
Rolls, E. T., Loh, M., Deco, G., et al. (2008). Computational models of schizophrenia and dopamine modulation in the prefrontal cortex. Nature Review Neuroscience, 9, 696.
Sakai, K., & Miyashita, Y. (1991). Neural organization for the long-term memory of paired associates. Nature, 354, 152–55.
Schaal, S., Mohajerian, P., & Ijspeert, A. (2007). Dynamics systems vs. optimal control-a unifying view. Progress in Brain Research, 165, 425–445.
Tamura, H., & Tanaka, K. (2001). Visual response properties of cells in the ventral and dorsal parts of the macaque inferotemporal cortex. Cerebral Cortex, 11, 384–99.
Tanaka, K. (1996). Inferotemporal cortex and object vision. Annual Review of Neuroscience, 19, 109–39.
Tanaka, K. (2003). Columns for complex visual object features in the inferotemporal cortex: clustering of cells with similar but slightly different stimulus selectivities. Cereb Cortex, 13, 90–99.
Thurley, K., Senn, W., & Luscher, H. R. (2008). Dopamine increases the gain of the input-output response of rat prefrontal pyramidal neurons. Journal of Neurophysiology, 99(6), 2985–97.
Tremblay, R., Lee, S., & Rudy, B. (2016). GABAergic interneurons in the neocortex: From cellular properties to circuits. Neuron, 91, 260–92.
Tsao, D. Y., Freiwald, W. A., Tootell, R. B., et al. (2006). A cortical region consisting entirely of face-selective cells. Science, 311(5761), 670–4.
Tsodyks, M. V. (1990). Hierarchical associative memory in neural networks with low activity level. Modern Physics Letters B, 4, 259–65.
Tsodyks, M. V., & Markram, H. (1997). The neural code between neocortical pyramidal neurons depends on neurotransmitter release probability. Proceedings of the National Academy of Sciences, 94, 719–23.
Van Petten, C. (2014). Examining the N400 semantic context effect item-by-item: Relationship to corpus-based measures of word co-occurrence. International Journal of Psychophysiology, 94, 407–19.
Vander Weele, C. M., Siciliano, C. A., Matthews, G. A., et al. (2018). Dopamine enhances signal-to-noise ratio in cortical-brainstem encoding of aversive stimuli. Nature, 563, 397–401.
Varela, J. A., Sen, K., Gibson, J., et al. (1997). A quantitative description of short-term plasticity at excitatory synapses in layer 2/3 of rat primary visual cortex. Journal of Neuroscience, 17(20), 7926–40.
Volman, V., Behrens, M. M., & Sejnowski, T. J. (2011). Downregulation of parvalbumin at cortical GABA synapses reduces network gamma oscillatory activity. Journal of Neuroscience, 31(49), 18137–48.
Wallis, J. D., & Miller, E. K. (2003). From rule to response: neuronal processes in the premotor and prefrontal cortex. Journal of Neuophysiology, 90(3), 1790–806.
Wallis, J. D., Anderson, K. C., & Miller, E. K. (2001). Single neurons in prefrontal cortex encode abstract rules. Nature, 411(6840), 953–6.
Wang, X. (2002). Probabilistic decision making by slow reverberation in cortical circuits. Neuron, 36, 955–68.
Weinberger, N. M. (1998). Physiological memory in primary auditory cortex: characteristics and mechanisms. Neurobiology of Learning and Memory, 70(1–2), 226–51.
Willems, R. M., Frank, S. L., Nijhof, A. D., et al. (2015). Prediction during natural language comprehension. Cerebral Cortex, 26(6), 2506–16.
Wilson, H. R., & Cowan, J. D. (2012). Excitatory and inhibitory interactions in localized populations of model neurons. Biophysics Journal, 12(1), 1–24.
Wirth, S., Yanike, M., Frank, L. M., et al. (2003). Single neurons in the monkey hippocampus and learning of new associations. Science, 300(5625), 1578–81.
Yakovlev, V., Fusi, S., E., B., et al. (1998). nter-trial neuronal activity in inferior temporal cortex: A putative vehicle to generate long-term visual associations. Nature Neuroscience, 1(4), 310–17.
Yizhar, O., Fenno, L., Prigge, M., et al. (2011). Neocortical excitation/inhibition balance in information processing and social dysfunction. Nature, 477(7363), 171–8.
Yoshida, M., Naya, Y., & Miyashita, Y. (2003). Anatomical organization of forward fiber projections from area TE to perirhinal neurons representing visual long-term memory in monkeys. Proceedings of the National Academy of Sciences, 100(4257–62).
