A Thalamocortical Pathway for Decoding Communicative Syllables

A Thalamocortical Pathway for Decoding Communicative Syllables | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Biological Sciences - Article A Thalamocortical Pathway for Decoding Communicative Syllables Wei Xiong, Zhikai Zhao, Xiaojing Tang, Yunlong Wu, Yangzhen Wang, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8163343/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Social communication in rodents depends on ultrasonic vocalizations, yet the neural circuits that selectively process these complex acoustic signals remain poorly defined. Here, using wide-field calcium imaging in awake mice, we identify a previously unknown region of the auditory cortex that we termed communicative auditory field (CAF). CAF is located between the anterior auditory field and ultrasonic field and responds selectively to ultrasonic syllables. Two-photon imaging revealed that individual CAF neurons are tuned to distinct syllable categories. Transsynaptic tracing showed that CAF receives its predominant thalamic input from the posterior nucleus (POL), defining a pathway in parallel to the canonical lemniscal pathway from the ventral medial geniculate body to the primary auditory cortex. Chemogenetic inhibition of either CAF or POL neurons projecting to CAF impaired innate social behaviors, including pup retrieval and courtship, whereas inhibition of A1 was ineffective. These findings define a specialized thalamocortical circuit for decoding ethologically relevant communication sounds. Biological sciences/Neuroscience/Auditory system/Cortex Biological sciences/Neuroscience/Sensory processing Figures Figure 1 Figure 2 Figure 3 Figure 4 INTRODUCTION The cerebral cortex is anatomically and functionally organized into distinct sensory areas that contain specialized subregions for processing specific features of sensory input. This functional segregation enables efficient and high-dimensional information processing 1–5 . Among these specialized regions, those involved in language and speech processing in humans represent a particularly sophisticated example of cortical specialization for social communication. In mice, social communication relies heavily on the auditory cortex (AuC) to process complex ultrasonic vocalizations (USV), which are composed of distinct syllables 6–9 . The mouse AuC has traditionally been subdivided into tonotopically organized regions, including the primary auditory cortex (A1), ultrasonic field (UF), secondary auditory cortex (A2), and anterior auditory field (AAF). However, in contrast to the well-mapped speech-processing regions in humans (ref), the functional specialization of mouse auditory subregions for processing natural vocalizations remains poorly characterized. Although a putative vocalization-sensitive region between A2 and AAF was previously identified 10 , its thalamocortical connectivity and necessity for behavior have not been established. In contrast, the canonical A1 pathway is well-defined: it receives frequency-specific, lemniscal inputs from the ventral division of the medial geniculate body (MGBv) to encode basic acoustic features 11–13 . In this study, we combined wide-field and two-photon calcium imaging in awake mice to identify and characterize a previously undefined subregion within the AuC that is specialized for processing ultrasonic communication sounds. Using anatomical tracing, we defined its unique thalamocortical connectivity, and through functional disruption experiments, we established its necessity for natural auditory-guided social behaviors. RESULTS CAF processes pup calls over tones To identify cortical subregions specialized for processing complex vocalizations, we performed wide-field calcium imaging in awake, head-fixed mice expressing GCaMP6f in AuC. Mice were presented with pure tones and natural pup calls—a well-characterized, ethologically relevant ultrasonic vocalization (Fig. 1 a) 14,15 . While A1 responded robustly to both 63-kHz tones (matching the dominant frequency of the pup calls) and the calls themselves, we identified a discrete region, termed Communicative Auditory Field (CAF), that was selectively activated by pup calls, with minimal responses to pure tones (Fig. 1 b). We next mapped the spatial organization of CAF relative to established auditory fields. Wide-field imaging across multiple mice confirmed that CAF is localized between AAF and UF, outside the classically defined tonotopic gradients of A1 (Fig. 1 c and Extended Data Fig. 1 ). To resolve CAF’s functional properties at cellular resolution, we performed two-photon calcium imaging of neuronal responses to pure tones (54, 63, and 80 kHz) and pup calls (Fig. 1 d). Population analysis revealed a striking preference for natural vocalizations: 57% of imaged CAF neurons responded to pup calls, compared to only 11–13% that responded to any of the pure tones (Fig. 1 e). Collectively, these data establish CAF as a discrete cortical module preferentially tuned to process complex communicative syllables over simple acoustic features. CAF neurons are tuned to syllables. We next investigated whether CAF neurons display selectivity for 11 distinct syllable types (Fig. 2 a), a critical feature for decoding complex vocalization 16 . Using two-photon calcium imaging, we recorded from CAF neurons in awake mice during presentation of categorized ultrasonic syllables (Extended Data Figs. 2 a,b). Individual CAF neurons exhibited distinct response profiles (Fig. 2 b) and showed preferential tuning on each syllable type (Fig. 2 D). Each syllable type has a subpopulation of CAF neurons preferentially tuned (Fig. 2 c and Extended Data Fig. 2 c). We then asked whether syllable-tuned neurons are organized into spatially discrete clusters. Analysis of their spatial distribution revealed that neurons responsive to different syllables were intermingled throughout CAF, showing no apparent topographic organization (Fig. 2 d). The observed overlap between populations tuned to different syllables was not greater than that expected by chance, as confirmed by comparison to shuffled data (Fig. 2 e). Consistent with this, the spatial distribution of neurons responsive to multiple syllables was indistinguishable from that of the general responsive population (Fig. 2 f). Together, these findings demonstrate that CAF contains a distributed functional code for syllables, where selectivity for specific vocalization elements is maintained without spatial clustering. CAF inhibition impairs communicative behaviors. To investigate the functional role of CAF neurons in communicative behaviors, we employed chemogenetic inhibition using the hM4Di-DREADD system. Effective suppression of CAF neuronal activity was validated in acute brain slices, where DCZ application significantly reduced firing rates (Extended Data Fig. 3 ). We focused behavioral assays on two key auditory communication tasks: pup retrieval and courtship behavior. In the pup retrieval task, postpartum female mice with CAF expressing hM4Di were tested in a Y-maze, where recorded pup calls and narrow-band noise (NBN) controls were randomly played from the two side arms (Figs. 2 g,h). Chemogenetic inhibition of CAF significantly reduced pup call-directed responses in the hM4Di group (52%) compared to the mCherry controls (83%; Fig. 3 C). Additionally, the hM4Di group exhibited prolonged retrieval latency and lower correct response rates (Figs. 2 i-k). During courtship behavior assays, where male USSs facilitate mating (Extended Data Figs. 4 a,b), CAF inhibition significantly impaired female receptivity. This was evidenced by reduced time spent in lordosis (9% in hM4Di group vs. 26% in controls), fewer lordosis events, and increased rejection behaviors (Extended Data Figs. 4 c-e). These results underscore the necessity of CAF for mediating auditory-driven communicative behaviors. CAF receives input from POL. To elucidate afferent inputs to CAF, we performed retrograde tracing using scAAV2/2-Retro-tdTomato (Fig. 3 a). Within the thalamus, the predominant fugal source of sensory input to cortex, over 80% of CAF-projecting neurons originated from the posterior limiting nucleus (POL), with additional contributions from the posterior intralaminar thalamic nucleus (PIL) (Fig. 3 b). In contrast, A1 primarily received input from the ventral division of the medial geniculate body (MGBv), consistent with its role in tonotopic processing 11 . Cortical input to CAF originated from auditory-related areas including temporal association cortex (TeA) and ectorhinal cortex (ECT) (Extended Data Fig. 5). Fiber photometry recordings in POL neurons expressing GCaMP6f revealed stronger responses to complex syllables (Chevron and Multi-steps) compared to pure tones (Figs. 3 c-e). Using monosynaptic rabies virus tracing from POL neurons projecting to CAF (Figs. 3 f,g and Extended Data Fig. 6), we found that this circuit receives input predominantly from the external (ICe, ~ 60%) and dorsal (ICd, ~ 30%) cortices of the inferior colliculus, with minor contributions from the central nucleus (ICc, ~ 7%; Figs. 3 h,i). These findings establish POL as the major thalamic relay providing non-lemniscal input to CAF. POL→CAF inhibition disrupts syllable-mediated communications. To investigate the functional relevance of the POL→CAF circuit, we selectively inhibited CAF-projecting POL neurons by expressing hM4Di specifically in POL neurons projecting to CAF using a retrograde-Cre approach, while monitoring CAF neurons simultaneously expressing GCaMP6f using two-photon calcium imaging in awake mice or examining communicative behaviors (Fig. 4 a). Inhibition of the POL→CAF circuit significantly reduced Ca 2+ responses to chevron and multi-steps syllables (Fig. 4 b). Quantitative analysis of GCaMP6f signals revealed a significant decrease in response amplitude following POL inhibition (Figs. 4 c,d). Behaviorally, inhibition of the POL→CAF circuit impaired both pup retrieval and courtship behaviors. In the pup retrieval task (Fig. 4 e), inhibition reduced response probability (57% in hM4Di group vs. 85% in controls), increased retrieval latency, and decreased correct response rates (Figs. 4 f,g). Similarly, courtship behavior was disrupted, with reduced lordosis time (8% in hM4Di group vs. 26% in controls), fewer lordosis events, and increased rejections (Extended Data Fig. 7). These findings demonstrate that POL→CAF circuit is specifically required in processing complex syllables and guiding communicative behaviors. POL→CAF operates independently of the canonical auditory pathway. To determine whether CAF-mediated syllable processing depends on the primary auditory pathway, we inhibited A1 while monitoring CAF neuronal responses (Extended Data Fig. 8a). Chemogenetic silencing of A1 neurons had no effect on CAF responses to complex syllables (Extended Data Figs. 8b). Quantitative analysis confirmed no significant change in response amplitudes in CAF neurons following A1 inhibition (Figs. 8c,d). These results demonstrate that CAF processing of complex syllables is independent of the tonotopically organized A1, supporting the existence of a parallel IC→POL→CAF circuit dedicated to complex sound processing and communicative behavior (Fig. 4 h). DISCUSSION In this study, we identified the CAF, a previously uncharacterized region of the mouse AuC that is selectively tuned to natural, complex syllabic vocalizations and is essential for guiding communicative behaviors. Our results establish that CAF’s selectivity for socially relevant vocalizations emerges from a noncanonical auditory pathway: IC shell→ POL → CAF, which is distinct from the classical lemniscal route (ICc→MGBv→A1), indicating a dedicated cortical module for processing the temporal and semantic features of complex communication sounds. This circuit also demonstrates functional necessity: chemogenetic inhibition of either CAF or POL CAF selectively disrupts socially relevant behaviors, including pup retrieval and courtship. CAF’s major thalamic input originates from POL, a subdivision of the non-classical auditory thalamus whose functional role has remained enigmatic 17 . Recent anatomical work by Liu et al. demonstrated that somatostatin-expressing (SOM⁺) neurons within the IC selectively innervate POL, which itself integrates polymodal inputs from midbrain structure involved in sensorimotor and defensive behaviors, such as the superior colliculus and periaqueductal gray 18 . Our functional data now assign a clear role to this IC→POL pathway, positioning it as a critical conduit for socially salient auditory information en route to the cortex, rather than a mere modulator of arousal. The IC serves as the principal source of auditory input to higher centers, and its population activity encodes vocal signals that form the foundation for processing in the thalamus, cortex, and limbic regions 19,20 . Our transsynaptic tracing reveals that inputs to POL arise predominantly from dorsal and lateral shell regions of the IC (IC shell), rather than its tonotopically organized central nucleus (ICc). The IC shell lacks precise tonotopy but is known