Common and distinct patterns underlying different linguistic tasks: multivariate disconnectome symptom mapping in poststroke patients.

Numerous studies have been devoted to neural mechanisms of a variety of linguistic tasks (e.g. speech comprehension and production). To date, however, whether and how the neural patterns underlying different linguistic tasks are similar or differ remains elusive. In this study, we compared the neural patterns underlying 3 linguistic tasks mainly concerning speech comprehension and production. To address this, multivariate regression approaches with lesion/disconnection symptom mapping were applied to data from 216 stroke patients with damage to the left hemisphere. The results showed that lesion/disconnection patterns could predict both poststroke scores of speech comprehension and production tasks; these patterns exhibited shared regions on the temporal pole of the left hemisphere as well as unique regions contributing to the prediction for each domain. Lower scores in speech comprehension tasks were associated with lesions/abnormalities in the superior temporal gyrus and middle temporal gyrus, while lower scores in speech production tasks were associated with lesions/abnormalities in the left inferior parietal lobe and frontal lobe. These results suggested an important role of the ventral and dorsal stream pathways in speech comprehension and production (i.e. supporting the dual stream model) and highlighted the applicability of the novel multivariate disconnectome-based symptom mapping in cognitive neuroscience research.

Phase-encoded fMRI tracks down brainstorms of natural language processing with subsecond precision.

Natural language processing unfolds information overtime as spatially separated, multimodal, and interconnected neural processes. Existing noninvasive subtraction-based neuroimaging techniques cannot simultaneously achieve the spatial and temporal resolutions required to visualize ongoing information flows across the whole brain. Here we have developed rapid phase-encoded designs to fully exploit the temporal information latent in functional magnetic resonance imaging data, as well as overcoming scanner noise and head-motion challenges during overt language tasks. We captured real-time information flows as coherent hemodynamic waves traveling over the cortical surface during listening, reading aloud, reciting, and oral cross-language interpreting tasks. We were able to observe the timing, location, direction, and surge of traveling waves in all language tasks, which were visualized as "brainstorms" on brain "weather" maps. The paths of hemodynamic traveling waves provide direct evidence for dual-stream models of the visual and auditory systems as well as logistics models for crossmodal and cross-language processing. Specifically, we have tracked down the step-by-step processing of written or spoken sentences first being received and processed by the visual or auditory streams, carried across language and domain-general cognitive regions, and finally delivered as overt speeches monitored through the auditory cortex, which gives a complete picture of information flows across the brain during natural language functioning. PRACTITIONER POINTS: Phase-encoded fMRI enables simultaneous imaging of high spatial and temporal resolution, capturing continuous spatiotemporal dynamics of the entire brain during real-time overt natural language tasks. Spatiotemporal traveling wave patterns provide direct evidence for constructing comprehensive and explicit models of human information processing. This study unlocks the potential of applying rapid phase-encoded fMRI to indirectly track the underlying neural information flows of sequential sensory, motor, and high-order cognitive processes.

Dual Stream Long Short-Term Memory Feature Fusion Classifier for Surface Electromyography Gesture Recognition.

Gesture recognition using electromyography (EMG) signals has prevailed recently in the field of human-computer interactions for controlling intelligent prosthetics. Currently, machine learning and deep learning are the two most commonly employed methods for classifying hand gestures. Despite traditional machine learning methods already achieving impressive performance, it is still a huge amount of work to carry out feature extraction manually. The existing deep learning methods utilize complex neural network architectures to achieve higher accuracy, which will suffer from overfitting, insufficient adaptability, and low recognition accuracy. To improve the existing phenomenon, a novel lightweight model named dual stream LSTM feature fusion classifier is proposed based on the concatenation of five time-domain features of EMG signals and raw data, which are both processed with one-dimensional convolutional neural networks and LSTM layers to carry out the classification. The proposed method can effectively capture global features of EMG signals using a simple architecture, which means less computational cost. An experiment is conducted on a public DB1 dataset with 52 gestures, and each of the 27 subjects repeats every gesture 10 times. The accuracy rate achieved by the model is 89.66%, which is comparable to that achieved by more complex deep learning neural networks, and the inference time for each gesture is 87.6 ms, which can also be implied in a real-time control system. The proposed model is validated using a subject-wise experiment on 10 out of the 40 subjects in the DB2 dataset, achieving a mean accuracy of 91.74%. This is illustrated by its ability to fuse time-domain features and raw data to extract more effective information from the sEMG signal and select an appropriate, efficient, lightweight network to enhance the recognition results.

