The inferior colliculus (IC) represents a crucial relay station in the auditory pathway, located in the midbrain's tectum and primarily projecting to the thalamus. Despite the identification of distinct cell classes based on various biomarkers in the IC, their specific contributions to the organization of auditory tectothalamic pathways have remained poorly understood. In this study, we demonstrate that IC neurons expressing parvalbumin (IC PV+ ) or somatostatin (IC SOM+ ) represent two minimally overlapping cell classes throughout the three IC subdivisions in mice of both sexes. Strikingly, regardless of their location within the IC, these neurons predominantly project to the primary and secondary auditory thalamic nuclei, respectively. Cell class-specific input tracing suggested that IC PV+ neurons primarily receive auditory inputs, whereas IC SOM+ neurons receive significantly more inputs from the periaqueductal gray and the superior colliculus (SC), which are sensorimotor regions critically involved in innate behaviors. Furthermore, IC PV+ neurons exhibit significant heterogeneity in both intrinsic electrophysiological properties and presynaptic terminal size compared with IC SOM+ neurons. Notably, approximately one-quarter of IC PV+ neurons are inhibitory neurons, whereas all IC SOM+ neurons are excitatory neurons. Collectively, our findings suggest that parvalbumin and somatostatin expression in the IC can serve as biomarkers for two functionally distinct, parallel tectothalamic pathways. This discovery suggests an alternative way to define tectothalamic pathways and highlights the potential usefulness of Cre mice in understanding the multifaceted roles of the IC at the circuit level.

Audio-visual approaches involving visual inputs have laid the foundation for recent progress in speech separation. However, the optimization of the concurrent usage of auditory and visual inputs is still an active research area. Inspired by the cortico-thalamo-cortical circuit, in which the sensory processing mechanisms of different modalities modulate one another via the non-lemniscal sensory thalamus, we propose a novel cortico-thalamo-cortical neural network (CTCNet) for audio-visual speech separation (AVSS). First, the CTCNet learns hierarchical auditory and visual representations in a bottom-up manner in separate auditory and visual subnetworks, mimicking the functions of the auditory and visual cortical areas. Then, inspired by the large number of connections between cortical regions and the thalamus, the model fuses the auditory and visual information in a thalamic subnetwork through top-down connections. Finally, the model transmits this fused information back to the auditory and visual subnetworks, and the above process is repeated several times. The results of experiments on three speech separation benchmark datasets show that CTCNet remarkably outperforms existing AVSS methods with considerably fewer parameters. These results suggest that mimicking the anatomical connectome of the mammalian brain has great potential for advancing the development of deep neural networks.

The inferior colliculus (IC) is the hub along the auditory pathway. Although it is fundamentally an auditory structure, the neurons in the IC, especially its non-lemniscal part also respond to multimodal stimuli. However, the sources of these non-auditory inputs are unclear. In this study, we injected the rAAV2-retro virus, a virus with efficient retrograde function, into the non-lemniscal IC of the Ai14 reporter line. The majority of cortical and subcortical brain areas, including cognitive, motor, somatosensory, auditory, and visual-related regions were revealed. The quantified whole brain input data have showed that the non-lemniscal IC received a higher proportion of inputs from ipsilateral cortical brain regions. The non-lemniscal IC integrates different multimodal patterns, for the dorsal cortex (ICD) receives primarily auditory inputs, and the external cortex (ICE) receives primarily auditory and somatosensory inputs. These findings demonstrate that auditory integration is shaped by a network of multi-sensory connections in the non-lemniscal IC subregions.

The expression of defensive responses to alerting sensory cues requires both general arousal and a specific arousal state associated with defensive emotions. However, it remains unclear whether these two forms of arousal can be regulated by common brain regions. We discovered that the medial sector of the auditory thalamus (ATm) in mice is a thalamic hub controlling both general and defensive arousal. The spontaneous activity of VGluT2-expressing ATm (ATmVGluT2+) neurons was correlated with and causally contributed to wakefulness. In sleeping mice, sustained ATmVGluT2+ population responses were predictive of sensory-induced arousal, the likelihood of which was markedly decreased by inhibiting ATmVGluT2+ neurons or multiple downstream pathways. In awake mice, ATmVGluT2+ activation led to heightened arousal accompanied by excessive anxiety and avoidance behavior. Notably, blocking their neurotransmission abolished alerting stimuli-induced defensive behaviors. These findings may shed light on the comorbidity of sleep disturbances and abnormal sensory sensitivity in specific brain disorders.

Largely topographical projections from different modules of the thalamus, such as the primary, secondary and association sensory thalamus, to hierarchically defined cortical areas have been recognized across sensory systems. However, how corticothalamic projections, which are believed to be crucial for the remarkable flexibility and precision exhibited by our sensory systems, are organized remained poorly understood compared with the thalamocortical counterpart. Here we report that, first, the primary auditory thalamus received direct inputs from cortical L5 neurons. Second, in contrast to the robust thalamocortical topography, L5 neurons in each of the primary, secondary and association auditory cortical regions project to each individual module of the auditory thalamus at the macroscale. Third, the association cortex provided the most L5 inputs to all thalamic modules followed by the secondary and primary auditory cortices. Lastly, L5 axon terminals were mainly varicosity-type and evenly distributed across thalamic modules, but those in the polymodal association module were the largest. Our data suggest that all the modules of the auditory thalamus may be under the modulation of common L5 inputs. This fully-connected-like corticothalamic architecture urges a revision of the traditional hierarchical model in the sensory systems.