to receive convergent input from both ascending brainstem pathways and descending cortical projections, endowing it with a capacity for encoding complex, temporally dynamic sounds like USVs 21–23 . This aligns with emerging evidence that the IC shell contains neurons tuned to ultrasonic frequencies and is engaged in translating sensory input into adaptive social behaviors 24,25 . Notably, a recent study identified a specialized pathway from the IC to the PIL, adjacent to POL, that is essential for processing pup retrieval calls and driving maternal behavior avia hypothalamic oxytocin release 26 . Our findings on the ICe→ POL → CAF pathway thus complement and expand this evolving understanding of the IC’s role in social acoustics. Previous studies have proposed several higher-order regions within the rodent AuC, such as A2, dorsoposterior (DP), and dorsomedial (DM) fields, as candidates for processing natural sounds. These areas are characterized by broader frequency tuning and inputs from non-lemniscal thalamic or cortical sources. For instance, DM shows selective activation to courtship vocalizations 27 and A2 exhibits heightened responses to pup calls linked to maternal behavior 28 . Furthermore, a frequency-insensitive region between A1 and AAF was previously noted for its heterogeneous responses to vocalizations 10 . However, none of these studies delineated the specific region we identify as CAF. We propose this omission likely stems from differences in sound stimulus design. We used ethologically relevant pup calls that preserve the natural temporal and syllabic structure crucial for eliciting robust behavioral responses. Maternal retrieval, for example, is known to be exquisitely sensitive to the inter-syllable interval (ISI) rhythm of pup calls 29 . The use of simplified acoustic stimuli lacking this critical temporal structure in prior work may have failed to engage the CAF module, highlighting the importance of ethological stimulus design in functional mapping 30–33 . The organization of communicative brain regions has been extensively studied in the human auditory system 34 . Beyond the primary auditory cortex (Heschl's gyrus), which processes basic acoustic features, the superior temporal gyrus (STG) contains subregions specialized for complex speech-related features. Notably, patients with lesions to Heschl’s gyrus can retain intact speech perception, indicating that the STG operates as a distinct, high-order module. The existence of a similarly parallel and distributed organization in the rodent brain, with a dedicated CAF region for complex social vocalizations operating alongside the A1, suggests an evolutionarily conserved strategy for efficient decoding of communication sounds. This convergence implies that fundamental principles of hierarchical and parallel processing may underline complex sound perception across mammals. In summary, our findings support a modular and parallel organizational principle within the mouse auditory cortex. We define CAF as a specialized region dedicated to decoding complex, socially salient vocal cues via a non-canonical thalamocortical pathway from the IC shell through POL. This functional segregation, reminiscent of specialized pathways in the human auditory system, enables the efficient processing of ethologically critical communication sounds. METHODS Animals All animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) guidelines at Chongqing University, Tsinghua University, and CIBR. CBA/J mice (8–12 weeks old) of both sexes were used. Both males and females were used for non-behavioral experiments; only females were used for pup retrieval task. Mice were housed under a 12 h light/dark cycle (lights on at 7:00 a.m.) with ad libitum access to food and water unless otherwise specified. Virus and fluorescent indicator injections All surgical procedures were conducted under isoflurane anesthesia (3% for induction; 1.5% for maintenance) in oxygen-enriched air. Mice were positioned in a stereotaxic apparatus, and ophthalmic ointment was applied to prevent corneal drying. Body temperature was maintained at 36.5–37.5°C with a heating pad. After shaving and disinfecting the scalp, the skull was exposed via a midline incision. For viral delivery, small craniotomies were made above the targeted regions. In widefield and two-photon calcium imaging experiments, 300 nL of AAV2/9-hSyn-GCaMP6f (5.0 × 10¹² viral particles/mL) was pressure-injected into the left AuC using a glass pipette. Retrograde tracing of CAF-projecting neurons was performed using retro-scAAV-tdTomato. For behaviorally targeted inhibition of the POL, CBA/J mice received bilateral injections of AAV2/9-DIO-hM4Di-mCherry or control AAV2/9-DIO-mCherry into the CAF, along with bilateral injections of AAV2/2-Retro-Cre into the CAF. For A1 inhibition experiments, AAV2/9-hSyn-hM4Di-mCherry or its control was bilaterally injected into A1. Monosynaptic rabies tracing was used to identify upstream inputs to POL neurons projecting to CAF. To this end, helper viruses (AAV2/5-Ef1a-DIO-EGFP-T2A-TVA and AAV2/5-Ef1a-DIO-RVG) were injected into POL, while AAV2/2-Retro-Cre was delivered to the CAF. Three weeks later, EnvA-pseudotyped rabies virus were injected into the CAF. CAF was targeted based on widefield imaging. POL was injected at three stereotaxic coordinates (relative to bregma, from dura) to cover its elongated structure: − 3.05 AP, ± 1.87 ML, − 2.80 DV; − 3.05 AP, ± 1.81 ML, − 3.23 DV; and − 3.05 AP, − 1.65 ML, − 3.60 DV. Wide-field fluorescence imaging To locate the position of CAF, we performed widefield calcium imaging (2-CAFE, NewLight Optoinstrument; www.newlightxhr.com) equipped with a 4×, 0.2 NA objective (Olympus). Cortical fluorescence was excited by a 470 nm LED (Thorlabs M470L4), and green emission was captured at 10 Hz using a high-speed sCMOS camera (Zyla 4.2, Andor Technology) through a standard filter cube. A 3 mm diameter cranial window was implanted over the auditory cortex (AuC) following a craniotomy. For signal quantification, fluorescence intensity was normalized to the baseline (f₀), defined as the average signal during the 1 s preceding stimulus presentation. Δf/f was calculated as (f – f₀)/f₀ and averaged across trials. Resulting Δf/f maps were visualized using a pseudocolor scale to reveal sound-evoked responses across cortical regions. For widefield image, tonotopic maps were generated from widefield calcium imaging data using custom MATLAB scripts. Images were downsampled to 150×150 pixels for analysis. For each stimulus frequency, frequency-specific response regions were identified based on calcium signals exceeding two standard deviations above baseline. The best frequency (BF) for each pixel was computed as the weighted average of responses across all frequencies, following a previously established method 10 . Two-photon Ca imaging Four weeks after viral injection, a small cranial window was implanted over the left AuC. A custom-made plastic imaging chamber was affixed to the skull using cyanoacrylate glue (UHU), and a circular portion of the skull (~ 3 mm diameter) was removed and replaced with glass coverslip. In vivo imaging was carried out using a LOTOS2 two-photon microscope (New Light Opto lnstrument Co., Ltd; see: www.newlightxhr.com), which equipped with a 12.0 kHz resonant scanner (model details, Spectra-Physics DeepSee Mai Tai). Excitation was provided by a mode-locked Ti:sapphire laser tuned to 920 nm. A 40× objective with a 0.8 numerical aperture (Nikon) was used to acquire images. For cellular imaging of L2/3 neurons, the field of view was typically 400 × 400 µm. Laser power at the brain surface ranged from 30 to 120 mW, depending on imaging depth. For two-photo imaging data analysis, calcium signals were analyzed from manually defined regions of interest (ROIs) based on fluorescence intensity. We calculated ΔF/F relative to the average baseline fluorescence within each ROI. A neuron was considered responsive if its average response across all trials also exceeded 3× SD within 2 s post-stimulation. Data were analyzed using custom scripts in LabVIEW 2016, MATLAB 2020a, and GraphPad Prism 9.0. Optical fiber recording For Ca 2+ indicator expression and fiber optic implantation in CAF-projecting POL neurons, we injected AAV2/2-Retro-Cre into the CAF and AAV2/8-Dio-GCamp6f into the POL, delivering 200 nL. A 200 mm fiber optic with an attached cannula (R-FOC-BL-200C-39N, RWD) was then implanted 2.8 mm deep above the injection site. Auditory stimulation For all calcium imaging experiments, including two-photon imaging, widefield fluorescence imaging, and fiber photometry, auditory stimuli were presented using consistent temporal and acoustic parameters across conditions. Pure-tone stimuli were generated using a custom program developed in LabVIEW 2012 (National Instruments) and converted to analog voltages via a PCI-6731 card (National Instruments). Sounds were delivered through an ED1 electrostatic speaker driver and a free-field ES1 speaker (Tucker-Davis Technologies) 35,36 . Syllable-based stimuli were derived from ultrasonic vocalizations (USVs) recorded from freely behaving mice using the Avisoft RECORDER system. Based on previous study 16 , we identified 11 distinct syllable types. During each trial, a single syllable was repeatedly played for 2 seconds, preserving its natural inter-syllable interval (ISI) ranging from 100 to 275 ms 29,33 . All stimuli—whether pure tone or syllable-based—were presented at 65 dB SPL and repeated 10 times per condition. The 5-second cycle (2-s stimulation + 3-s silence) ensured sufficient time for calcium signals to return to baseline between trials. Pup retrieval behavior This experiment utilized a customized Y-maze apparatus, comprising a 40×30×30 cm central hall connected to two 15×15×30 cm symmetrical side chambers. A 10×10 cm nest area was positioned at the center of the rear wall in the central region, with each side chamber equipped with an UltraSoundGate 116H ultrasonic speaker. Bedding was laid at the bottom, and water and food were provided for the female mice to access freely during the experiment. The experimental subjects were female CBA/J mice 5–8 days postpartum. One day prior to the test, original nest materials were transferred to the nest area on the rear wall of the central region in the test cage, and nest reconstruction by the female mice was observed and confirmed. Pups were individually placed in a 20×20×20 cm soundproof box maintained at 30°C with a heating pad. Ultrasonic vocalizations (USVs) were recorded for 5 minutes per session using an Avisoft-UltraSoundGate CM16 microphone (15 cm above the box bottom) at a sampling rate of 250 kHz with 16-bit precision. Thirty minutes before the test, female mice were intraperitoneally injected with DCZ (0.1 mg/kg body weight) or physiological saline. During the behavioral test, the female mice were first allowed to acclimate to the central region for 5 minutes. A pup was then randomly introduced into the left or right chamber, and the time taken by the female mice to retrieve the pup was recorded. After the female mice remained quiet in the nest for 10 seconds, the pup’s vocalizations filtered to 20–100 kHz (USV group) or 50–75 kHz narrow-band noise (NBN group) were randomly played from the left or right speaker. To generate narrow-band noise (NBN) as control, each USV syllable was temporally matched with a band-limited noise segment (50–75 kHz) of the same duration. The amplitude of each NBN segment was adjusted to match the envelope of its corresponding syllable. This preserved the temporal envelope of the vocalizations while disrupting spectral structure. The trial ended when the female mouse fully entered the side chamber (defined as making a choice) and was terminated if no choice was made within 4 minutes. Videos were recorded using a SONY FDR-AX60.The latencies to correct choices, probabilities of correct choices, and probabilities of making choices by the mice were statistically analyzed using GraphPad Prism. Courtship behavior This experiment utilized a 30×40×40 cm transparent observation chamber partitioned by a removable divider, conducted in a quiet environment with constant temperature (22 ± 2°C) and humidity (55 ± 5%). Mating behavior tests were performed in CBA/J mice aged 2–3 months. Male mice were pre-screened to confirm at least one prior successful mating, while female mice were virgins. Estrus was determined by vaginal lavage: secretions were stained with Giemsa, observed under a bright-field microscope, and females with ≥ 70% cornified epithelial cells were classified as estrous. Tests began 1 hour after the onset of the dark phase in the mice's housing room. Thirty minutes before testing, mice received intraperitoneal injections of DCZ (0.1 mg/kg body weight) or saline. Estrous females and sexually experienced males were acclimated separately on either side of the divider for 10 minutes. After removing the divider, interactions were observed for 30 minutes, with simultaneous video and audio recording using an RMONCAM-S900H USB camera (overhead view), SONY FDR-AX60 camera (frontal view), and an