A Dual-Stream Cross AGFormer-GPT Network for Traffic Flow Prediction Based on Large-Scale Road Sensor Data.

Traffic flow prediction can provide important reference data for managers to maintain traffic order, and can also be based on personal travel plans for optimal route selection. On account of the development of sensors and data collection technology, large-scale road network historical data can be effectively used, but their high non-linearity makes it meaningful to establish effective prediction models. In this regard, this paper proposes a dual-stream cross AGFormer-GPT network with prompt engineering for traffic flow prediction, which integrates traffic occupancy and speed as two prompts into traffic flow in the form of cross-attention, and uniquely mines spatial correlation and temporal correlation information through the dual-stream cross structure, effectively combining the advantages of the adaptive graph neural network and large language model to improve prediction accuracy. The experimental results on two PeMS road network data sets have verified that the model has improved by about 1.2% in traffic prediction accuracy under different road networks.

A Novel End-to-End Deep Learning Framework for Chip Packaging Defect Detection.

As semiconductor chip manufacturing technology advances, chip structures are becoming more complex, leading to an increased likelihood of void defects in the solder layer during packaging. However, identifying void defects in packaged chips remains a significant challenge due to the complex chip background, varying defect sizes and shapes, and blurred boundaries between voids and their surroundings. To address these challenges, we present a deep-learning-based framework for void defect segmentation in chip packaging. The framework consists of two main components: a solder region extraction method and a void defect segmentation network. The solder region extraction method includes a lightweight segmentation network and a rotation correction algorithm that eliminates background noise and accurately captures the solder region of the chip. The void defect segmentation network is designed for efficient and accurate defect segmentation. To cope with the variability of void defect shapes and sizes, we propose a Mamba model-based encoder that uses a visual state space module for multi-scale information extraction. In addition, we propose an interactive dual-stream decoder that uses a feature correlation cross gate module to fuse the streams' features to improve their correlation and produce more accurate void defect segmentation maps. The effectiveness of the framework is evaluated through quantitative and qualitative experiments on our custom X-ray chip dataset. Furthermore, the proposed void defect segmentation framework for chip packaging has been applied to a real factory inspection line, achieving an accuracy of 93.3% in chip qualification.

Limb Apraxias: The Influence of Higher Order Perceptual and Semantic Deficits in Motor Recovery After Stroke.

Stroke is a leading cause of disability worldwide. Limb apraxia is a group of higher order motor disorders associated with greater disability and dependence after stroke. Original neuropsychology studies distinguished separate brain pathways involved in perception and action, known as the dual stream hypothesis. This framework has allowed a better understanding of the deficits identified in Limb Apraxia. In this review, we propose a hierarchical organization of this disorder, in which a distinction can be made between several visuomotor pathways that lead to purposeful actions. Based on this, executive apraxias (such as limb kinetic apraxia) cause deficits in executing fine motor hand skills, and intermediate apraxias (such as optic ataxia and tactile apraxia) cause deficits in reaching to grasp and manipulating objects in space. These disorders usually affect the contralesional limb. A further set of disorders collectively known as limb apraxias include deficits in gesture imitation, pantomime, gesture recognition, and object use. These deficits are due to deficits in integrating perceptual and semantic information to generate complex movements. Limb apraxias are usually caused by left-hemisphere lesions in right-handed stroke patients, affecting both limbs. The anterior- to posterior-axis of brain areas are disrupted depending on the increasing involvement of perceptual and semantic processes with each condition. Lower-level executive apraxias are linked to lesions in the frontal lobe and the basal ganglia, while intermediate apraxias are linked to lesions in dorso-dorsal subdivisions of the dorsal fronto-parietal networks. Limb apraxias can be caused by lesions in both dorsal and ventral subdivisions including the ventro-dorsal stream and a third visuomotor pathway, involved in body schema and social cognition. Rehabilitation of these disorders with behavioral therapies has aimed to either restore perceptuo-semantic deficits or compensate to overcome these deficits. Further studies are required to better stratify patients, using modern neurophysiology and neuroimaging techniques, to provide targeted and personalized therapies for these disorders in the future.

Controversy over the temporal cortical terminations of the left arcuate fasciculus: a reappraisal.

