I. Introduction

Acoustic communication relies on detecting, encoding, and transmitting sound cues by converting auditory stimuli into neural signals. Spiral Ganglion Neurons (SGNs) serve as the essential neurosensory interface between the cochlea and the brain. Hair Cells (HCs) release neurotransmitters at ribbon synapses, triggering Postsynaptic Potentials (PSPs) in SGNs, which subsequently generate Action Potentials (APs). As the first generators of APs, SGNs relay these neural signals to the brainstem, establishing precise connections with Cochlear Nucleus (CN).^[1]

SGNs are predominantly located in the modiolus, with most residing in Rosenthal’s canal. Their peripheral processes extend through the osseous spiral lamina to connect with HCs in the organ of Corti, while their central processes bundle together to form the auditory nerve.^[2] Dysfunction or degeneration of SGNs results in Sensorineural Hearing Loss (SNHL), a major cause of permanent hearing impairment. Nevertheless, investigations into the auditory system have advanced at a slower pace compared to other sensory systems, largely due to the minute size and inaccessibility of the inner ear.^[3]

Like other neurons, SGNs generate APs through fundamental mechanisms, including depolarization beyond the threshold, voltage-gated ion channel opening, and the subsequent all-or-none spikes. However, SGNs exhibit highly specialized morphological features, synaptic structures, innervation patterns, and firing properties that distinguish them from other neuronal types. These distinctive attributes are crucial for encoding auditory information, ensuring the rapid and precise auditory transmission to the central auditory system.

For individuals with severe SGN damage, Auditory Brainstem Implants (ABIs) provide an alternative by directly stimulating second-order neurons in the CN. Nevertheles, clinical efficacy of ABIs remains limited by their restricted tonotopic selectivity within the CN, resulting in spectral smearing that diminishes frequency resolution. Moreover, bypassing the specialized neural circuitry may compromise the precise coding required for natural auditory perception. Recent advances in neural prosthetics and stereotactic techniques have contributed to the development of a new generation of ABIs designed to provide more precise and direct neural stimulation. While their benefits require further investigation, they hold promise for improving auditory outcomes.^[4,5] In addtion, emerging approaches in gene therapy, neuroprotection, and neural regeneration strategies offer potential strategies for restoring SGN function though most of them remain in the preclinical setting.^[6,7,8]

The present study aims to comprehensively investigate the structure and function of SGNs, with particular emphasis on their roles in neural transmission and signal encoding within the auditory system. Given their critical role in conveying auditory information from the cochlea to the central auditory system, this study seeks to deepen the understanding of physiological mechanisms underlying SGN function and provide valuable insights that may facilitate the development of effective interventions for SNHL.

II. Structure and function of SGNs

2.1 Types of SGNs

SGNs serve as the essential neurosensory link between the cochlea and the brain, forming the primary neural pathway for auditory signal transmission. SGNs fall into two primary categories, Type I and Type II, depending on the pattern of peripheral innervation.^[9]

The bipolar Type I SGNs, often referred to as radial fibers, are more numerous and have larger-diameter peripheral processes. In Type I SGNs, both the peripheral and central processes are myelinated, except for a short initial segment near the synapse.^[10,11] They make up 90 % to 95 % of the total number of SGNs, and largely innervate the IHCs which constitute fewer than 25 % of the HCs. Due to the greater number of Type I SGNs compared to IHCs, each IHC establishes connections with multiple (approximately 10 to 20) Type I SGNs, forming a radial innervation pattern, while each Type I SGN innervates only a single IHC, ensuring fast and precise auditory signal transmission. Notably, such exclusive connection of each Type I SGN to a single IHC ensures a highly specific and one-to-one transmission of auditory information, serving as the primary afferent neurons in the auditory system. The density of Type I SGNs per IHC varies along the cochlear tonotopic axis, with 3 to 4 fibers per IHC at the apex and base, and a higher density (up to 15 fibers per IHC) around the 1 kHz region, which corresponds to the region of peak sensitivity for human speech perception.^[1,11,12,13]

On the other hand, the bipolar or pseudounipolar Type II SGNs, also known as spiral or longitudinal fibers, are fewer and have smaller-diameter peripheral processes. Both the dendrites and axons of Type II SGNs are unmyelinated. They represent only 5 % to 10 % of the total SGN population, and primarily innervate Outer Hair Cells (OHCs) which constitute more than 75 % of the HCs.^[1,11] Given the greater number of OHCs compared to Type II SGNs, each fiber establishes connections with multiple OHCs, running in longitudinal fashion. Each OHC contacts only one Type II SGN with each Type II SGN innervating approximately 15 to 20 OHCs. The Type II SGNs connect primarily to the outer row of OHCs in the basal region of the cochlea, with their innervation progressing toward the middle and then the innermost rows as they extend toward the apical region of the cochlea.^{[1,11,12,14,15]}

2.2 Auditory nerve

Auditory nerve, also known as cochlear nerve, consists of a bundle of the central axons of SGNs, rather than their entire neurons including dendrites and cell bodies.^[16] Although the term is sometimes used interchangeably with SGNs, the auditory nerve specifically refers to the axonal projection transmitting auditory information from the cochlea to the brainstem.