Avisoft-UltraSoundGate CM16 microphone (30 cm above the floor, 250 kHz sampling rate). A custom MATLAB program was used for frame-by-frame manual annotation of frontal-view videos, analyzing the 10-minute period following the female's first mounting event. Behavioral criteria were defined as: lordosis—female hindlimb support with spinal arching during mating or intromission; mounting—male climbing onto the female's back without pelvic thrusting; rejection behaviors including three subtypes: active escape (running > three body lengths), postural avoidance (lateral evasion), and defensive posture (forelimb pushing). Data analysis All results are presented as mean ± SD. Specific experimental details and animal numbers are indicated in the figure panels and legends. For comparisons between groups, the two-sided Mann–Whitney U test was used; for paired observations, the two-sided Wilcoxon signed-rank test was applied. When evaluating more than two groups, the Kruskal–Wallis test was conducted, followed by Dunn’s post hoc test for pairwise comparisons when overall significance was detected. Statistical significance in the figures is denoted as follows: ns (p > 0.05), * (p < 0.05), ** (p < 0.01),, *** (p < 0.001) and **** (p < 0.0001). Declarations Acknowledgments The authors thank Fenghua Liu and Yu Wang for laboratory management. We also thank the Laboratory Animal Resource Center of Tsinghua University, CIBR, and Chongqing University for maintenance of mice; the Imaging Core of CIBR for assistance in imaging. W.X. discloses support for the research described in this study from China Ministry of Science and Technology (2021ZD0203304), National Natural Science Foundation of China (32300833 to Z.Z.; 32430044 to X.C.; U23A20442 to W.X.; 32300830 to S.L.), and Shenzhen Medical Research Fund (B2402008). Author contributions: In vivo imaging: Z.Z.; Data analysis: X.T., Y-Z.W., and S.L.; Mouse surgery, viral injection, and histology: Z.Z. and Y-L.W.; Mouse behavior: Z.Z., Y-L.W., M.P., and J.L.; Brain slice recording: J.T.; Photometry: Z.Y.; Mouse breeding: Z.Z., Y.-L.W., and Q.L.; Experiment design and supervision: Z.Z., X.C., and W.X.; Manuscript writing: Z.Z. and W.X.. Competing interests W.X. is a co-founder of SimpGen Therapeutics. This relationship did not influence this study. The other authors declare no competing interests. References Kanold, P. O., Nelken, I. & Polley, D. B. Local versus global scales of organization in auditory cortex. Trends Neurosci 37 , 502-510 (2014). https://doi.org/10.1016/j.tins.2014.06.003 Funkhouser, E. B. The visual cortex, its location, histological structure, and physiological function. J Exp Med 21 , 617-628 (1915). https://doi.org/10.1084/jem.21.6.617 Beltramo, R. & Scanziani, M. A collicular visual cortex: Neocortical space for an ancient midbrain visual structure. Science 363 , 64-69 (2019). https://doi.org/10.1126/science.aau7052 Chen, X., Gabitto, M., Peng, Y., Ryba, N. J. & Zuker, C. S. A gustotopic map of taste qualities in the mammalian brain. Science 333 , 1262-1266 (2011). https://doi.org/10.1126/science.1204076 Rothschild, G., Nelken, I. & Mizrahi, A. Functional organization and population dynamics in the mouse primary auditory cortex. Nat Neurosci 13 , 353-360 (2010). https://doi.org/10.1038/nn.2484 Tasaka, G. I. et al. Genetic tagging of active neurons in auditory cortex reveals maternal plasticity of coding ultrasonic vocalizations. Nat Commun 9 , 871 (2018). https://doi.org/10.1038/s41467-018-03183-2 Zucca, S., La Rosa, C., Fellin, T., Peretto, P. & Bovetti, S. Developmental encoding of natural sounds in the mouse auditory cortex. Cereb Cortex 34 (2024). https://doi.org/10.1093/cercor/bhae438 Rauschecker, J. P. Cortical processing of complex sounds. Curr Opin Neurobiol 8 , 516-521 (1998). https://doi.org/10.1016/s0959-4388(98)80040-8 Scott, S. K. & Johnsrude, I. S. The neuroanatomical and functional organization of speech perception. Trends Neurosci 26 , 100-107 (2003). https://doi.org/10.1016/s0166-2236(02)00037-1 Issa, J. B. et al. Multiscale optical Ca2+ imaging of tonal organization in mouse auditory cortex. Neuron 83 , 944-959 (2014). https://doi.org/10.1016/j.neuron.2014.07.009 Hackett, T. A., Barkat, T. R., O'Brien, B. M., Hensch, T. K. & Polley, D. B. Linking topography to tonotopy in the mouse auditory thalamocortical circuit. J Neurosci 31 , 2983-2995 (2011). https://doi.org/10.1523/JNEUROSCI.5333-10.2011 Tsukano, H. et al. Reconsidering Tonotopic Maps in the Auditory Cortex and Lemniscal Auditory Thalamus in Mice. Front Neural Circuits 11 , 14 (2017). https://doi.org/10.3389/fncir.2017.00014 Nakata, S., Takemoto, M. & Song, W. J. Differential cortical and subcortical projection targets of subfields in the core region of mouse auditory cortex. Hear Res 386 , 107876 (2020). https://doi.org/10.1016/j.heares.2019.107876 Lecca, S. et al. A neural substrate for negative affect dictates female parental behavior. Neuron 111 , 1094-1103.e1098 (2023). https://doi.org/10.1016/j.neuron.2023.01.003 Burenkova, O. V., Averkina, A. A., Aleksandrova, E. A. & Zarayskaya, I. Y. Brief but enough: 45-min maternal separation elicits behavioral and physiological responses in neonatal mice and changes in dam maternal behavior. Physiol Behav 222 , 112877 (2020). https://doi.org/10.1016/j.physbeh.2020.112877 Fonseca, A. H., Santana, G. M., Bosque Ortiz, G. M., Bampi, S. & Dietrich, M. O. Analysis of ultrasonic vocalizations from mice using computer vision and machine learning. Elife 10 (2021). https://doi.org/10.7554/eLife.59161 Márquez-Legorreta, E., Horta-Júnior Jde, A., Berrebi, A. S. & Saldaña, E. Organization of the zone of transition between the pretectum and the thalamus, with emphasis on the pretectothalamic lamina. Front Neuroanat 10 , 82 (2016). https://doi.org/10.3389/fnana.2016.00082 Liu, M. et al. Parvalbumin and Somatostatin: Biomarkers for Two Parallel Tectothalamic Pathways in the Auditory Midbrain. J Neurosci 44 (2024). https://doi.org/10.1523/JNEUROSCI.1655-23.2024 Woolley, S. M. & Portfors, C. V. Conserved mechanisms of vocalization coding in mammalian and songbird auditory midbrain. Hear Res 305 , 45-56 (2013). https://doi.org/10.1016/j.heares.2013.05.005 Lyzwa, D. & Wörgötter, F. Neural and response correlations to complex natural sounds in the auditory midbrain. Front Neural Circuits 10 , 89 (2016). https://doi.org/10.3389/fncir.2016.00089 Chen, C., Cheng, M., Ito, T. & Song, S. Neuronal organization in the inferior colliculusrevisited with Cell-type-dependent monosynaptic tracing. J Neurosci 38 , 3318-3332 (2018). https://doi.org/10.1523/jneurosci.2173-17.2018 Lesicko, A. M., Hristova, T. S., Maigler, K. C. & Llano, D. A. Connectional modularity of Top-down and Bottom-up multimodal inputs to the lateral cortex of the mouse inferior colliculus. J Neurosci 36 , 11037-11050 (2016). https://doi.org/10.1523/jneurosci.4134-15.2016 Casseday, J. H. & Covey, E. A neuroethological theory of the operation of the inferior colliculus. Brain Behav Evol 47 , 311-336 (1996). https://doi.org/10.1159/000113249 Shi, K. et al. Population coding of time-varying sounds in the nonlemniscal inferior colliculus. J Neurophysiol 131 , 842-864 (2024). https://doi.org/10.1152/jn.00013.2024 Holmstrom, L. A., Eeuwes, L. B., Roberts, P. D. & Portfors, C. V. Efficient encoding of vocalizations in the auditory midbrain. J Neurosci 30 , 802-819 (2010). https://doi.org/10.1523/JNEUROSCI.1964-09.2010 Valtcheva, S. et al. Neural circuitry for maternal oxytocin release induced by infant cries. Nature 621 , 788-795 (2023). https://doi.org/10.1038/s41586-023-06540-4 Tsukano, H. et al. Delineation of a frequency-organized region isolated from the mouse primary auditory cortex. J Neurophysiol 113 , 2900-2920 (2015). https://doi.org/10.1152/jn.00932.2014 Chong, K. K., Anandakumar, D. B., Dunlap, A. G., Kacsoh, D. B. & Liu, R. C. Experience-dependent coding of time-dependent frequency trajectories by off responses in secondary auditory cortex. J Neurosci 40 , 4469-4482 (2020). https://doi.org/10.1523/jneurosci.2665-19.2020 Schiavo, J. K. et al. Innate and plastic mechanisms for maternal behaviour in auditory cortex. Nature 587 , 426-431 (2020). https://doi.org/10.1038/s41586-020-2807-6 Gaub, S. & Ehret, G. Grouping in auditory temporal perception and vocal production is mutually adapted: the case of wriggling calls of mice. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 191 , 1131-1135 (2005). https://doi.org/10.1007/s00359-005-0036-y Ehret, G. Infant rodent ultrasounds -- a gate to the understanding of sound communication. Behav Genet 35 , 19-29 (2005). https://doi.org/10.1007/s10519-004-0853-8 Uematsu, A. et al. Maternal approaches to pup ultrasonic vocalizations produced by a nanocrystalline silicon thermo-acoustic emitter. Brain Res 1163 , 91-99 (2007). https://doi.org/10.1016/j.brainres.2007.05.056 Castellucci, G. A., Calbick, D. & McCormick, D. The temporal organization of mouse ultrasonic vocalizations. PLoS One 13 , e0199929 (2018). https://doi.org/10.1371/journal.pone.0199929 Hamilton, L. S., Oganian, Y., Hall, J. & Chang, E. F. Parallel and distributed encoding of speech across human auditory cortex. Cell 184 , 4626-4639.e4613 (2021). https://doi.org/10.1016/j.cell.2021.07.019 Li, J. et al. PIEZO2 mediates ultrasonic hearing via cochlear outer hair cells in mice. Proc Natl Acad Sci U S A 118 , e2101207118 (2021). https://doi.org/10.1073/pnas.2101207118 Li, J. et al. Prestin-Mediated Frequency Selectivity Does not Cover Ultrahigh Frequencies in Mice. Neurosci Bull 38 , 769-784 (2022). https://doi.org/10.1007/s12264-022-00839-4 Additional Declarations Yes there is potential Competing Interest. W.X. is a co-founder of SimpGen Therapeutics. This relationship did not influence this study. The other authors declare no competing interests. Supplementary Files CAFfiguresED.pdf Extended Data Figure Legends Extended Data Fig. 1 | Tonotopic organization confirms CAF localization. a, Auditory cortical maps in 6 additional mice via wide-field imaging. Frequency gradients (4-80 kHz) demarcate primary regions (A1, AAF, UF), with CAF consistently localized outside tonotopic gradients. Extended Data Fig. 2 | Two-photon Ca 2+ responses of CAF neurons. a, Example two-photon imaging plane in CAF. b, Calcium traces from representative neurons responding to distinct syllables. c, Distribution of CAF neurons by response amplitude on syllable types. Neurons sorted by peak response amplitude to their preferred syllable (ΔF/F, normalized). Extended Data Fig. 3 | hM4Di-DREADD of CAF neurons a, Schematic of AAV2/9-DIO-hM4Di injection in CAF for electrophysiology. b, Example current-evoked action potentials in a CAF neuron before, during, and after DCZ application. c, DCZ significantly reduced firing rates vs. ACSF (**p < 0.01; Kruskal-Wallis test; n = 6 cells from 3 mice). Extended Data Fig. 4 | hM4Di-DREADD of CAF inhibition disrupts courtship behaviors. a, Schematic of AAV2/9-DIO-hM4Di injection in CAF for courtship behaviors. b, Courtship behavior schematic. c, Lordosis duration. Proportion of time in lordosis posture (10-min window) in control vs. hM4Di groups. d, Lordosis frequency. Fewer lordosis events in hM4Di group vs. controls (****p < 0.0001, Mann-Whitney test; control: n = 9 mice, hM4Di: n = 8 mice). e, Rejection frequency. Increased rejection behaviors in hM4Di group vs. controls (****p < 0.0001; same group as H). Extended Data Fig. 5 | Cortical inputs to CAF. a, Retrograde-traced tdTomato + neurons in AuC providing input to CAF (scale bar: 200 μm). b, Quantification of ipsilateral (red) vs. contralateral (black) cortical inputs to CAF. Data normalized to total labeled cells; n = 6 mice. AuC, Auditory areas; SS, Somatosensory areas; MO, Somatomotor areas; TeA, Temporal association areas; ECT, Ectorhinal area; RHP, Retrohippocampal region; VIS, Visual areas; VISC, Visceral area; PTLp, Posterior parietal association areas; AI, Agranular insular area; ACA, Anterior cingulate area; OLF, Olfactory areas; PL, Prelimbic area; ILA, Infralimbic area; GU, Gustatory areas; PERI, Perirhinal area; RSP, Retrosplenial area; CLA, Claustrum; BLA, Basolateral amygdala nucleus; VENT, Ventral group of the dorsal thalamus; LAT, Lateral group of the dorsal thalamus; ILM, Intralaminar nuclei of the dorsal thalamus; GEND, Geniculate group, dorsal thalamus; HB, Hindbrain; MB, Midbrain. Extended Data Fig. 6 | Brain-wide inputs to POL CAF . a, Retrograde-traced presynaptic neurons for POL (scale bar: 200 μm). b, Ipsilateral (red) vs. contralateral (black) input proportions to POL. n = 5 mice. TeA , Temporal association areas; VISC, Visceral area; AuC, Auditory areas; GU, Gustatory areas; SS, Somatosensory areas; MO, Somatomotor areas; VIS, Visual areas; ACA, Anterior cingulate area; AI, Agranular insular area; RSP, Retrosplenial area; ECT, Ectorhinal area; PIR, Piriform area; STR, Striatum; PALd, Pallidum dorsal region; DORsm, Thalamus, sensory-motor cortex related; DORpm Thalamus, polymodal association cortex related; HY, Hypothalamus; P-sen, Pons, sensory related; P-mot, Pons, motor related; P-sat, Pons, behavioral state related; MY-mot, Medulla, motor related. Extended Data Fig. 7 | hM4Di-DREADD of POL→CAF inhibition disrupts courtship behaviors. a, POL→CAF inhibition strategy by injecting AAV2/9-DIO-hM4Di in POL and AAV2/2-Retro-Cre in CAF. b, Lordosis duration. Time in lordosis posture (10 min window; control vs. hM4Di). c, Lordosis frequency. Reduced lordosis events in hM4Di group (***p < 0.001; Mann-Whitney test; control: n=7; hM4Di: n = 9 mice). d, Rejection frequency. Increased rejection frequency in hM4Di group (***p < 0.001; Mann-Whitney test; control: n=7; hM4Di: n = 9 mice). Extended Data Fig. 8 | A1 inhibition does not disrupt syllable processing in CAF. a, Schematics of experimental design for inhibition and imaging. Left: Chemogenetic inhibition of A1 during CAF recording. Right: Imaging setup. b, CAF responses under A1 inhibition. Left: Two-photon imaging of CAF neurons. Right: Calcium traces to syllables (8 trials). c, Population response profile. Heatmap of neurons sorted by response amplitude (one plane). d, No effect on CAF activity. Unchanged ΔF/F during A1 inhibition (ns, p > 0.5; one-sample Wilcoxon test). Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8163343","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":555928847,"identity":"e3b5b9dc-9c02-4ab6-8c3c-492b3acd83f7","order_by":0,"name":"Wei Xiong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIiWNgGAWjYDADPgYexgdAWoaBgY1ILWwMPMwGDAwGPCRpYZMgSos5e4/Zhw81NtFs7GePVXxs+8PDz96WwPCjYhtOLZY9Z4xnzjiWltvGk5d2c2abAY9kz7EDjD1nbuPUYnAjx5iZt+FwbhtDjtltXqAWgxvpDcyMbXi03H8D1cL/xqyYOC03eKBaJHLMmCFa0g7g13ImrZgR7BeJd8mSM84Zg/yScBCvX44f3swADLHcfv7cgx8+lMnJAUPM8MGPCtxasIMDJKofBaNgFIyCUYAGAGykUMl+lOK8AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-2784-7696","institution":"Chinese Institute for Brain Research, Beijing","correspondingAuthor":true,"prefix":"","firstName":"Wei","middleName":"","lastName":"Xiong","suffix":""},{"id":555928848,"identity":"31eaf7d2-26c5-4e05-8be8-79ee3dc35310","order_by":1,"name":"Zhikai Zhao","email":"","orcid":"","institution":"Chongqing University","correspondingAuthor":false,"prefix":"","firstName":"Zhikai","middleName":"","lastName":"Zhao","suffix":""},{"id":555928849,"identity":"1e0b3b24-9b4b-426a-85d3-6fdbd6c3d110","order_by":2,"name":"Xiaojing Tang","email":"","orcid":"","institution":"Chongqing University","correspondingAuthor":false,"prefix":"","firstName":"Xiaojing","middleName":"","lastName":"Tang","suffix":""},{"id":555928850,"identity":"65f50d13-36c2-4df4-94bb-452825e03109","order_by":3,"name":"Yunlong Wu","email":"","orcid":"","institution":"Chinese Institute for Brain Research, Beijing","correspondingAuthor":false,"prefix":"","firstName":"Yunlong","middleName":"","lastName":"Wu","suffix":""},{"id":555928851,"identity":"6c35a90e-4b30-48a3-b6ba-5332d5036a28","order_by":4,"name":"Yangzhen Wang","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Yangzhen","middleName":"","lastName":"Wang","suffix":""},{"id":555928852,"identity":"9c6607a5-3e0b-445c-8420-ab3ea92a8335","order_by":5,"name":"Shuang Liu","email":"","orcid":"","institution":"Chinese Institute for Brain Research, Beijing (CIBR), Beijing, China","correspondingAuthor":false,"prefix":"","firstName":"Shuang","middleName":"","lastName":"Liu","suffix":""},{"id":555928853,"identity":"33d95223-ec2c-4417-b7cd-98252bb32461","order_by":6,"name":"Mahiber Polat","email":"","orcid":"","institution":"Chongqing University","correspondingAuthor":false,"prefix":"","firstName":"Mahiber","middleName":"","lastName":"Polat","suffix":""},{"id":555928854,"identity":"1d68c529-fcc0-45b2-aeb5-e48d94f3cc2a","order_by":7,"name":"Jie Tao","email":"","orcid":"","institution":"Guangxi University","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Tao","suffix":""},{"id":555928855,"identity":"f3aed58d-750e-4e0e-9f0b-95c49ba8a165","order_by":8,"name":"Zhiqi Yang","email":"","orcid":"","institution":"Third Military Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhiqi","middleName":"","lastName":"Yang","suffix":""},{"id":555928856,"identity":"cccc515a-ea70-4818-b55e-1f304f5108e8","order_by":9,"name":"Jie Li","email":"","orcid":"","institution":"School of Life Sciences, Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Li","suffix":""},{"id":555928857,"identity":"d2bdf473-c513-4c85-86fc-6ad8592edef0","order_by":10,"name":"Qingling Liu","email":"","orcid":"","institution":"Chinese Institute for Brain Research, Beijing","correspondingAuthor":false,"prefix":"","firstName":"Qingling","middleName":"","lastName":"Liu","suffix":""},{"id":555928858,"identity":"0bb45095-f4a6-43f1-b492-51c22822f46a","order_by":11,"name":"Xiaowei Chen","email":"","orcid":"https://orcid.org/0000-0003-0906-6666","institution":"Third Military Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiaowei","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2025-11-20 10:19:03","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8163343/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8163343/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":99223494,"identity":"878f1ca2-d20f-4768-a911-cd246221e947","added_by":"auto","created_at":"2025-12-30 10:00:56","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":398229,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA cortical subregion (CAF) preferentially processes ultrasonic social syllables.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Wide-field imaging reveals syllable-evoked cortical activity. Top: Auditory stimulus-evoked cortical activity (GCaMP6f). Bottom: calcium traces from a responsive subregion.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e CAF selectivity for natural pup calls. Top: Wide-field maps show A1 responses to 63 kHz tones and pup calls, versus CAF’s selective responses to pup calls. Middle: Magnified views of A1 and CAF with representative ΔF/F traces. Bottom: Pup calls evoke stronger responses in CAF than tones (n = 7 mice; **p \u0026lt; 0.01, Mann-Whitney test).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec,\u003c/strong\u003e Spatial localization of CAF. Auditory cortical map (representative mouse; bottom-right, average of 7 recorded mice). CAF resides between ultrasonic field (UF) and anterior auditory field (AAF), distinct from tonotopically organized A1 (demarcated by 4-80 kHz gradients).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed,\u003c/strong\u003e Single-neuron validation of CAF tuning. Top: two-photon imaging in CAF during presentation of tones (54/63/80 kHz) and pup calls. Bottom: Example calcium traces from individual neurons (6 trials each).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee,\u003c/strong\u003e Population-level tuning in CAF. Top: Response amplitude of all imaged neurons (cells sorted by peak ΔF/F). Bottom: Proportions of neurons responsive to each stimulus (n = 742 neurons from 6 mice).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8163343/v1/ab1b5aa963ac7aca25bbabd0.png"},{"id":99223496,"identity":"92d2a2a5-c094-4be2-ac4f-b8f62a64f333","added_by":"auto","created_at":"2025-12-30 10:00:56","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":153799,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatially intermingled syllable tuning in CAF neurons.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea, \u003c/strong\u003eSpectrograms of 11 ultrasonic syllable (USS) types (S1: Flat, S2: Down, S3: Step down, S4: Short, S5: Step up, S6: Up, S7: Re-chevron, S8: Chevron, S9: Two steps, S10: Complex, S11: Multi-steps).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Normalized peak ΔF/F responses of representative CAF neuron to each syllable type.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec, \u003c/strong\u003eDistribution of CAF neurons by selectivity on syllable types. Heatmap of maximal responses across 914 neurons (rows) to each syllable (S1-S11 columns). Black pixels indicate the syllable evoking the highest ΔF/F for a given neuron. Totally 914 neurons from 10 focus planes of 5 mice.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed,\u003c/strong\u003e Spatial organization of syllable tuning. Locations of neurons responsive to specific syllables (S1-S11: colored dots; gray: non-responsive).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee,\u003c/strong\u003e Histogram of overlap counts in shuffled data (10,000 repetitions). Red line indicates observed count of neurons responsive to ≥2 syllables.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee,\u003c/strong\u003e Spatial dispersion analysis. Cumulative distribution functions (CDF) of distances between neuron centroids. Distributions for all neurons and overlap-responsive neurons (≥2 Syllables) were indistinguishable (Kolmogorov–Smirnov test, p = 1.0, n = 914 neurons).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg-k,\u003c/strong\u003e Pup retrieval behavior of animals with CAF disruption. Task design and inhibition\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eg\u003c/strong\u003e). Pup retrieval schematic (\u003cstrong\u003eh\u003c/strong\u003e). Response probability. Proportion of trials with successful pup retrieval (mCherry-control vs. hM4Di groups) (\u003cstrong\u003ei\u003c/strong\u003e). Retrieval latency. Increased latency in hM4Di group vs. controls (*p \u0026lt; 0.05, Mann-Whitney test; control: n = 5 mice, hM4Di: n = 6 mice) (\u003cstrong\u003ej\u003c/strong\u003e). Correct response rate. Reduced accuracy in hM4Di group vs. controls (**p \u0026lt; 0.01, Mann-Whitney test) (\u003cstrong\u003ek\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8163343/v1/dd292726c599e960ede411b3.png"},{"id":99319948,"identity":"49a60da6-ca88-4207-97a4-7ac55223f1f3","added_by":"auto","created_at":"2025-12-31 16:38:02","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":240553,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNon-lemniscal projection to CAF.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Retrograde tracing strategy. Schematic of scAAV-tdTomato injection in CAF for labeling input neurons.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Thalamic input to CAF. Left: tdTomato\u003csup\u003e+\u003c/sup\u003e neurons in auditory thalamus (scale bar: 100 μm). Right: \u0026gt;80% of CAF inputs originate from POL and secondary PIL (n = 6 mice).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec,\u003c/strong\u003e Fiber photometry setup. Schematic for Ca²⁺ recording in POL.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed,\u003c/strong\u003e POL responses to auditory stimuli. ΔF/F traces to 63-kHz tones and USSs, including Chevron and Multi-steps (representative mouse).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee,\u003c/strong\u003e Stimulus selectivity. POL responses were stronger to USSs than to 63-kHz tones (*p \u0026lt; 0.05, **p \u0026lt; 0.01; Kruskal-Wallis test; n = 7 mice).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef,\u003c/strong\u003e Transsynaptic tracing strategy. Rabies virus (RV) injection in POL for retrograde input mapping.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg,\u003c/strong\u003e RV construct design. Schematic of RV for transsynaptic labeling.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eh,\u003c/strong\u003e IC input to POL, which was shown as mCherry\u003csup\u003e+\u003c/sup\u003e neurons in IC subregions (scale bar: 500 mm and 200 mm).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ei,\u003c/strong\u003e Quantification of IC inputs. Fraction of inputs from IC subdivisions to POL (n = 5 mice).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8163343/v1/6863d22b6d9d6b622d71f119.png"},{"id":99223495,"identity":"12d29385-5e45-482d-af9a-cf395fbb48d7","added_by":"auto","created_at":"2025-12-30 10:00:56","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":227686,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePOL→CAF circuit is required for syllable processing.\u003cbr\u003e\na,\u003c/strong\u003e Schematic of experimental design for inhibition and imaging. Left: Chemogenetic inhibition of POL→CAF projections during CAF recording. Right: Wide-field and two-photon imaging setup.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Calcium traces to Chevron and Multi-steps (average from 8 trials).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec,\u003c/strong\u003e Population response profile. Heatmap of neurons sorted by response amplitude (ΔF/F; one imaging plan).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed,\u003c/strong\u003e Suppression of CAF activity. Reduced ΔF/F to syllables during POL inhibition (**p \u0026lt; 0.001, one-sample Wilcoxon test).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee,\u003c/strong\u003e Response probability of pup retrieval. Retrieval success in control (mCherry) vs. hM4Di groups.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef,\u003c/strong\u003e Retrieval latency quantification of pup retrieval. Increased latency in hM4Di group (*p \u0026lt; 0.05; Mann-Whitney test; control: n = 5, hM4Di: n = 7 mice).