Alterations in internal states, such as elevated arousal level and increased anxiety or fear, triggered by alerting environmental cues are required for behavioral state transitions promoting survival. However, the specific brain region that plays an interfacing role between alerting stimuli and internal states remains to be identified. Here, we report that the medial sector of the auditory thalamus (ATm), which consists of a group of non-lemniscal thalamic nuclei, can fulfill this function. VGluT2-expressing ATm (ATm VGluT2+ ) neurons receive direct and strong inputs from both visual and auditory midbrain regions, and project to multiple downstream structures critically involved in brain state regulation. Their activity was correlated with, and indispensable for, both blue light- and sound-induced NREM sleep-to-Wake transition, and their arousing effects were mainly mediated by, but not limited to, the temporal association cortices. ATm VGluT2+ neuron activation in awake behaving mice induced pupil dilation and behavioral responses suggestive of anxiety. Blocking the neurotransmitter release of ATm VGluT2+ neurons receiving auditory inputs selectively abolished loud noise-triggered escape behavior but not locomotion. Thus, the ATm is an interface in mouse brain that can transform alerting environmental cues into internal arousal and emotional state alterations that promote survival.

The protracted maturational process of temporal processing in layer 4 (L4) of primary auditory cortex (A1) has been extensively studied. Accumulating evidences show that layer 5 (L5) receives direct thalamic inputs as well. How the temporal responses in L5 may developmentally emerge remains unclear. Using in vivo loose-patch recordings in rat A1, we found that putative pyramidal (Pyr) neurons in developing L5 exhibited adult-like stimulus-following ability but less bursting shortly after hearing onset. L5 Pyr neurons in adult A1 exhibited phase-locking similar to L4 neurons, while L5 fast-spiking (FS) neurons showed greater phase-locking at 7 and 12.5 pps. In developing L5, whole-cell recordings revealed inhibition with decay constant comparable to that in adult L5, thereby avoiding the summation of inhibition that contributed to the strong adaptation in L4. Given the targets of L5 outputs, the relatively precocious temporal processing in L5 might contribute to temporal response maturation in connected cortical and subcortical areas. Our findings were in agreement with the idea that L5 may be a "hub" for processing cortical inputs and outputs that can operate independently of L4.

Systematic examination of the inputs and outputs of the nonlemniscal auditory thalamus will facilitate the functional elucidation of this complex structure in the central auditory system. In mice, comprehensive tracing studies that reveal the long-range connectivity of the nonlemniscal auditory thalamus are lacking. To this end, we used Cre-inducible anterograde and monosynaptic retrograde viruses in Calbindin-2A-dgCre-D and Calretinin-IRES-Cre mice, focusing on the differences across subdivisions of the nonlemniscal auditory thalamus. We found that, 1) the dorsal and medial parts of the auditory thalamus were predominantly connected to sensory processing centers, whereas the posterior intralaminar (PIN) and peripeduncular nucleus (PP) were additionally connected to emotion and motivation modulation centers; 2) ventral auditory cortical areas were the major source of cortical inputs for all subdivisions, and the PIN/PP received more inputs from cortical layer 5 than other subdivisions did; 3) deep layers of the superior colliculus and rostral part of the nonlemniscal inferior colliculus preferentially projected to the PIN/PP; and 4) compared with the dorsal auditory thalamus, the PIN/PP mainly innervated association cortices. In addition, new brain areas connected to the nonlemniscal auditory thalamus, mostly the PIN/PP, were identified. Our results suggested subdivision-specific function of the nonlemniscal auditory thalamus in sound processing.

Faithful representation of sound envelopes in primary auditory cortex (A1) is vital for temporal processing and perception of natural sounds. However, the emergence of cortical temporal processing mechanisms during development remains poorly understood. Although cortical inhibition has been proposed to play an important role in this process, direct in-vivo evidence has been lacking. Using loose-patch recordings in rat A1 immediately after hearing onset, we found that stimulus-following ability in fast-spiking neurons was significantly better than in regular-spiking (RS) neurons. In-vivo whole-cell recordings of RS neurons revealed that inhibition in the developing A1 demonstrated much weaker adaptation to repetitive stimuli than in adult A1. Furthermore, inhibitory synaptic inputs were of longer duration than observed in vitro and in adults. Early in development, overlap of the prolonged inhibition evoked by 2 closely following stimuli disrupted the classical temporal sequence between excitation and inhibition, resulting in slower following capacity. During maturation, inhibitory duration gradually shortened accompanied by an improving temporal following ability of RS neurons. Both inhibitory duration and stimulus-following ability demonstrated exposure-based plasticity. These results demonstrate the role of inhibition in setting the pace for experience-dependent maturation of temporal processing in the auditory cortex.