The arcuate fasciculus has been considered a major dorsal fronto-temporal white matter pathway linking frontal language production regions with auditory perception in the superior temporal gyrus, the so-called Wernicke's area. In line with this tradition, both historical and contemporary models of language function have assigned primacy to superior temporal projections of the arcuate fasciculus. However, classical anatomical descriptions and emerging behavioural data are at odds with this assumption. On one hand, fronto-temporal projections to Wernicke's area may not be unique to the arcuate fasciculus. On the other hand, dorsal stream language deficits have been reported also for damage to middle, inferior and basal temporal gyri that may be linked to arcuate disconnection. These findings point to a reappraisal of arcuate projections in the temporal lobe. Here, we review anatomical and functional evidence regarding the temporal cortical terminations of the left arcuate fasciculus by incorporating dissection and tractography findings with stimulation data using cortico-cortical evoked potentials and direct electrical stimulation mapping in awake patients. First, we discuss the fibres of the arcuate fasciculus projecting to the superior temporal gyrus and the functional rostro-caudal gradient in this region where both phonological encoding and auditory-motor transformation may be performed. Caudal regions within the temporoparietal junction may be involved in articulation and associated with temporoparietal projections of the third branch of the superior longitudinal fasciculus, while more rostral regions may support encoding of acoustic phonetic features, supported by arcuate fibres. We then move to examine clinical data showing that multimodal phonological encoding is facilitated by projections of the arcuate fasciculus to superior, but also middle, inferior and basal temporal regions. Hence, we discuss how projections of the arcuate fasciculus may contribute to acoustic (middle-posterior superior and middle temporal gyri), visual (posterior inferior temporal/fusiform gyri comprising the visual word form area) and lexical (anterior-middle inferior temporal/fusiform gyri in the basal temporal language area) information in the temporal lobe to be processed, encoded and translated into a dorsal phonological route to the frontal lobe. Finally, we point out surgical implications for this model in terms of the prediction and avoidance of neurological deficit.

Solar Power Prediction Using Dual Stream CNN-LSTM Architecture.

The integration of solar energy with a power system brings great economic and environmental benefits. However, the high penetration of solar power is challenging due to the operation and planning of the existing power system owing to the intermittence and randomicity of solar power generation. Achieving accurate predictions for power generation is important to provide high-quality electric energy for end-users. Therefore, in this paper, we introduce a deep learning-based dual-stream convolutional neural network (CNN) and long short-term nemory (LSTM) network followed by a self-attention mechanism network (DSCLANet). Here, CNN is used to learn spatial patterns and LSTM is incorporated for temporal feature extraction. The output spatial and temporal feature vectors are then fused, followed by a self-attention mechanism to select optimal features for further processing. Finally, fully connected layers are incorporated for short-term solar power prediction. The performance of DSCLANet is evaluated on DKASC Alice Spring solar datasets, and it reduces the error rate up to 0.0136 MSE, 0.0304 MAE, and 0.0458 RMSE compared to recent state-of-the-art methods.

DMU-Net: A Dual-Stream Multi-Scale U-Net Network Using Multi-Dimensional Spatial Information for Urban Building Extraction.

Automatically extracting urban buildings from remote sensing images has essential application value, such as urban planning and management. Gaofen-7 (GF-7) provides multi-perspective and multispectral satellite images, which can obtain three-dimensional spatial information. Previous studies on building extraction often ignored information outside the red-green-blue (RGB) bands. To utilize the multi-dimensional spatial information of GF-7, we propose a dual-stream multi-scale network (DMU-Net) for urban building extraction. DMU-Net is based on U-Net, and the encoder is designed as the dual-stream CNN structure, which inputs RGB images, near-infrared (NIR), and normalized digital surface model (nDSM) fusion images, respectively. In addition, the improved FPN (IFPN) structure is integrated into the decoder. It enables DMU-Net to fuse different band features and multi-scale features of images effectively. This new method is tested with the study area within the Fourth Ring Road in Beijing, and the conclusions are as follows: (1) Our network achieves an overall accuracy (OA) of 96.16% and an intersection-over-union (IoU) of 84.49% for the GF-7 self-annotated building dataset, outperforms other state-of-the-art (SOTA) models. (2) Three-dimensional information significantly improved the accuracy of building extraction. Compared with RGB and RGB + NIR, the IoU increased by 7.61% and 3.19% after using nDSM data, respectively. (3) DMU-Net is superior to SMU-Net, DU-Net, and IEU-Net. The IoU is improved by 0.74%, 0.55%, and 1.65%, respectively, indicating the superiority of the dual-stream CNN structure and the IFPN structure.

