Intrinsic Properties of Basal-Ganglia Projecting Neurons Modulate the Spectro-Temporal Features of Zebra Finches’ Vocalizations

Abstract

While synaptic plasticity is widely recognized as the primary mechanism for information processing and storage in the brain, intrinsic plasticity represents another critical neural mechanism. The type and magnitude of ion currents expressed by a neuron significantly influence the number, timing, and patterns of action potentials generated in response to specific inputs, thereby shaping the neuron’s role in network dynamics. Despite their fundamental importance, the role of non-synaptic mechanisms in memory and learning remains poorly understood. The birdsong system has emerged as a prominent model for investigating the neural mechanisms that underlie vocal production and learning. However, the translation of neural impulses into precise motor behaviors governing the spectral and temporal features of the complex songs is still not well characterized. Cortical nucleus High Vocal Center (HVC) in songbirds plays a crucial role in song learning and production and is comprised of three major neuronal populations that have distinct anatomical and electrophysiological properties. One class of HVC neurons is the basal ganglia projecting neurons (HVCX), which are known to play crucial roles in motor control and learning, yet their specific contributions to the production of intricate vocalizations in songbirds remain unclear. In this study, we investigated the relationship between the intrinsic properties of HVCX neurons and the spectro-temporal features of adult zebra finch vocalizations. Utilizing electrophysiological and mathematical modeling techniques combined with acoustic analysis of zebra finch songs, we first recorded the neural activity from a large set of HVCX neurons while simultaneously monitoring the birds' vocal output. We then constructed a mathematical model that replicates the firing patterns of HVCX neurons by varying the maximal conductances of the underlying ionic currents. Consistent with recent electrophysiological results that showed a consistent pattern of homogeneity of HVCX intrinsic properties within individual adult birds and heterogeneity across birds, our mathematical model simulations indicated similar magnitudes in two principal ion currents that govern these neurons firing patterns, the Ca2+ - dependent K+ (SK) current and the transient Na+ current. Strikingly, when we extracted the spectro-temporal features for the songs of each bird, we found a strong correlation between the 2D conductance space and the spectro-temporal space mapped by the mean pitch and the mean amplitude features. Our results demonstrate an explicit link between neuronal intrinsic properties and learned behavior and represent a shift from a solely synaptic focused perspective of learning to a more cellular-focused perspective. These results indicate that investigating the interdependent links between the synaptic and intrinsic properties governing neuronal dynamics will yield valuable insights into the mechanisms underlying birdsong learning.