A Temporal Signal-Processing Circuit Based on Spiking Neuron and Synaptic Learning

• Main Authors: Hui Wei, Yi-Fan Du
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2019-06-01
• Series: Frontiers in Computational Neuroscience
• Subjects: time-related neuron; time-processing circuit; spiking-neuron; synaptic learning; ramp activity; SCT
• Online Access: https://www.frontiersin.org/article/10.3389/fncom.2019.00041/full

Time is a continuous, homogeneous, one-way, and independent signal that cannot be modified by human will. The mechanism of how the brain processes temporal information remains elusive. According to previous work, time-keeping in medial premotor cortex (MPC) is governed by four kinds of ramp cell populations (Merchant et al., 2011). We believe that these cell populations participate in temporal information processing in MPC. Hence, in this the present study, we present a model that uses spiking neuron, including these cell populations, to construct a complete circuit for temporal processing. By combining the time-adaptive drift-diffusion model (TDDM) with the transmission of impulse information between neurons, this new model is able to successfully reproduce the result of synchronization-continuation tapping task (SCT). We also discovered that the neurons that we used exhibited some of the firing properties of time-related neurons detected by electrophysiological experiments in other studies. Therefore, we believe that our model reflects many of the physiological of neural circuits in the biological brain and can explain some of the phenomena in the temporal-perception process.