Using HVS Dual-Pathway and Contrast Sensitivity to Blindly Assess Image Quality.

Blind image quality assessment (BIQA) aims to evaluate image quality in a way that closely matches human perception. To achieve this goal, the strengths of deep learning and the characteristics of the human visual system (HVS) can be combined. In this paper, inspired by the ventral pathway and the dorsal pathway of the HVS, a dual-pathway convolutional neural network is proposed for BIQA tasks. The proposed method consists of two pathways: the "what" pathway, which mimics the ventral pathway of the HVS to extract the content features of distorted images, and the "where" pathway, which mimics the dorsal pathway of the HVS to extract the global shape features of distorted images. Then, the features from the two pathways are fused and mapped to an image quality score. Additionally, gradient images weighted by contrast sensitivity are used as the input to the "where" pathway, allowing it to extract global shape features that are more sensitive to human perception. Moreover, a dual-pathway multi-scale feature fusion module is designed to fuse the multi-scale features of the two pathways, enabling the model to capture both global features and local details, thus improving the overall performance of the model. Experiments conducted on six databases show that the proposed method achieves state-of-the-art performance.

Musician developed left putaminal hemorrhage and lost absolute pitch ability-case report.

Absolute pitch (AP) recognizes and labels the pitch chroma of a given tone without external reference. Its neural mechanism remains unclear. We report a 68-year-old AP musician who developed a left putaminal hemorrhage edematous lesion under the posterior insular cortex. Diffusion tensor tractography with the region of interest, including Heschl's gyrus, was performed. In the left hemisphere, the middle longitudinal fasciculus was absent, especially at the parietal lobe. Her AP ability was lost. As the hematoma was absorbed and the left MdLF was observed on the tractography, her AP ability recovered. Our case suggested that the left middle longitudinal fasciculus, a part of the ventral auditory pathway, plays a role in AP.

Absolute Camera Pose Regression Using an RGB-D Dual-Stream Network and Handcrafted Base Poses.

Absolute pose regression (APR) for camera localization is a single-shot approach that encodes the information of a 3D scene in an end-to-end neural network. The camera pose result of APR methods can be observed as the linear combination of the base poses. Previous APR methods' base poses are learned from training data. However, the training data can limit the performance of the methods, which cannot be generalized to cover the entire scene. To solve this issue, we use handcrafted base poses instead of learning-based base poses, which prevents overfitting the camera poses of the training data. Moreover, we use a dual-stream network architecture to process color and depth images separately to get more accurate localization. On the 7 Scenes dataset, the proposed method is among the best in median rotation error, and in median translation error, it outperforms previous APR methods. On a more difficult dataset-Oxford RobotCar dataset, the proposed method achieves notable improvements in median translation and rotation errors compared to the state-of-the-art APR methods.

mm-DSF: A Method for Identifying Dangerous Driving Behaviors Based on the Lateral Fusion of Micro-Doppler Features Combined.

To address the dangerous driving behaviors prevalent among current car drivers, it is necessary to provide real-time, accurate warning and correction of driver's driving behaviors in a small, movable, and enclosed space. In this paper, we propose a method for detecting dangerous behaviors based on frequency-modulated continuous-wave radar (mm-DSF). The highly packaged millimeter-wave radar chip has good in-vehicle emotion recognition capability. The acquired millimeter-wave differential frequency signal is Fourier-transformed to obtain the intermediate frequency signal. The physiological decomposition of the local micro-Doppler feature spectrum of the target action is then used as the eigenvalue. Matrix signal intensity and clutter filtering are performed by analyzing the signal echo model of the input channel. The signal classification is based on the estimation and variety of the feature vectors of the target key actions using a modified and optimized level fusion method of the SlowFast dual-channel network. Nine typical risky driving behaviors were set up by the Dula Hazard Questionnaire and TEIQue-SF, and the accuracy of the classification results of the self-built dataset was analyzed to verify the high robustness of the method. The recognition accuracy of this method increased by 1.97% compared with the traditional method.

Structural white matter connectometry of word production in aphasia: an observational study.