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eg,\u003c/strong\u003e Retrieval accuracy of pup retrieval. Reduced accuracy in hM4Di group (**p \u0026lt; 0.01; Mann-Whitney test; control: n = 5, hM4Di: n = 7 mice)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eh,\u003c/strong\u003e Model of parallel pathway for syllable processing. Canonical pathway, ICc→MGBv→A1 (tonotopic processing); syllable pathway, IC shell→POL→CAF (syllable processing; essential for social behaviors).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8163343/v1/52facf8913e576243690a448.png"},{"id":99323858,"identity":"8a52fb4f-8a07-443d-8c60-be36d5b34193","added_by":"auto","created_at":"2025-12-31 16:46:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1660181,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8163343/v1/4b8e48e3-3fa2-47e9-a313-d008f47d94c3.pdf"},{"id":99223499,"identity":"d91f5c32-e435-499b-a3db-94ca6d4fcaf5","added_by":"auto","created_at":"2025-12-30 10:00:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":30934262,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExtended Data Figure Legends\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;Extended Data Fig. 1 | Tonotopic organization confirms CAF localization.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Auditory cortical maps in 6 additional mice via wide-field imaging. Frequency gradients (4-80 kHz) demarcate primary regions (A1, AAF, UF), with CAF consistently localized outside tonotopic gradients. \u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eExtended Data Fig. 2 | Two-photon Ca\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e2+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e responses of CAF neurons.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Example two-photon imaging plane in CAF.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Calcium traces from representative neurons responding to distinct syllables.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec,\u003c/strong\u003e Distribution of CAF neurons by response amplitude on syllable types. Neurons sorted by peak response amplitude to their preferred syllable (ΔF/F, normalized).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eExtended Data Fig. 3 | hM4Di-DREADD of CAF neurons\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Schematic of AAV2/9-DIO-hM4Di injection in CAF for electrophysiology.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Example current-evoked action potentials in a CAF neuron before, during, and after DCZ application. \u003cstrong\u003ec,\u003c/strong\u003e DCZ significantly reduced firing rates vs. ACSF (**p \u0026lt; 0.01; Kruskal-Wallis test; n = 6 cells from 3 mice).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eExtended Data Fig. 4 | hM4Di-DREADD of CAF inhibition disrupts courtship behaviors.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Schematic of AAV2/9-DIO-hM4Di injection in CAF for courtship behaviors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Courtship behavior schematic.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec,\u003c/strong\u003e Lordosis duration. Proportion of time in lordosis posture (10-min window) in control vs. hM4Di groups.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed,\u003c/strong\u003e Lordosis frequency. Fewer lordosis events in hM4Di group vs. controls (****p \u0026lt; 0.0001, Mann-Whitney test; control: n = 9 mice, hM4Di: n = 8 mice).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee,\u003c/strong\u003e Rejection frequency. Increased rejection behaviors in hM4Di group vs. controls (****p \u0026lt; 0.0001; same group as H). \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;Extended Data Fig. 5 | Cortical inputs to CAF.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Retrograde-traced tdTomato\u003csup\u003e+\u003c/sup\u003e neurons in AuC providing input to CAF (scale bar: 200 μm).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb, \u003c/strong\u003eQuantification of ipsilateral (red) vs. contralateral (black) cortical inputs to CAF. Data normalized to total labeled cells; n = 6 mice. AuC, Auditory areas; SS, Somatosensory areas; MO, Somatomotor areas; TeA, Temporal association areas; ECT, Ectorhinal area; RHP, Retrohippocampal region; VIS, Visual areas; VISC, Visceral area; PTLp,\u0026nbsp; Posterior parietal association areas; AI,\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; Agranular insular area; ACA, Anterior cingulate area; OLF,\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; Olfactory areas; PL, Prelimbic area; ILA, Infralimbic area; GU, Gustatory areas; PERI, Perirhinal area; RSP, Retrosplenial area; CLA, Claustrum; BLA, Basolateral amygdala nucleus; VENT, Ventral group of the dorsal thalamus; LAT, Lateral group of the dorsal thalamus; ILM, Intralaminar nuclei of the dorsal thalamus; GEND, Geniculate group, dorsal thalamus; HB, Hindbrain; MB, Midbrain.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eExtended Data Fig. 6 | Brain-wide inputs to POL\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003eCAF\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Retrograde-traced presynaptic neurons for POL (scale bar: 200 μm).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Ipsilateral (red) vs. contralateral (black) input proportions to POL. n = 5 mice. TeA\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; , Temporal association areas; VISC, Visceral area; AuC, Auditory areas; GU, Gustatory areas; SS, Somatosensory areas; MO, Somatomotor areas; VIS, Visual areas; ACA, Anterior cingulate area; AI, Agranular insular area; RSP, Retrosplenial area; ECT, Ectorhinal area; PIR, Piriform area; STR, Striatum; PALd, Pallidum dorsal region; DORsm, Thalamus, sensory-motor cortex related; DORpm\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp; Thalamus, polymodal association cortex related; HY, Hypothalamus; P-sen, Pons, sensory related; P-mot, Pons, motor related; P-sat, Pons, behavioral state related; MY-mot, Medulla, motor related.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eExtended Data Fig. 7 | hM4Di-DREADD of POL→CAF inhibition disrupts courtship behaviors.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e POL→CAF inhibition strategy by injecting AAV2/9-DIO-hM4Di in POL and AAV2/2-Retro-Cre in CAF.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e Lordosis duration. Time in lordosis posture (10 min window; control vs. hM4Di).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec,\u003c/strong\u003e Lordosis frequency. Reduced lordosis events in hM4Di group (***p \u0026lt; 0.001; Mann-Whitney test; control: n=7; hM4Di: n = 9 mice).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed,\u003c/strong\u003e Rejection frequency. Increased rejection frequency in hM4Di group (***p \u0026lt; 0.001; Mann-Whitney test; control: n=7; hM4Di: n = 9 mice).\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eExtended Data Fig. 8 | A1 inhibition does not disrupt syllable processing in CAF.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea,\u003c/strong\u003e Schematics of experimental design for inhibition and imaging. Left: Chemogenetic inhibition of A1 during CAF recording. Right: Imaging setup.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb,\u003c/strong\u003e CAF responses under A1 inhibition. Left: Two-photon imaging of CAF neurons. Right: Calcium traces to syllables (8 trials).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec,\u003c/strong\u003e Population response profile. Heatmap of neurons sorted by response amplitude (one plane).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed,\u003c/strong\u003e No effect on CAF activity. Unchanged ΔF/F during A1 inhibition (ns, p \u0026gt; 0.5; one-sample Wilcoxon test).\u003c/p\u003e","description":"","filename":"CAFfiguresED.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8163343/v1/e2a066999ff396de394adb43.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nW.X. is a co-founder of SimpGen Therapeutics. This relationship did not influence this study. The other authors declare no competing interests.","formattedTitle":"A Thalamocortical Pathway for Decoding Communicative Syllables","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eThe cerebral cortex is anatomically and functionally organized into distinct sensory areas that contain specialized subregions for processing specific features of sensory input. This functional segregation enables efficient and high-dimensional information processing\u003csup\u003e1\u0026ndash;5\u003c/sup\u003e. Among these specialized regions, those involved in language and speech processing in humans represent a particularly sophisticated example of cortical specialization for social communication.\u003c/p\u003e \u003cp\u003eIn mice, social communication relies heavily on the auditory cortex (AuC) to process complex ultrasonic vocalizations (USV), which are composed of distinct syllables\u003csup\u003e6\u0026ndash;9\u003c/sup\u003e. The mouse AuC has traditionally been subdivided into tonotopically organized regions, including the primary auditory cortex (A1), ultrasonic field (UF), secondary auditory cortex (A2), and anterior auditory field (AAF). However, in contrast to the well-mapped speech-processing regions in humans (ref), the functional specialization of mouse auditory subregions for processing natural vocalizations remains poorly characterized. Although a putative vocalization-sensitive region between A2 and AAF was previously identified\u003csup\u003e10\u003c/sup\u003e, its thalamocortical connectivity and necessity for behavior have not been established. In contrast, the canonical A1 pathway is well-defined: it receives frequency-specific, lemniscal inputs from the ventral division of the medial geniculate body (MGBv) to encode basic acoustic features\u003csup\u003e11\u0026ndash;13\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn this study, we combined wide-field and two-photon calcium imaging in awake mice to identify and characterize a previously undefined subregion within the AuC that is specialized for processing ultrasonic communication sounds. Using anatomical tracing, we defined its unique thalamocortical connectivity, and through functional disruption experiments, we established its necessity for natural auditory-guided social behaviors.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCAF processes pup calls over tones\u003c/h2\u003e \u003cp\u003eTo identify cortical subregions specialized for processing complex vocalizations, we performed wide-field calcium imaging in awake, head-fixed mice expressing GCaMP6f in AuC. Mice were presented with pure tones and natural pup calls\u0026mdash;a well-characterized, ethologically relevant ultrasonic vocalization (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea)\u003csup\u003e14,15\u003c/sup\u003e. While A1 responded robustly to both 63-kHz tones (matching the dominant frequency of the pup calls) and the calls themselves, we identified a discrete region, termed Communicative Auditory Field (CAF), that was selectively activated by pup calls, with minimal responses to pure tones (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe next mapped the spatial organization of CAF relative to established auditory fields. Wide-field imaging across multiple mice confirmed that CAF is localized between AAF and UF, outside the classically defined tonotopic gradients of A1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec and Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo resolve CAF\u0026rsquo;s functional properties at cellular resolution, we performed two-photon calcium imaging of neuronal responses to pure tones (54, 63, and 80 kHz) and pup calls (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed). Population analysis revealed a striking preference for natural vocalizations: 57% of imaged CAF neurons responded to pup calls, compared to only 11\u0026ndash;13% that responded to any of the pure tones (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee).\u003c/p\u003e \u003cp\u003eCollectively, these data establish CAF as a discrete cortical module preferentially tuned to process complex communicative syllables over simple acoustic features.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCAF neurons are tuned to syllables.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWe next investigated whether CAF neurons display selectivity for 11 distinct syllable types (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea), a critical feature for decoding complex vocalization\u003csup\u003e16\u003c/sup\u003e. Using two-photon calcium imaging, we recorded from CAF neurons in awake mice during presentation of categorized ultrasonic syllables (Extended Data Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea,b). Individual CAF neurons exhibited distinct response profiles (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb) and showed preferential tuning on each syllable type (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). Each syllable type has a subpopulation of CAF neurons preferentially tuned (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec and Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe then asked whether syllable-tuned neurons are organized into spatially discrete clusters. Analysis of their spatial distribution revealed that neurons responsive to different syllables were intermingled throughout CAF, showing no apparent topographic organization (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed). The observed overlap between populations tuned to different syllables was not greater than that expected by chance, as confirmed by comparison to shuffled data (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee). Consistent with this, the spatial distribution of neurons responsive to multiple syllables was indistinguishable from that of the general responsive population (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef).\u003c/p\u003e \u003cp\u003eTogether, these findings demonstrate that CAF contains a distributed functional code for syllables, where selectivity for specific vocalization elements is maintained without spatial clustering.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCAF inhibition impairs communicative behaviors.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo investigate the functional role of CAF neurons in communicative behaviors, we employed chemogenetic inhibition using the hM4Di-DREADD system. Effective suppression of CAF neuronal activity was validated in acute brain slices, where DCZ application significantly reduced firing rates (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). We focused behavioral assays on two key auditory communication tasks: pup retrieval and courtship behavior.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn the pup retrieval task, postpartum female mice with CAF expressing hM4Di were tested in a Y-maze, where recorded pup calls and narrow-band noise (NBN) controls were randomly played from the two side arms (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg,h). Chemogenetic inhibition of CAF significantly reduced pup call-directed responses in the hM4Di group (52%) compared to the mCherry controls (83%; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Additionally, the hM4Di group exhibited prolonged retrieval latency and lower correct response rates (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ei-k).\u003c/p\u003e \u003cp\u003eDuring courtship behavior assays, where male USSs facilitate mating (Extended Data Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea,b), CAF inhibition significantly impaired female receptivity. This was evidenced by reduced time spent in lordosis (9% in hM4Di group vs. 26% in controls), fewer lordosis events, and increased rejection behaviors (Extended Data Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec-e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThese results underscore the necessity of CAF for mediating auditory-driven communicative behaviors.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCAF receives input from POL.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo elucidate afferent inputs to CAF, we performed retrograde tracing using scAAV2/2-Retro-tdTomato (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). Within the thalamus, the predominant fugal source of sensory input to cortex, over 80% of CAF-projecting neurons originated from the posterior limiting nucleus (POL), with additional contributions from the posterior intralaminar thalamic nucleus (PIL) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). In contrast, A1 primarily received input from the ventral division of the medial geniculate body (MGBv), consistent with its role in tonotopic processing\u003csup\u003e11\u003c/sup\u003e. Cortical input to CAF originated from auditory-related areas including temporal association cortex (TeA) and ectorhinal cortex (ECT) (Extended Data Fig.\u0026nbsp;5).\u003c/p\u003e \u003cp\u003eFiber photometry recordings in POL neurons expressing GCaMP6f revealed stronger responses to complex syllables (Chevron and Multi-steps) compared to pure tones (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec-e). Using monosynaptic rabies virus tracing from POL neurons projecting to CAF (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef,g and Extended Data Fig.\u0026nbsp;6), we found that this circuit receives input predominantly from the external (ICe, ~\u0026thinsp;60%) and dorsal (ICd, ~\u0026thinsp;30%) cortices of the inferior colliculus, with minor contributions from the central nucleus (ICc, ~\u0026thinsp;7%; Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eh,i).\u003c/p\u003e \u003cp\u003eThese findings establish POL as the major thalamic relay providing non-lemniscal input to CAF.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePOL\u0026rarr;CAF inhibition disrupts syllable-mediated communications.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo investigate the functional relevance of the POL\u0026rarr;CAF circuit, we selectively inhibited CAF-projecting POL neurons by expressing hM4Di specifically in POL neurons projecting to CAF using a retrograde-Cre approach, while monitoring CAF neurons simultaneously expressing GCaMP6f using two-photon calcium imaging in awake mice or examining communicative behaviors (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea).\u003c/p\u003e \u003cp\u003eInhibition of the POL\u0026rarr;CAF circuit significantly reduced Ca\u003csup\u003e2+\u003c/sup\u003e responses to chevron and multi-steps syllables (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Quantitative analysis of GCaMP6f signals revealed a significant decrease in response amplitude following POL inhibition (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec,d).\u003c/p\u003e \u003cp\u003eBehaviorally, inhibition of the POL\u0026rarr;CAF circuit impaired both pup retrieval and courtship behaviors. In the pup retrieval task (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee), inhibition reduced response probability (57% in hM4Di group vs. 85% in controls), increased retrieval latency, and decreased correct response rates (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ef,g). Similarly, courtship behavior was disrupted, with reduced lordosis time (8% in hM4Di group vs. 26% in controls), fewer lordosis events, and increased rejections (Extended Data Fig.\u0026nbsp;7).\u003c/p\u003e \u003cp\u003eThese findings demonstrate that POL\u0026rarr;CAF circuit is specifically required in processing complex syllables and guiding communicative behaviors.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePOL\u0026rarr;CAF operates independently of the canonical auditory pathway.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo determine whether CAF-mediated syllable processing depends on the primary auditory pathway, we inhibited A1 while monitoring CAF neuronal responses (Extended Data Fig.\u0026nbsp;8a). Chemogenetic silencing of A1 neurons had no effect on CAF responses to complex syllables (Extended Data Figs.\u0026nbsp;8b). Quantitative analysis confirmed no significant change in response amplitudes in CAF neurons following A1 inhibition (Figs.\u0026nbsp;8c,d).\u003c/p\u003e \u003cp\u003eThese results demonstrate that CAF processing of complex syllables is independent of the tonotopically organized A1, supporting the existence of a parallel IC\u0026rarr;POL\u0026rarr;CAF circuit dedicated to complex sound processing and communicative behavior (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eh).\u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eIn this study, we identified the CAF, a previously uncharacterized region of the mouse AuC that is selectively tuned to natural, complex syllabic vocalizations and is essential for guiding communicative behaviors. Our results establish that CAF\u0026rsquo;s selectivity for socially relevant vocalizations emerges from a noncanonical auditory pathway: IC shell\u0026rarr; POL \u0026rarr; CAF, which is distinct from the classical lemniscal route (ICc\u0026rarr;MGBv\u0026rarr;A1), indicating a dedicated cortical module for processing the temporal and semantic features of complex communication sounds. This circuit also demonstrates functional necessity: chemogenetic inhibition of either CAF or POL\u003csup\u003eCAF\u003c/sup\u003e selectively disrupts socially relevant behaviors, including pup retrieval and courtship.\u003c/p\u003e \u003cp\u003eCAF\u0026rsquo;s major thalamic input originates from POL, a subdivision of the non-classical auditory thalamus whose functional role has remained enigmatic\u003csup\u003e17\u003c/sup\u003e. Recent anatomical work by Liu et al. demonstrated that somatostatin-expressing (SOM⁺) neurons within the IC selectively innervate POL, which itself integrates polymodal inputs from midbrain structure involved in sensorimotor and defensive behaviors, such as the superior colliculus and periaqueductal gray\u003csup\u003e18\u003c/sup\u003e. Our functional data now assign a clear role to this IC\u0026rarr;POL pathway, positioning it as a critical conduit for socially salient auditory information en route to the cortex, rather than a mere modulator of arousal. The IC serves as the principal source of auditory input to higher centers, and its population activity encodes vocal signals that form the foundation for processing in the thalamus, cortex, and limbic regions\u003csup\u003e19,20\u003c/sup\u003e. Our transsynaptic tracing reveals that inputs to POL arise predominantly from dorsal and lateral shell regions of the IC (IC shell), rather than its tonotopically organized central nucleus (ICc). The IC shell lacks precise tonotopy but is known to receive convergent input from both ascending brainstem pathways and descending cortical projections, endowing it with a capacity for encoding complex, temporally dynamic sounds like USVs\u003csup\u003e21\u0026ndash;23\u003c/sup\u003e. This aligns with emerging evidence that the IC shell contains neurons tuned to ultrasonic frequencies and is engaged in translating sensory input into adaptive social behaviors\u003csup\u003e24,25\u003c/sup\u003e. Notably, a recent study identified a specialized pathway from the IC to the PIL, adjacent to POL, that is essential for processing pup retrieval calls and driving maternal behavior avia hypothalamic oxytocin release\u003csup\u003e26\u003c/sup\u003e. Our findings on the ICe\u0026rarr; POL \u0026rarr; CAF pathway thus complement and expand this evolving understanding of the IC\u0026rsquo;s role in social acoustics.\u003c/p\u003e \u003cp\u003ePrevious studies have proposed several higher-order regions within the rodent AuC, such as A2, dorsoposterior (DP), and dorsomedial (DM) fields, as candidates for processing natural sounds. These areas are characterized by broader frequency tuning and inputs from non-lemniscal thalamic or cortical sources. For instance, DM shows selective activation to courtship vocalizations\u003csup\u003e27\u003c/sup\u003e and A2 exhibits heightened responses to pup calls linked to maternal behavior\u003csup\u003e28\u003c/sup\u003e. Furthermore, a frequency-insensitive region between A1 and AAF was previously noted for its heterogeneous responses to vocalizations\u003csup\u003e10\u003c/sup\u003e. However, none of these studies delineated the specific region we identify as CAF. We propose this omission likely stems from differences in sound stimulus design. We used ethologically relevant pup calls that preserve the natural temporal and syllabic structure crucial for eliciting robust behavioral responses. Maternal retrieval, for example, is known to be exquisitely sensitive to the inter-syllable interval (ISI) rhythm of pup calls\u003csup\u003e29\u003c/sup\u003e. The use of simplified acoustic stimuli lacking this critical temporal structure in prior work may have failed to engage the CAF module, highlighting the importance of ethological stimulus design in functional mapping\u003csup\u003e30\u0026ndash;33\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe organization of communicative brain regions has been extensively studied in the human auditory system\u003csup\u003e34\u003c/sup\u003e. Beyond the primary auditory cortex (Heschl's gyrus), which processes basic acoustic features, the superior temporal gyrus (STG) contains subregions specialized for complex speech-related features. Notably, patients with lesions to Heschl\u0026rsquo;s gyrus can retain intact speech perception, indicating that the STG operates as a distinct, high-order module. The existence of a similarly parallel and distributed organization in the rodent brain, with a dedicated CAF region for complex social vocalizations operating alongside the A1, suggests an evolutionarily conserved strategy for efficient decoding of communication sounds. This convergence implies that fundamental principles of hierarchical and parallel processing may underline complex sound perception across mammals.\u003c/p\u003e \u003cp\u003eIn summary, our findings support a modular and parallel organizational principle within the mouse auditory cortex. We define CAF as a specialized region dedicated to decoding complex, socially salient vocal cues via a non-canonical thalamocortical pathway from the IC shell through POL. This functional segregation, reminiscent of specialized pathways in the human auditory system, enables the efficient processing of ethologically critical communication sounds.