The centrally projecting axons of SGNs converge within the modiolus, forming approximately 30,000 neurons in each auditory nerve in human.^[17] At the porus acousticus within the medial Internal Auditory Canal (IAC), the superior and inferior vestibular nerves and the cochlear nerve merge, often forming a crescent shape. As the vestibulocochlear nerve and facial nerve travel through the Cerebellopontine Angle (CPA), they remain distinct and surrounded by Cerebrospinal Fluid (CSF) before entering the pontomedullary junction of the brainstem.^[18]

Auditory nerve preserves a precise tonotopic organization, with low Characteristic Frequency (CF) fibers located centrally and high CF positioned fibers peripherally, forming a helical arrangement that mirrors the cochlear spiral. This spatial organization extends through the IAC into the central nervous system, where the fibers project topographically into CN, with low-frequency fibers located ventrally and high-frequency fibers dorsally.^[11,19]

2.3 Afferent SGNs

Both Type I and Type II SGNs function as afferent neurons, transmitting auditory signals to the central nervous system. IHCs receive approximately 90 % ~ 95 % of afferent innervation, transmitting precise auditory information to the brainstem. In contrast, OHCs account for only about 5 % ~ 10 % of afferent innervation, and their associated functional properties remain poorly understood.^[11,20] These primary auditory neurons terminate in the CN, located on the dorsolateral side of the brain stem. Synapses are established with various classes of cells throughout CN, giving rise to multiple parallel representations of the acoustic information. Each SGN projects centrally to the brainstem, where it bifurcates into two branches. The anterior or ascending branch projects to the Anteroventral CN (AVCN) while the descending branch targets the Posteroventral CN (PVCN) and the Dorsal CN (DCN). Each branch independently targets tonotopically appropriate post-synaptic sites, forming large endbulbs of Held in the VCN and boutons in the DCN.^[3,21,22]

In addition to the afferent pathway, HCs have efferent innervation that conveys descending feedback from the brain. Unlike vision and touch, auditory transduction is uniquely modulated at the periphery by Olivocochlear (OC) efferent fibers, which travel from the brain back to the inner ear.^[23] OHCs are the primary targets of efferent neurons, which regulate cochlear mechanics and auditory sensitivity by Acetylcholine (ACh) to inhibit OHC while also making sparse connections with afferent Type II SGNs. Conversely, IHCs receive no direct efferent input, serving as the primary presynaptic source for afferent Type I SGNs.^[24] Thus, OHCs are directly targeted by Medial OC (MOC) efferent fibers while IHCs are indirectly influenced by Lateral OC (LOC) fibers, which modulate Type I SGN terminals rather than forming direct synapses onto IHCs.^[23,24,25] Table 1 summarizes the overall structure and function of afferent SGNs.

III. Neural transmission of SGNs

3.1 PSPs in SGNs

In SGNs, depolarization is initiated by Na⁺ influx through voltage-gated Na⁺ channels whereas in cochlear HCs, it is driven by K⁺ influx through Mechanoelectrical Transduction (MET) channels, reflecting their specialized mechanisms. In HCs, MET channel-mediated depolarization activates L-type (CaV1.3) voltage-gated Ca²⁺ channels at presynaptic active zones, and the resulting Ca²⁺ influx triggers synaptic vesicle fusion and subsequent glutamate release onto the postsynaptic membrane of SGNs.^[26,27] Such Ca²⁺-dependent exocytosis shares conventional features of chemical neurotransmission observed elsewhere in both the central and peripheral nervous systems, wheasre ribbon synapses also show specialized mechanisms that enable rapid and sustained neurotransmitter.^[28]