While current dual-steam neurocognitive models of language function have coalesced around the view that distinct neuroanatomical networks subserve semantic and phonological processing, respectively, the specific white matter components of these networks remain a matter of debate. To inform this debate, we investigated relationships between structural white matter connectivity and word production in a cross-sectional study of 42 participants with aphasia due to unilateral left hemisphere stroke. Specifically, we reconstructed a local connectome matrix for each participant from diffusion spectrum imaging data and regressed these matrices on indices of semantic and phonological ability derived from their responses to a picture-naming test and a computational model of word production. These connectometry analyses indicated that both dorsally located (arcuate fasciculus) and ventrally located (inferior frontal-occipital, uncinate, and middle longitudinal fasciculi) tracts were associated with semantic ability, while associations with phonological ability were more dorsally situated, including the arcuate and middle longitudinal fasciculi. Associations with limbic pathways including the posterior cingulum bundle and the fornix were also found. All analyses controlled for total lesion volume and all results showing positive associations obtained false discovery rates < 0.05. These results challenge dual-stream accounts that deny a role for the arcuate fasciculus in semantic processing, and for ventral-stream pathways in language production. They also illuminate limbic contributions to both semantic and phonological processing for word production.

The Cortical Organization of Syntax.

Syntax, the structure of sentences, enables humans to express an infinite range of meanings through finite means. The neurobiology of syntax has been intensely studied but with little consensus. Two main candidate regions have been identified: the posterior inferior frontal gyrus (pIFG) and the posterior middle temporal gyrus (pMTG). Integrating research in linguistics, psycholinguistics, and neuroscience, we propose a neuroanatomical framework for syntax that attributes distinct syntactic computations to these regions in a unified model. The key theoretical advances are adopting a modern lexicalized view of syntax in which the lexicon and syntactic rules are intertwined, and recognizing a computational asymmetry in the role of syntax during comprehension and production. Our model postulates a hierarchical lexical-syntactic function to the pMTG, which interconnects previously identified speech perception and conceptual-semantic systems in the temporal and inferior parietal lobes, crucial for both sentence production and comprehension. These relational hierarchies are transformed via the pIFG into morpho-syntactic sequences, primarily tied to production. We show how this architecture provides a better account of the full range of data and is consistent with recent proposals regarding the organization of phonological processes in the brain.

Connectivity Gradient in the Human Left Inferior Frontal Gyrus: Intraoperative Cortico-Cortical Evoked Potential Study.

In the dual-stream model of language processing, the exact connectivity of the ventral stream to the anterior temporal lobe remains elusive. To investigate the connectivity between the inferior frontal gyrus (IFG) and the lateral part of the temporal and parietal lobes, we integrated spatiotemporal profiles of cortico-cortical evoked potentials (CCEPs) recorded intraoperatively in 14 patients who had undergone surgical resection for a brain tumor or epileptic focus. Four-dimensional visualization of the combined CCEP data showed that the pars opercularis (Broca's area) is connected to the posterior temporal cortices and the supramarginal gyrus, whereas the pars orbitalis is connected to the anterior lateral temporal cortices and angular gyrus. Quantitative topographical analysis of CCEP connectivity confirmed an anterior-posterior gradient of connectivity from IFG stimulus sites to the temporal response sites. Reciprocality analysis indicated that the anterior part of the IFG is bidirectionally connected to the temporal or parietal area. This study shows that each IFG subdivision has different connectivity to the temporal lobe with an anterior-posterior gradient and supports the classical connectivity concept of Dejerine; that is, the frontal lobe is connected to the temporal lobe through the arcuate fasciculus and also a double fan-shaped structure anchored at the limen insulae.

Verbal and musical short-term memory: Variety of auditory disorders after stroke.

Auditory cognitive deficits after stroke may concern language and/or music processing, resulting in aphasia and/or amusia. The aim of the present study was to assess the potential deficits of auditory short-term memory for verbal and musical material after stroke and their underlying cerebral correlates with a Voxel-based Lesion Symptom Mapping approach (VLSM). Patients with an ischemic stroke in the right (N=10) or left (N=10) middle cerebral artery territory and matched control participants (N=14) were tested with a detailed neuropsychological assessment including global cognitive functions, music perception and language tasks. All participants then performed verbal and musical auditory short-term memory (STM) tasks that were implemented in the same way for both materials. Participants had to indicate whether series of four words or four tones presented in pairs, were the same or different. To detect domain-general STM deficits, they also had to perform a visual STM task. Behavioral results showed that patients had lower performance for the STM tasks in comparison with control participants, regardless of the material (words, tones, visual) and the lesion side. The individual patient data showed a double dissociation between some patients exhibiting verbal deficits without musical deficits or the reverse. Exploratory VLSM analyses suggested that dorsal pathways are involved in verbal (phonetic), musical (melodic), and visual STM, while the ventral auditory pathway is involved in musical STM.