Young, M., & Yamane, S. (1992). Sparse population coding of faces in the inferotemporal cortex. Science, 256(5061), 1327–31.
Acknowledgements
The authors gratefully thank the reviewers for their invaluable comments to improve the paper.
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception, design, writing. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Ethics approval
Non applicable.
Consent to participate
Not applicable.
Consent for publication
Not applicable
Conflicts of interest
The authors declare that they have no conflict of interest.
Additional information
Action Editor: Nicolas Brunel
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: The latching dynamics
Appendix: The latching dynamics
1.1 A.1 Sketch of the analysis in Köksal Ersöz et al. (2020)
A sequence of \(N-1\) learned patterns \(\xi ^i=(\xi ^i_1\dots ,\xi ^i_{N-1})\) was considered in a network of N units, where \(\xi ^k_i=\xi ^k_{i+1}=1\) and the other coordinates are 0. These patterns sit on vertices of the hypercube \([0,1]^N\) and the trajectories of the activity Eq. (5) lie within the hypercube. The analysis will be simplified by the presence of the factor \(x_i(1-x_i)\) in (5), which forces the trajectories starting on one face of the hypercube to lie within that face, whatever the dimension of the face. In particular, the vertices are the steady-states of (5). The evolution of the system is governed by the activity Eq. (5) and the STD is driven by (1). The corresponding connectivity matrix is (7), according to the learning rule (3). In the following we sketch without proofs the analysis in (Köksal Ersöz et al., 2020).
The first question is about the stability of learned patterns \(\xi ^i\). This is determined by the sign of the eigenvalues at \(\xi ^i\) of the Jacobian matrix of the system (5), which we write \(\sigma ^i_j\) (\(j=1,\dots ,N\)). These eigenvalues are easily computed because the edges emanating from \(\xi ^i\) on the hypercube are the eigendirections. It can be seen that all eigenvalues with \(j<i\) or \(j>i+1\) are negative if the condition \(2\lambda +I>S\) is satisfied (\(S=(1+\rho )^{-1}<1\) is the value towards which the synaptic strength \(s_i\) decays as time elapses). The two remaining eigenvalues are
where \(m=1\) if \(i=1\), \(m'=1\) if \(i=N-1\), and \(m=m'=2\) for all other values of i.
Without STD the learned patterns \(\xi ^i\) with \(1<i<N-1\) are stable under the mild condition \(S<2\lambda +I<3-\mu\). When \(i=1\) or \(N-1\) the condition becomes \(S<2\lambda +I<2-\mu\), which is more restrictive. We assumed in Köksal Ersöz et al. (2020) that \(I=0\). The results are still valid with \(I>0\) small enough. In this study we have set \(I=0\).
If STD is on, \(s_i\) and \(s_{i+1}\) decay towards \(0<S<1\), so that eventually \(\sigma ^i_i\) or \(\sigma ^i_{i+1}\) may become positive at finite time. When \(i=1\), (11) shows that \(\sigma ^1_1>\sigma ^1_2\), therefore \(\xi ^i\) becomes first unstable and the direction of this instability is \(x_1\). When \(i>1\), \(m=m'\) in (11), so that \(\sigma ^i_i\) and \(\sigma ^i_{i+1}\) may become positive simultaneously. However, thanks to the previous switch \(\xi ^{i-1}\rightarrow \xi ^i\), \(s_i\) has been decaying towards S while \(s_{i+1}\) was still close to 1 (\(x_{i+1}=0\) at \(\xi ^{i-1}\)). Hence in any case \(\xi ^i\) is first destabilized along \(x_i\).
We claim that “typical” trajectories starting near \(\xi ^i\) will converge towards \(\xi ^{i+1}\) by first decreasing \(x_i\) towards 0, so that the trajectory converges for some time towards the intermediate state \(\hat{\xi }^i=(0,1,0,\dots ,0)\) (which represents the “overlap” between the learned patterns \(\xi ^i\) and \(\xi ^{i+1}\)), then increasing \(x_{i+2}\) towards 1, so that \(\xi ^{i+1}\) is reached. Noise is an essential ingredient of this behavior as we will see. The proof of the claim relies on the following properties of the system: (i) the synaptic variables \(s_j\) are slow compared to the activity variables \(x_j\) and can be seen, in the limit of “stationary variables”, as bifurcation parameters for the \(x_j\)’s. This is the idea of “dynamic bifurcation”. These bifurcations drive the dynamics of the activity variables. (ii) The analysis can be restricted to the flow-invariant 2 dimensional face \(F_i\) on the hypercube, with vertices \(\xi ^{i}\), \(\hat{\xi }^{i}\) and \(\xi ^{i+1}\). In \(F_i\), \(x_{i+1}=1\), \(x_j=0\) when \(j<i\) or \(j>i+2\) and the coordinates are \(x_i\), \(x_{i+2}\). In these coordinates \(\xi ^i=(1,0)\), \(\xi ^{i+1}=(0,1)\) and \(\hat{\xi }^i=(0,0)\).