PSPs are graded changes in the membrane potential of a neuron that occur following synaptic transmission. PSPs vary in magnitude, unlike APs which follow an all- or-none principle. They can be Excitatory PSPs (EPSPs), which depolarize the membrane and increase the likelihood of AP generation, or Inhibitory PSPs (IPSPs), which hyperpolarize the membrane and decrease excitability. Their magnitude depends on neurotransmitter release and receptor density on the postsynaptic membrane.^[29] EPSPs are typically large enough to exceed the spike threshold, likely due to coordinated multi-vesicular release.^[30] In the central auditory pathway following SGNs, the integration of depolarizing EPSPs and hyperpolarizing IPSPs is fundamental to neuronal signal processing. A hyperpolarizing IPSP occurs when the synaptic current has a reversal potential more negative than the Resting Membrane Potential (RMP). When an IPSP coincides with an EPSP, it reduces the EPSP amplitude, thus modulating synaptic excitation and overall neuronal output.^[31]

3.2 Glutamate onto SGNs

Glutamate, released from the ribbon synapses of HCs, acts as the main neurotransmitter mediating communication with SGNs. These specialized synapses enable precise and sustained neurotransmitter release, ensuring accurate auditory signal encoding. Glutamate, the principal excitatory neurotransmitter in both the auditory periphery and the central nervous system, facilitates synaptic transmission by activating receptors located on the postsynaptic membranes of SGNs.^[30,32] Glutamate receptors are classified into ionotropic and metabotropic types. Ionotropic receptors, such as α-Amino-3-hydroxy- 5-methyl-4-isoxazolepropionic acid (AMPA), N-methyl- D-aspartate (NMDA), and kainate, are nonselective cation channels that permit Na⁺, K⁺, and sometimes Ca²⁺ influx. Metabotropic glutamate receptors modulate neuronal excitability and synaptic transmission via G-protein- coupled pathways.^[10,33]

Excessive glutamate release or impaired clearance leads to synaptic accumulation, causing AMPA receptor desensitization and extrasynaptic receptor activation. Acoustic overstimulation can induce excitotoxicity, resulting in SGN degeneration and hearing impairment.^[34] To prevent this, extracellular glutamate levels are regulated through passive diffusion and active transport mediated by glutamate transporters, including excitatory amino acid transporters. These transporters rapidly bind glutamate near synapses, competing with receptors and shaping glutamate transients, thereby preventing unintended receptor activation at neighboring synapses.^[35]

Excitotoxicity by the overactivation of glutamate receptors induces massive Na⁺ and Ca²⁺ influx and water entry, leading to dendritic edema and disruption of SGNs.^[36] Glutamate agonists causes dendritic damage of SGNs, suggesting that excessive glutamate release may underlie such injury.^[37] Also, excessive K⁺ efflux from overstimulated HCs and SGNs elevates perilymphatic K⁺ levels, further contributing to ribbon synapse degeneration.^[38]

Extensive animal studies have investigated strategies to regulate glutamate activity, including NMDA receptor antagonists, which mitigate excitotoxic damage and demonstrate efficacy in noise-induced tinnitus models.^[39,40] Glutamate modulators like riluzole have shown neuroprotective effects against noise- and drug-induced ototoxicity.^[41,42] Additionally, the delivery of the wild-type VGLUT3 gene via an Adeno-Associated Virus (AAV1) vector has been shown to enhance glutamate clearance and restores auditory function.^[43]

3.3 AP in SGN

Each SGN transmits information through brief electrical pulses, known as APs, nerve impulses, or spikes, which propagate along the axon. The generation of a spike is often referred to as firing and its magnitude remains constant. Instead, information is conveyed in the number and timing of these spikes.^[44]

In the absence of sound, most SGNs generate spontaneous APs with random interspike intervals. A fraction of MET channels remains open at rest,^[45] generating a depolarizing current that drives spontaneous glutamate release at IHC ribbon synapses.^[30] This baseline neurotransmitter release modulates spontaneous AP rates in SGNs, influencing their thresholds and sensitivity to sound.^[27]

Spontaneous firing Rate (SR), defined as the average number of APs per second in the absence of stimulation, fluctuates over time. SGNs are classified based on their SR into three groups: high-SR fibers (61 %) firing at rates above 18 spike/s, medium-SR fibers (23 %) firing between 0.5 spike/s ~ 18 spike/s, and low-SR fibers (16 %) firing below than 0.5 spike/s.^[1] SR is inversely related to threshold sensitivity, such that high-SR fibers have low thresholds while low-SR fibers have high thresholds.^[46,47]

SGNs are highly responsive, exhibiting precise electrical properties and large excitatory postsynaptic currents. Remarkably, even a single event of presynaptic release, potentially involving just one vesicle, can be sufficient to elicit an APs.^[48] Nearly every neurotransmitter release event triggers a single spike unless the fiber is refractory.^[49,50] After a spike occurs, there is an absolute refractory period lasting approximately 0.5 ms ~ 0.75 ms, during which no new spike can be generated. This is followed by a relative refractory period of approximately 2 ms ~ 3 ms, during which spike generation is possible but less likely.^[44]