Investigating category- and shape-selective neural processing in ventral and dorsal visual stream under interocular suppression.

Recent behavioral and neuroimaging studies using continuous flash suppression (CFS) have suggested that action-related processing in the dorsal visual stream might be independent of perceptual awareness, in line with the "vision-for-perception" versus "vision-for-action" distinction of the influential dual-stream theory. It remains controversial if evidence suggesting exclusive dorsal stream processing of tool stimuli under CFS can be explained by their elongated shape alone or by action-relevant category representations in dorsal visual cortex. To approach this question, we investigated category- and shape-selective functional magnetic resonance imaging-blood-oxygen level-dependent responses in both visual streams using images of faces and tools. Multivariate pattern analysis showed enhanced decoding of elongated relative to non-elongated tools, both in the ventral and dorsal visual stream. The second aim of our study was to investigate whether the depth of interocular suppression might differentially affect processing in dorsal and ventral areas. However, parametric modulation of suppression depth by varying the CFS mask contrast did not yield any evidence for differential modulation of category-selective activity. Together, our data provide evidence for shape-selective processing under CFS in both dorsal and ventral stream areas and, therefore, do not support the notion that dorsal "vision-for-action" processing is exclusively preserved under interocular suppression.

Weighing the evidence for a dorsal processing bias under continuous flash suppression.

With the introduction of continuous flash suppression (CFS) as a method to render stimuli invisible and study unconscious visual processing, a novel hypothesis has gained popularity. It states that processes typically ascribed to the dorsal visual stream can escape CFS and remain functional, while ventral stream processes are suppressed when stimuli are invisible under CFS. This notion of a CFS-specific "dorsal processing bias" has been argued to be in line with core characteristics of the influential dual-stream hypothesis of visual processing which proposes a dissociation between dorsally mediated vision-for-action and ventrally mediated vision-for-perception. Here, we provide an overview of neuroimaging and behavioral studies that either examine this dorsal processing bias or base their conclusions on it. We show that both evidence for preserved ventral processing as well as lack of dorsal processing can be found in studies using CFS. To reconcile the diverging results, differences in the paradigms and their effects are worthy of future research. We conclude that given the current level of information a dorsal processing bias under CFS cannot be universally assumed.

Supramodal processing optimizes visual perceptual learning and plasticity.

Multisensory interactions are ubiquitous in cortex and it has been suggested that sensory cortices may be supramodal i.e. capable of functional selectivity irrespective of the sensory modality of inputs (Pascual-Leone and Hamilton, 2001; Renier et al., 2013; Ricciardi and Pietrini, 2011; Voss and Zatorre, 2012). Here, we asked whether learning to discriminate visual coherence could benefit from supramodal processing. To this end, three groups of participants were briefly trained to discriminate which of a red or green intermixed population of random-dot-kinematograms (RDKs) was most coherent in a visual display while being recorded with magnetoencephalography (MEG). During training, participants heard no sound (V), congruent acoustic textures (AV) or auditory noise (AVn); importantly, congruent acoustic textures shared the temporal statistics - i.e. coherence - of visual RDKs. After training, the AV group significantly outperformed participants trained in V and AVn although they were not aware of their progress. In pre- and post-training blocks, all participants were tested without sound and with the same set of RDKs. When contrasting MEG data collected in these experimental blocks, selective differences were observed in the dynamic pattern and the cortical loci responsive to visual RDKs. First and common to all three groups, vlPFC showed selectivity to the learned coherence levels whereas selectivity in visual motion area hMT+ was only seen for the AV group. Second and solely for the AV group, activity in multisensory cortices (mSTS, pSTS) correlated with post-training performances; additionally, the latencies of these effects suggested feedback from vlPFC to hMT+ possibly mediated by temporal cortices in AV and AVn groups. Altogether, we interpret our results in the context of the Reverse Hierarchy Theory of learning (Ahissar and Hochstein, 2004) in which supramodal processing optimizes visual perceptual learning by capitalizing on sensory-invariant representations - here, global coherence levels across sensory modalities.