Without synaptic plasticity, the learned patterns are stable and the dynamics in \(F_i\) looks like Fig. 9. When STD is on, synaptic strengths \(s_i\) of active neurons diminish slowly, which triggers successive transitions in \(F_i\). The proof of the claim is provided in Köksal Ersöz et al. (2020) and relies upon slow-fast dynamical systems theory. Here we only describe the possible scenarios.
The eigenvalue at \(\xi ^i\) along coordinate \(x_i\) is given in (11) and note that \(\hat{\xi }^i\) has a double eigenvalue in \(F_i\), given by
Initially, \(\xi ^i\) is stable. Let \(t_{(1,0)}\) be the time at which \(ms_i(t_{(1,0)})+s_{i+1}(t_{(1,0)})=\mu +2\lambda\) so that \(\sigma ^i_i\) becomes positive, and let \(t_{0,0}\) be the time at which the initially positive eigenvalue \(\hat{\sigma }^i_i\) becomes negative. This happens when \(s_{i+1}(t_{(0,0)})=\lambda\). Depending on the order of these two critical times two scenarios can occur.
Scenario 1
\(t_{(1,0)}<t_{(0,0)}\). A (dynamic) bifurcation of a stable equilibrium occurs on the edge \(x_{i+2}=0\) from the equilibrium (1, 0). This equilibrium travels towards (0, 0), which is reached at time \(t_{(0,0)}\). A trajectory starting at initial time from the vicinity of \(\xi ^i=(1,0)\) will therefore follow the edge \(x_{i+2}=0\) until it arrives in a neighborhood of \(\hat{\xi }^i\).
Scenario 2
\(t_{(1,0)}>t_{(0,0)}\), so that an equilibrium bifurcates first from (0, 0) along the edge \(x_{i+2}=0\). The point (1, 0) is still stable but its basin of attraction shrinks. Then sufficient noise may allow the trajectory starting in the vicinity of (1, 0) to escape its basin of attraction and converge towards \(\hat{\xi }^i\).
In both scenarios, in order for the trajectory to converge to \(\xi ^{i+1}\), one must have \(\sigma ^{i+1}_{i+2}<0\), which by (11) requires \(s_{i+1}+m's_{i+2}>\mu +2\lambda\). Since \(s_{i+2}\) is still close to 1 this condition is easily satisfied, as was seen above, unless \(m'=1\). In this case \(i=N-1\), which implies that the last pattern of the chain can hardly be attained by the system.
Additional constraints shared by the two scenarios must be satisfied, notably that \(\hat{\xi }^i\) be stable in the transverse directions to \(F_i\). This gives the condition \(\mu <\lambda\) (see Fig. 10).
1.2 A.2 Latching dynamics in the case of n-ways branching
We analyze how the presence of a n-ways branching node in the graph of learned patterns affects the occurrence of latching dynamics. For the sake of clarity we concentrate on the example associated with the connectivity matrix (8), where the node \(\xi ^2\) has multiplicity 3 (Example 1, Sect. 2.2). The generalization to higher multiplicity is straightforward.
Now the equations for the neural activity are (10) where the term \(\nu _i\) is 0 if \(i\ne 3\) and \(\nu _3=\lambda\). Along chains which do not encounter the branching “choice” \(\xi ^2\rightarrow \xi ^3\) or \(\xi ^2\rightarrow \xi ^5\), the latching dynamics analysis proceeds as sketched in A.1. Indeed the higher excitation at unit 3, which in the factor of \(x_i(1-x_i)\) in (10) reads \(3s_3x_3\), is compensated by the higher inhibition \(-3\lambda x_3\). We therefore concentrate on the piece of the graph defined by the multi node \(\xi ^2\) and its adjacent patterns \(\xi ^3\) and \(\xi ^5\).