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec6\"\u003e\n \u003ch2\u003eAnimals\u003c/h2\u003e\n \u003cp\u003eAll animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) guidelines at Chongqing University, Tsinghua University, and CIBR. CBA/J mice (8–12 weeks old) of both sexes were used. Both males and females were used for non-behavioral experiments; only females were used for pup retrieval task. Mice were housed under a 12 h light/dark cycle (lights on at 7:00 a.m.) with ad libitum access to food and water unless otherwise specified.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eVirus and fluorescent indicator injections\u003c/h3\u003e\n\u003cp\u003eAll surgical procedures were conducted under isoflurane anesthesia (3% for induction; 1.5% for maintenance) in oxygen-enriched air. Mice were positioned in a stereotaxic apparatus, and ophthalmic ointment was applied to prevent corneal drying. Body temperature was maintained at 36.5–37.5°C with a heating pad. After shaving and disinfecting the scalp, the skull was exposed via a midline incision. For viral delivery, small craniotomies were made above the targeted regions. In widefield and two-photon calcium imaging experiments, 300 nL of AAV2/9-hSyn-GCaMP6f (5.0 × 10¹² viral particles/mL) was pressure-injected into the left AuC using a glass pipette. Retrograde tracing of CAF-projecting neurons was performed using retro-scAAV-tdTomato. For behaviorally targeted inhibition of the POL, CBA/J mice received bilateral injections of AAV2/9-DIO-hM4Di-mCherry or control AAV2/9-DIO-mCherry into the CAF, along with bilateral injections of AAV2/2-Retro-Cre into the CAF. For A1 inhibition experiments, AAV2/9-hSyn-hM4Di-mCherry or its control was bilaterally injected into A1. Monosynaptic rabies tracing was used to identify upstream inputs to POL neurons projecting to CAF. To this end, helper viruses (AAV2/5-Ef1a-DIO-EGFP-T2A-TVA and AAV2/5-Ef1a-DIO-RVG) were injected into POL, while AAV2/2-Retro-Cre was delivered to the CAF. Three weeks later, EnvA-pseudotyped rabies virus were injected into the CAF. CAF was targeted based on widefield imaging. POL was injected at three stereotaxic coordinates (relative to bregma, from dura) to cover its elongated structure: − 3.05 AP, ± 1.87 ML, − 2.80 DV; − 3.05 AP, ± 1.81 ML, − 3.23 DV; and − 3.05 AP, − 1.65 ML, − 3.60 DV.\u003c/p\u003e\n\u003cdiv id=\"Sec8\"\u003e\n \u003ch2\u003eWide-field fluorescence imaging\u003c/h2\u003e\n \u003cp\u003eTo locate the position of CAF, we performed widefield calcium imaging (2-CAFE, NewLight Optoinstrument; www.newlightxhr.com) equipped with a 4×, 0.2 NA objective (Olympus). Cortical fluorescence was excited by a 470 nm LED (Thorlabs M470L4), and green emission was captured at 10 Hz using a high-speed sCMOS camera (Zyla 4.2, Andor Technology) through a standard filter cube. A 3 mm diameter cranial window was implanted over the auditory cortex (AuC) following a craniotomy. For signal quantification, fluorescence intensity was normalized to the baseline (f₀), defined as the average signal during the 1 s preceding stimulus presentation. Δf/f was calculated as (f – f₀)/f₀ and averaged across trials. Resulting Δf/f maps were visualized using a pseudocolor scale to reveal sound-evoked responses across cortical regions. For widefield image, tonotopic maps were generated from widefield calcium imaging data using custom MATLAB scripts. Images were downsampled to 150×150 pixels for analysis. For each stimulus frequency, frequency-specific response regions were identified based on calcium signals exceeding two standard deviations above baseline. The best frequency (BF) for each pixel was computed as the weighted average of responses across all frequencies, following a previously established method \u003csup\u003e10\u003c/sup\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eTwo-photon Ca imaging\u003c/h3\u003e\n\u003cp\u003eFour weeks after viral injection, a small cranial window was implanted over the left AuC. A custom-made plastic imaging chamber was affixed to the skull using cyanoacrylate glue (UHU), and a circular portion of the skull (~ 3 mm diameter) was removed and replaced with glass coverslip. In vivo imaging was carried out using a LOTOS2 two-photon microscope (New Light Opto lnstrument Co., Ltd; see: www.newlightxhr.com), which equipped with a 12.0 kHz resonant scanner (model details, Spectra-Physics DeepSee Mai Tai). Excitation was provided by a mode-locked Ti:sapphire laser tuned to 920 nm. A 40× objective with a 0.8 numerical aperture (Nikon) was used to acquire images. For cellular imaging of L2/3 neurons, the field of view was typically 400 × 400 µm. Laser power at the brain surface ranged from 30 to 120 mW, depending on imaging depth. For two-photo imaging data analysis, calcium signals were analyzed from manually defined regions of interest (ROIs) based on fluorescence intensity. We calculated ΔF/F relative to the average baseline fluorescence within each ROI. A neuron was considered responsive if its average response across all trials also exceeded 3× SD within 2 s post-stimulation. Data were analyzed using custom scripts in LabVIEW 2016, MATLAB 2020a, and GraphPad Prism 9.0.\u003c/p\u003e\n\u003ch3\u003eOptical fiber recording\u003c/h3\u003e\n\u003cp\u003eFor Ca\u003csup\u003e2+\u003c/sup\u003e indicator expression and fiber optic implantation in CAF-projecting POL neurons, we injected AAV2/2-Retro-Cre into the CAF and AAV2/8-Dio-GCamp6f into the POL, delivering 200 nL. A 200 mm fiber optic with an attached cannula (R-FOC-BL-200C-39N, RWD) was then implanted 2.8 mm deep above the injection site.\u003c/p\u003e\n\u003cdiv id=\"Sec11\"\u003e\n \u003ch2\u003eAuditory stimulation\u003c/h2\u003e\n \u003cp\u003eFor all calcium imaging experiments, including two-photon imaging, widefield fluorescence imaging, and fiber photometry, auditory stimuli were presented using consistent temporal and acoustic parameters across conditions. Pure-tone stimuli were generated using a custom program developed in LabVIEW 2012 (National Instruments) and converted to analog voltages via a PCI-6731 card (National Instruments). Sounds were delivered through an ED1 electrostatic speaker driver and a free-field ES1 speaker (Tucker-Davis Technologies) \u003csup\u003e35,36\u003c/sup\u003e. Syllable-based stimuli were derived from ultrasonic vocalizations (USVs) recorded from freely behaving mice using the Avisoft RECORDER system. Based on previous study \u003csup\u003e16\u003c/sup\u003e, we identified 11 distinct syllable types. During each trial, a single syllable was repeatedly played for 2 seconds, preserving its natural inter-syllable interval (ISI) ranging from 100 to 275 ms \u003csup\u003e29,33\u003c/sup\u003e. All stimuli—whether pure tone or syllable-based—were presented at 65 dB SPL and repeated 10 times per condition. The 5-second cycle (2-s stimulation + 3-s silence) ensured sufficient time for calcium signals to return to baseline between trials.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\"\u003e\n \u003ch2\u003ePup retrieval behavior\u003c/h2\u003e\n \u003cp\u003eThis experiment utilized a customized Y-maze apparatus, comprising a 40×30×30 cm central hall connected to two 15×15×30 cm symmetrical side chambers. A 10×10 cm nest area was positioned at the center of the rear wall in the central region, with each side chamber equipped with an UltraSoundGate 116H ultrasonic speaker. Bedding was laid at the bottom, and water and food were provided for the female mice to access freely during the experiment. The experimental subjects were female CBA/J mice 5–8 days postpartum. One day prior to the test, original nest materials were transferred to the nest area on the rear wall of the central region in the test cage, and nest reconstruction by the female mice was observed and confirmed. Pups were individually placed in a 20×20×20 cm soundproof box maintained at 30°C with a heating pad. Ultrasonic vocalizations (USVs) were recorded for 5 minutes per session using an Avisoft-UltraSoundGate CM16 microphone (15 cm above the box bottom) at a sampling rate of 250 kHz with 16-bit precision. Thirty minutes before the test, female mice were intraperitoneally injected with DCZ (0.1 mg/kg body weight) or physiological saline. During the behavioral test, the female mice were first allowed to acclimate to the central region for 5 minutes. A pup was then randomly introduced into the left or right chamber, and the time taken by the female mice to retrieve the pup was recorded. After the female mice remained quiet in the nest for 10 seconds, the pup’s vocalizations filtered to 20–100 kHz (USV group) or 50–75 kHz narrow-band noise (NBN group) were randomly played from the left or right speaker. To generate narrow-band noise (NBN) as control, each USV syllable was temporally matched with a band-limited noise segment (50–75 kHz) of the same duration. The amplitude of each NBN segment was adjusted to match the envelope of its corresponding syllable. This preserved the temporal envelope of the vocalizations while disrupting spectral structure. The trial ended when the female mouse fully entered the side chamber (defined as making a choice) and was terminated if no choice was made within 4 minutes. Videos were recorded using a SONY FDR-AX60.The latencies to correct choices, probabilities of correct choices, and probabilities of making choices by the mice were statistically analyzed using GraphPad Prism.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\"\u003e\n \u003ch2\u003eCourtship behavior\u003c/h2\u003e\n \u003cp\u003eThis experiment utilized a 30×40×40 cm transparent observation chamber partitioned by a removable divider, conducted in a quiet environment with constant temperature (22 ± 2°C) and humidity (55 ± 5%). Mating behavior tests were performed in CBA/J mice aged 2–3 months. Male mice were pre-screened to confirm at least one prior successful mating, while female mice were virgins. Estrus was determined by vaginal lavage: secretions were stained with Giemsa, observed under a bright-field microscope, and females with ≥ 70% cornified epithelial cells were classified as estrous. Tests began 1 hour after the onset of the dark phase in the mice's housing room. Thirty minutes before testing, mice received intraperitoneal injections of DCZ (0.1 mg/kg body weight) or saline. Estrous females and sexually experienced males were acclimated separately on either side of the divider for 10 minutes. After removing the divider, interactions were observed for 30 minutes, with simultaneous video and audio recording using an RMONCAM-S900H USB camera (overhead view), SONY FDR-AX60 camera (frontal view), and an Avisoft-UltraSoundGate CM16 microphone (30 cm above the floor, 250 kHz sampling rate). A custom MATLAB program was used for frame-by-frame manual annotation of frontal-view videos, analyzing the 10-minute period following the female's first mounting event. Behavioral criteria were defined as: lordosis—female hindlimb support with spinal arching during mating or intromission; mounting—male climbing onto the female's back without pelvic thrusting; rejection behaviors including three subtypes: active escape (running \u0026gt; three body lengths), postural avoidance (lateral evasion), and defensive posture (forelimb pushing).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\"\u003e\n \u003ch2\u003eData analysis\u003c/h2\u003e\n \u003cp\u003eAll results are presented as mean ± SD. Specific experimental details and animal numbers are indicated in the figure panels and legends. For comparisons between groups, the two-sided Mann–Whitney U test was used; for paired observations, the two-sided Wilcoxon signed-rank test was applied. When evaluating more than two groups, the Kruskal–Wallis test was conducted, followed by Dunn’s post hoc test for pairwise comparisons when overall significance was detected. Statistical significance in the figures is denoted as follows: ns (p \u0026gt; 0.05), * (p \u0026lt; 0.05), ** (p \u0026lt; 0.01),, *** (p \u0026lt; 0.001) and **** (p \u0026lt; 0.0001).\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors thank Fenghua Liu and Yu Wang for laboratory management. We also thank the Laboratory Animal Resource Center of Tsinghua University, CIBR, and Chongqing University for maintenance of mice; the Imaging Core of CIBR for assistance in imaging. W.X. discloses support for the research described in this study from China Ministry of Science and Technology (2021ZD0203304), National Natural Science Foundation of China (32300833 to Z.Z.; 32430044 to X.C.; U23A20442 to W.X.; 32300830 to S.L.), and Shenzhen Medical Research Fund (B2402008).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn vivo imaging: Z.Z.; Data analysis: X.T., Y-Z.W., and S.L.; Mouse surgery, viral injection, and histology: Z.Z. and Y-L.W.; Mouse behavior: Z.Z., Y-L.W., M.P., and J.L.; Brain slice recording: J.T.; Photometry: Z.Y.; Mouse breeding: Z.Z., Y.-L.W., and Q.L.; Experiment design and supervision: Z.Z., X.C., and W.X.; Manuscript writing: Z.Z. and W.X..