IV. Auditory encoding of SGNs

4.1 Frequency encoding

Frequency encoding depends on the tonotopic arrangement of HCs along the cochlea, where basal HCs detect high frequencies and apical HCs respond to low frequencies. SGNs preserve this spatial mapping through selective innervation of IHCs, such that each SGN exhibits a characteristic CF based on its cochlear location, reflecting the tonotopic organization of frequency encoding in the auditory periphery.^[44] Within the tonotopic arrangement, low-frequency SGNs connected to apical HCs are centrally positioned in the modiolar trunk, while high-frequency SGNs linked to basal HCs are located peripherally, making outer SGN damage a cause of high-frequency hearing loss. In the CN, fibers with low CFs are located ventrally, while those with higher CFs are found dorsally. This spatial organization remains intact through the IAC, where fibers twist as they enter the brainstem, thereby preserving tonotopy up to the auditory cortex.^[1,11]

Frequency information is conveyed not only by tonotopicity but also by the temporal pattern of neural activity. Phase locking, where APs synchronize with specific points in the stimulus cycle, is crucial for representing frequency. A neuron does not fire at a fixed rate, but its temporal spike pattern reliably encodes the stimulus period. In response to pure tones, nerve spikes exhibit phase locking, occurring at consistent stimulus phases. Although neurons do not fire on every cycle, their spike intervals approximate integer multiples of the stimulus period. For instance, a 500 Hz tone with a 2 ms period results in spike intervals near 2 ms, 4 ms, or 6 ms. While firing is not perfectly regular, spike intervals generally align with integer multiples of the tone’s period. For sinusoidal stimuli, phase locking is most effective below 4 kHz ~ 5 kHz.^[44] Phase locking changes along the auditory pathway, with a general decline in the upper frequency limit at higher synaptic levels.^[51]

4.2 Intensity encoding

The perception of loudness is believed to correlate with the total neural activity within the auditory system. This suggests that the perceived loudness of a tone depends not only on the activation of SGNs with CFs matching the tone but also on the spread of activity to adjacent CFs. Consequently, loudness may result from summation across different frequency channels.^[52]

The IHC-SGN innervation pattern also contributes to intensity coding. Each IHC connects to multiple SGNs, enabling the encoding of subtle intensity variations. This divergent innervation enhances the detection of soft sounds by distributing neural activation across several SGNs, compensating for variability in individual neuronal SRs and thresholds. The redundancy in neural connections improves sensitivity and reliability, facilitating precise intensity discrimination, especially at low sound levels.^[3]

Meanwhile, SGNs exhibit intrinsic variations in their firing properties, indicating their specialization for distinct aspects of auditory stimuli. Differences in threshold, SR, and operating range enable precise encoding across a wide dynamic range. The auditory system encodes sound pressure levels spanning nearly 120 dB, though individual auditory nerve fibers generally have dynamic ranges of only 20 dB ~ 50 dB from threshold to saturation. To accommodate this range, multiple nerve fibers with varying thresholds and frequency sensitivities collectively encode the full dynamic range of hearing.^[11,53] High-SR fibers with low threshold have the lowest detection thresholds at a given frequency, making them highly sensitive to low-level sounds and optimal for quiet conditions. In contrast, low-SR fibers, with high threshold and minimal saturation, are well-suited for encoding intensity variations across a broad range of sound pressure levels, enhancing discrimination in louder and noisier environments. High- and medium-SR fibers respond effectively to lower sound levels but saturate quickly, whereas low-SR fibers activate at higher levels, maintaining a more gradual increase in response.^[11,54,55]

In addition, lower SR fibers formed more extensive synaptic branches and contacted a greater number of cells within the CN compared to high-SR fibers, potentially contributing to loudness perception by increasing the pool of active neurons. At higher intensities, low-SR activation spreads activity within the AVCN, compensating for the saturation of high-SR fibers and ensuring continued intensity encoding.^[56,57]

V. Conclusions

SGNs, as the first AP generators within the auditory system, form the critical neurosensory interface between HCs and the central auditory pathway. These highly specialized primary afferent neurons possess distinct anatomical and physiological properties that are essential for precise auditory encoding. Disruption of SGNs, whether from genetic, environmental, or age-related factors, contributes to SNHL.

Recent progress in ABI technology, along with emerging therapeutic approaches such as gene transfer, neuroprotective agents, and stem cell-based regeneration, offer potential for functional recovery. Further elucidation of the SGN physiology, including their modulation with efferent pathways and integration within higher-order auditory circuits, will be crucial for advancing our understanding of auditory encoding and guiding the development of targeted interventions for hearing disorders.