We expect \(\xi ^2\) becomes first unstable along its coordinate \(x_2\), converging towards the intermediate state \(\hat{\xi }^2=(0,0,1,0,\dots ,0)\). But then either \(x_4\) or \(x_6\) should rise to 1 (transitions \(\xi ^2\rightarrow \xi ^3\) or \(\xi ^2\rightarrow \xi ^5\). Therefore we need to consider the cube \(F_3\) defined by setting \(x_3=1\) and \(x_j=0\) for \(j\ne 2,4,6\). In \(F_3\), \(\xi ^2=(1,0,0)\), \(\xi ^3=(0,1,0)\), \(\xi ^5=(0,0,1)\) and \(\hat{\xi }^2=(0,0,0)\). The equations in \(F_3\) are
Observe that restricting further the equations to the squares defined by \(x_4=0\) or \(x_6=0\) leads to systems completely analogous to those in \(F_i\) in A.1, so that the same analysis applies to both faces. Moreover the transverse eigenvalues to these faces at \(\xi ^3\) and \(\xi ^5\) are, in the notations of A.1, \(\sigma ^3_6=\sigma ^5_4=-2\lambda +s_3<0\). Therefore the dynamics will closely follow one of the faces and the transitions \(\xi ^2\rightarrow \xi ^3\) and \(\xi ^2\rightarrow \xi ^5\) will occur with equal probability. Figure 11 illustrates the global dynamics in \(F_3\) in a case corresponding to scenarios 1-(c) or 2-(c) in Fig. 10.
1.3 A.3 Relation between the number of excitatory connections and self-inhibition
Consider that the system is in \(F_3\) and \((x_2, x_4, x_6) = (1,0,0)\), which corresponds to \(\xi ^2\). As \(s_i\) variables varies in slow time, the system (13) will undergo several bifurcations that leading to either \(\xi ^2 \rightarrow \hat{\xi }^2 \rightarrow \xi ^3\) or \(\xi ^2 \rightarrow \hat{\xi }^2 \rightarrow \xi ^5\) transitions, with equal probabilities. The conditions for the \(\xi ^2 \rightarrow \hat{\xi }^2 \rightarrow \xi ^3\) transition can be derived form the eigenvalue expressions:
-
Initially stable equilibrium point \((x_2, x_4, x_6) = (1,0,0)\) becomes unstable at \(t = t^3_{(1, 0, 0)}\):
$$2 s_2(t^3_{(1, 0, 0)}) + s_3(t^3_{(1, 0, 0)}) = \mu + 2\lambda .$$ -
Initially unstable equilibrium point \((x_2, x_4, x_6) = (0,0,0)\) becomes stable at \(t = t^3_{(0, 0, 0)}\):
$$s_3(t^3_{(0, 0, 0)}) = \lambda .$$
The next transition is \(\xi ^3 \rightarrow \hat{\xi }^3 \rightarrow \xi ^4\) in \(F_4\) defined by setting \(x_4 = 1\) and \(x_j = 0\) for \(j\ne 3,5\). In \(F_4\) the equations of the neural activity are reduced to
The conditions for the transition \(\xi ^3 \rightarrow \hat{\xi }^3 \rightarrow \xi ^4\) are
-
Initially stable equilibrium point \((x_3, x_5) = (1,0)\) becomes unstable at \(t = t^4_{(1, 0)}\):
$$3 s_3(t^4_{(1, 0)}) + s_4(t^4_{(1, 0)}) = \mu + 2\lambda + \nu _3.$$ -
Initially unstable equilibrium point \((x_3, x_5) = (0,0)\) becomes stable at \(t = t^4_{(0, 0)}\):
$$s_4(t^4_{(0, 0)}) = \lambda .$$
Assume that \(t^3_{(1, 0, 0)} < t^3_{(0, 0, 0)}\) and \(t^4_{(1, 0)} < t^4_{(0, 0)}\). To have equally distributed pattern duration along a sequence, the \(s_i\) values at the bifurcation moments should read \(s_2(t^3_{(1, 0, 0)}) = s_3(t^4_{(1, 0)})\) and \(s_3(t^3_{(1, 0, 0)}) = s_4(t^4_{(1, 0)})\). With this restriction, the following algebraic condition can be deduced:
that is
More generally, the relation between the number excitatory connections \(d_i\) and the level of self-inhibition \(\nu _i\) is given by:
with \(s_i (t_{(1, 0)})\) being the value of \(s_i\) at the moment where a learned pattern changes stability. This implies that to succeed in sequential activation between the learned patterns, over-excitation imposed by the learning should be balanced by self-inhibition.
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Köksal Ersöz, E., Chossat, P., Krupa, M. et al. Dynamic branching in a neural network model for probabilistic prediction of sequences. J Comput Neurosci 50, 537–557 (2022). https://doi.org/10.1007/s10827-022-00830-y
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10827-022-00830-y