Synaptic competition: ‘to be or not to be’ the calyx of Held?

Bushy cells (BC) of the cochlear nucleus mono-innervate their target neuron, the principal cell of the medial nucleus of the trapezoid body (MNTB), via the calyx of Held (CH) terminal, which is a typically mammalian structure and perhaps the largest nerve terminal in the brain. CH:MNTB innervation has become an attractive model to study neural circuit formation because it forms quickly, passing through stages of competition in mice within 2–4 days. BCs innervate MNTB neurons by E17, but CHs do not begin to grow for another five days (P3). Progress has been made to identify molecular factors for axon guidance, CH growth, and physiological maturation of synaptic partners, but important details remain to be discovered. We summarize key events in CH formation and highlight unresolved issues in molecular and physiological signaling, roles for non-neural cells, and the nature of competition during the first postnatal week.

Download Full-text

The Calyx of Held: A Hypothesis on the Need for Reliable Timing in an Intensity-Difference Encoder

Neuron ◽

10.1016/j.neuron.2018.10.026 ◽

2018 ◽

Vol 100 (3) ◽

pp. 534-549 ◽

Cited By ~ 10

Author(s):

Philip X. Joris ◽

Laurence O. Trussell

Keyword(s):

Intensity Difference ◽

Calyx Of Held

Download Full-text

Competitive Learning in a Spiking Neural Network: Towards an Intelligent Pattern Classifier

Sensors ◽

10.3390/s20020500 ◽

2020 ◽

Vol 20 (2) ◽

pp. 500 ◽

Cited By ~ 3

Author(s):

Sergey A. Lobov ◽

Andrey V. Chernyshov ◽

Nadia P. Krilova ◽

Maxim O. Shamshin ◽

Victor B. Kazantsev

Keyword(s):

Supervised Learning ◽

Hebbian Learning ◽

Back Propagation ◽

Spike Timing ◽

Temporal Coding ◽

Competitive Learning ◽

Neuron Activity ◽

Synaptic Competition ◽

Rate Coding ◽

Emg Patterns

One of the modern trends in the design of human–machine interfaces (HMI) is to involve the so called spiking neuron networks (SNNs) in signal processing. The SNNs can be trained by simple and efficient biologically inspired algorithms. In particular, we have shown that sensory neurons in the input layer of SNNs can simultaneously encode the input signal based both on the spiking frequency rate and on varying the latency in generating spikes. In the case of such mixed temporal-rate coding, the SNN should implement learning working properly for both types of coding. Based on this, we investigate how a single neuron can be trained with pure rate and temporal patterns, and then build a universal SNN that is trained using mixed coding. In particular, we study Hebbian and competitive learning in SNN in the context of temporal and rate coding problems. We show that the use of Hebbian learning through pair-based and triplet-based spike timing-dependent plasticity (STDP) rule is accomplishable for temporal coding, but not for rate coding. Synaptic competition inducing depression of poorly used synapses is required to ensure a neural selectivity in the rate coding. This kind of competition can be implemented by the so-called forgetting function that is dependent on neuron activity. We show that coherent use of the triplet-based STDP and synaptic competition with the forgetting function is sufficient for the rate coding. Next, we propose a SNN capable of classifying electromyographical (EMG) patterns using an unsupervised learning procedure. The neuron competition achieved via lateral inhibition ensures the “winner takes all” principle among classifier neurons. The SNN also provides gradual output response dependent on muscular contraction strength. Furthermore, we modify the SNN to implement a supervised learning method based on stimulation of the target classifier neuron synchronously with the network input. In a problem of discrimination of three EMG patterns, the SNN with supervised learning shows median accuracy 99.5% that is close to the result demonstrated by multi-layer perceptron learned by back propagation of an error algorithm.

Download Full-text