\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eW.X. is a co-founder of SimpGen Therapeutics. This relationship did not influence this study. The other authors declare no competing interests.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eKanold, P. O., Nelken, I. \u0026amp; Polley, D. B. Local versus global scales of organization in auditory cortex. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 502-510 (2014). https://doi.org/10.1016/j.tins.2014.06.003\u003c/li\u003e\n\u003cli\u003eFunkhouser, E. B. The visual cortex, its location, histological structure, and physiological function. \u003cem\u003eJ Exp Med\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 617-628 (1915). https://doi.org/10.1084/jem.21.6.617\u003c/li\u003e\n\u003cli\u003eBeltramo, R. \u0026amp; Scanziani, M. A collicular visual cortex: Neocortical space for an ancient midbrain visual structure. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e363\u003c/strong\u003e, 64-69 (2019). https://doi.org/10.1126/science.aau7052\u003c/li\u003e\n\u003cli\u003eChen, X., Gabitto, M., Peng, Y., Ryba, N. J. \u0026amp; Zuker, C. S. A gustotopic map of taste qualities in the mammalian brain. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e333\u003c/strong\u003e, 1262-1266 (2011). https://doi.org/10.1126/science.1204076\u003c/li\u003e\n\u003cli\u003eRothschild, G., Nelken, I. \u0026amp; Mizrahi, A. Functional organization and population dynamics in the mouse primary auditory cortex. \u003cem\u003eNat Neurosci\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 353-360 (2010). https://doi.org/10.1038/nn.2484\u003c/li\u003e\n\u003cli\u003eTasaka, G. I.\u003cem\u003e et al.\u003c/em\u003e Genetic tagging of active neurons in auditory cortex reveals maternal plasticity of coding ultrasonic vocalizations. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 871 (2018). https://doi.org/10.1038/s41467-018-03183-2\u003c/li\u003e\n\u003cli\u003eZucca, S., La Rosa, C., Fellin, T., Peretto, P. \u0026amp; Bovetti, S. Developmental encoding of natural sounds in the mouse auditory cortex. \u003cem\u003eCereb Cortex\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e (2024). https://doi.org/10.1093/cercor/bhae438\u003c/li\u003e\n\u003cli\u003eRauschecker, J. P. Cortical processing of complex sounds. \u003cem\u003eCurr Opin Neurobiol\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 516-521 (1998). https://doi.org/10.1016/s0959-4388(98)80040-8\u003c/li\u003e\n\u003cli\u003eScott, S. K. \u0026amp; Johnsrude, I. S. The neuroanatomical and functional organization of speech perception. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 100-107 (2003). https://doi.org/10.1016/s0166-2236(02)00037-1\u003c/li\u003e\n\u003cli\u003eIssa, J. B.\u003cem\u003e et al.\u003c/em\u003e Multiscale optical Ca2+ imaging of tonal organization in mouse auditory cortex. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e83\u003c/strong\u003e, 944-959 (2014). https://doi.org/10.1016/j.neuron.2014.07.009\u003c/li\u003e\n\u003cli\u003eHackett, T. A., Barkat, T. R., O\u0026apos;Brien, B. M., Hensch, T. K. \u0026amp; Polley, D. B. Linking topography to tonotopy in the mouse auditory thalamocortical circuit. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 2983-2995 (2011). https://doi.org/10.1523/JNEUROSCI.5333-10.2011\u003c/li\u003e\n\u003cli\u003eTsukano, H.\u003cem\u003e et al.\u003c/em\u003e Reconsidering Tonotopic Maps in the Auditory Cortex and Lemniscal Auditory Thalamus in Mice. \u003cem\u003eFront Neural Circuits\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 14 (2017). https://doi.org/10.3389/fncir.2017.00014\u003c/li\u003e\n\u003cli\u003eNakata, S., Takemoto, M. \u0026amp; Song, W. J. Differential cortical and subcortical projection targets of subfields in the core region of mouse auditory cortex. \u003cem\u003eHear Res\u003c/em\u003e \u003cstrong\u003e386\u003c/strong\u003e, 107876 (2020). https://doi.org/10.1016/j.heares.2019.107876\u003c/li\u003e\n\u003cli\u003eLecca, S.\u003cem\u003e et al.\u003c/em\u003e A neural substrate for negative affect dictates female parental behavior. \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, 1094-1103.e1098 (2023). https://doi.org/10.1016/j.neuron.2023.01.003\u003c/li\u003e\n\u003cli\u003eBurenkova, O. V., Averkina, A. A., Aleksandrova, E. A. \u0026amp; Zarayskaya, I. Y. Brief but enough: 45-min maternal separation elicits behavioral and physiological responses in neonatal mice and changes in dam maternal behavior. \u003cem\u003ePhysiol Behav\u003c/em\u003e \u003cstrong\u003e222\u003c/strong\u003e, 112877 (2020). https://doi.org/10.1016/j.physbeh.2020.112877\u003c/li\u003e\n\u003cli\u003eFonseca, A. H., Santana, G. M., Bosque Ortiz, G. M., Bampi, S. \u0026amp; Dietrich, M. O. Analysis of ultrasonic vocalizations from mice using computer vision and machine learning. \u003cem\u003eElife\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e (2021). https://doi.org/10.7554/eLife.59161\u003c/li\u003e\n\u003cli\u003eM\u0026aacute;rquez-Legorreta, E., Horta-J\u0026uacute;nior Jde, A., Berrebi, A. S. \u0026amp; Salda\u0026ntilde;a, E. Organization of the zone of transition between the pretectum and the thalamus, with emphasis on the pretectothalamic lamina. \u003cem\u003eFront Neuroanat\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 82 (2016). https://doi.org/10.3389/fnana.2016.00082\u003c/li\u003e\n\u003cli\u003eLiu, M.\u003cem\u003e et al.\u003c/em\u003e Parvalbumin and Somatostatin: Biomarkers for Two Parallel Tectothalamic Pathways in the Auditory Midbrain. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e (2024). https://doi.org/10.1523/JNEUROSCI.1655-23.2024\u003c/li\u003e\n\u003cli\u003eWoolley, S. M. \u0026amp; Portfors, C. V. Conserved mechanisms of vocalization coding in mammalian and songbird auditory midbrain. \u003cem\u003eHear Res\u003c/em\u003e \u003cstrong\u003e305\u003c/strong\u003e, 45-56 (2013). https://doi.org/10.1016/j.heares.2013.05.005\u003c/li\u003e\n\u003cli\u003eLyzwa, D. \u0026amp; W\u0026ouml;rg\u0026ouml;tter, F. Neural and response correlations to complex natural sounds in the auditory midbrain. \u003cem\u003eFront Neural Circuits\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 89 (2016). https://doi.org/10.3389/fncir.2016.00089\u003c/li\u003e\n\u003cli\u003eChen, C., Cheng, M., Ito, T. \u0026amp; Song, S. Neuronal organization in the inferior colliculusrevisited with Cell-type-dependent monosynaptic tracing. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 3318-3332 (2018). https://doi.org/10.1523/jneurosci.2173-17.2018\u003c/li\u003e\n\u003cli\u003eLesicko, A. M., Hristova, T. S., Maigler, K. C. \u0026amp; Llano, D. A. Connectional modularity of Top-down and Bottom-up multimodal inputs to the lateral cortex of the mouse inferior colliculus. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 11037-11050 (2016). https://doi.org/10.1523/jneurosci.4134-15.2016\u003c/li\u003e\n\u003cli\u003eCasseday, J. H. \u0026amp; Covey, E. A neuroethological theory of the operation of the inferior colliculus. \u003cem\u003eBrain Behav Evol\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, 311-336 (1996). https://doi.org/10.1159/000113249\u003c/li\u003e\n\u003cli\u003eShi, K.\u003cem\u003e et al.\u003c/em\u003e Population coding of time-varying sounds in the nonlemniscal inferior colliculus. \u003cem\u003eJ Neurophysiol\u003c/em\u003e \u003cstrong\u003e131\u003c/strong\u003e, 842-864 (2024). https://doi.org/10.1152/jn.00013.2024\u003c/li\u003e\n\u003cli\u003eHolmstrom, L. A., Eeuwes, L. B., Roberts, P. D. \u0026amp; Portfors, C. V. Efficient encoding of vocalizations in the auditory midbrain. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 802-819 (2010). https://doi.org/10.1523/JNEUROSCI.1964-09.2010\u003c/li\u003e\n\u003cli\u003eValtcheva, S.\u003cem\u003e et al.\u003c/em\u003e Neural circuitry for maternal oxytocin release induced by infant cries. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e621\u003c/strong\u003e, 788-795 (2023). https://doi.org/10.1038/s41586-023-06540-4\u003c/li\u003e\n\u003cli\u003eTsukano, H.\u003cem\u003e et al.\u003c/em\u003e Delineation of a frequency-organized region isolated from the mouse primary auditory cortex. \u003cem\u003eJ Neurophysiol\u003c/em\u003e \u003cstrong\u003e113\u003c/strong\u003e, 2900-2920 (2015). https://doi.org/10.1152/jn.00932.2014\u003c/li\u003e\n\u003cli\u003eChong, K. K., Anandakumar, D. B., Dunlap, A. G., Kacsoh, D. B. \u0026amp; Liu, R. C. Experience-dependent coding of time-dependent frequency trajectories by off responses in secondary auditory cortex. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 4469-4482 (2020). https://doi.org/10.1523/jneurosci.2665-19.2020\u003c/li\u003e\n\u003cli\u003eSchiavo, J. K.\u003cem\u003e et al.\u003c/em\u003e Innate and plastic mechanisms for maternal behaviour in auditory cortex. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e587\u003c/strong\u003e, 426-431 (2020). https://doi.org/10.1038/s41586-020-2807-6\u003c/li\u003e\n\u003cli\u003eGaub, S. \u0026amp; Ehret, G. Grouping in auditory temporal perception and vocal production is mutually adapted: the case of wriggling calls of mice. \u003cem\u003eJ Comp Physiol A Neuroethol Sens Neural Behav Physiol\u003c/em\u003e \u003cstrong\u003e191\u003c/strong\u003e, 1131-1135 (2005). https://doi.org/10.1007/s00359-005-0036-y\u003c/li\u003e\n\u003cli\u003eEhret, G. Infant rodent ultrasounds -- a gate to the understanding of sound communication. \u003cem\u003eBehav Genet\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 19-29 (2005). https://doi.org/10.1007/s10519-004-0853-8\u003c/li\u003e\n\u003cli\u003eUematsu, A.\u003cem\u003e et al.\u003c/em\u003e Maternal approaches to pup ultrasonic vocalizations produced by a nanocrystalline silicon thermo-acoustic emitter. \u003cem\u003eBrain Res\u003c/em\u003e \u003cstrong\u003e1163\u003c/strong\u003e, 91-99 (2007). https://doi.org/10.1016/j.brainres.2007.05.056\u003c/li\u003e\n\u003cli\u003eCastellucci, G. A., Calbick, D. \u0026amp; McCormick, D. The temporal organization of mouse ultrasonic vocalizations. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e0199929 (2018). https://doi.org/10.1371/journal.pone.0199929\u003c/li\u003e\n\u003cli\u003eHamilton, L. S., Oganian, Y., Hall, J. \u0026amp; Chang, E. F. Parallel and distributed encoding of speech across human auditory cortex. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e184\u003c/strong\u003e, 4626-4639.e4613 (2021). https://doi.org/10.1016/j.cell.2021.07.019\u003c/li\u003e\n\u003cli\u003eLi, J.\u003cem\u003e et al.\u003c/em\u003e PIEZO2 mediates ultrasonic hearing via cochlear outer hair cells in mice. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e2101207118 (2021). https://doi.org/10.1073/pnas.2101207118\u003c/li\u003e\n\u003cli\u003eLi, J.\u003cem\u003e et al.\u003c/em\u003e Prestin-Mediated Frequency Selectivity Does not Cover Ultrahigh Frequencies in Mice. \u003cem\u003eNeurosci Bull\u003c/em\u003e\u003cstrong\u003e38\u003c/strong\u003e, 769-784 (2022). https://doi.org/10.1007/s12264-022-00839-4\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8163343/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8163343/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSocial communication in rodents depends on ultrasonic vocalizations, yet the neural circuits that selectively process these complex acoustic signals remain poorly defined. Here, using wide-field calcium imaging in awake mice, we identify a previously unknown region of the auditory cortex that we termed communicative auditory field (CAF). CAF is located between the anterior auditory field and ultrasonic field and responds selectively to ultrasonic syllables. Two-photon imaging revealed that individual CAF neurons are tuned to distinct syllable categories. Transsynaptic tracing showed that CAF receives its predominant thalamic input from the posterior nucleus (POL), defining a pathway in parallel to the canonical lemniscal pathway from the ventral medial geniculate body to the primary auditory cortex. Chemogenetic inhibition of either CAF or POL neurons projecting to CAF impaired innate social behaviors, including pup retrieval and courtship, whereas inhibition of A1 was ineffective. These findings define a specialized thalamocortical circuit for decoding ethologically relevant communication sounds.\u003c/p\u003e","manuscriptTitle":"A Thalamocortical Pathway for Decoding Communicative Syllables","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-30 10:00:51","doi":"10.21203/rs.3.rs-8163343/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"dfa9834e-b4b6-4a9f-b5db-f9e1c95faa0b","owner":[],"postedDate":"December 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":59171786,"name":"Biological sciences/Neuroscience/Auditory system/Cortex"},{"id":59171787,"name":"Biological sciences/Neuroscience/Sensory processing"}],"tags":[],"updatedAt":"2025-12-30T10:00:51+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-30 10:00:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8163343","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8163343","identity":"rs-8163343","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}